text
stringlengths 649
4.42k
| synonym_substitution
stringlengths 759
4.5k
| butter_fingers
stringlengths 649
4.42k
| random_deletion
stringlengths 453
2.31k
| change_char_case
stringlengths 649
4.42k
| whitespace_perturbation
stringlengths 764
5.02k
| underscore_trick
stringlengths 649
4.42k
|
|---|---|---|---|---|---|---|
} \left(p^{|E(C)|} (1-p)^{|\partial(C)|} - (1-p)^{|E(C)|} p^{|\partial(C)|} \right) \right),$$ where $C$ ranges over all connected subgraphs of $G(m)$ containing a given vertex, where $V(C),E(C)$ denote the vertex and edge set of $C$, and where $\partial (C)$ denotes the set of edges with at least one endpoint in $C$, which are not in $E(C)$. As mentioned, this expression for $D(p)$ does not involve the graphs $G_t(m)$ anymore, but its proof crucially relies on their existence.
Section \[section:approximation\] proves that replacing $D(p)$ in by a truncated series continues to yield an upper bound on the critical probability $p_c$ of $G(m)$ (Theorem \[theo:bound\]). This allows us to compute explicit numerical upper bounds on $p_c$. Finally, Section \[section:conclusion\] summarizes the results with Theorem \[theo:final\] and gives some concluding comments.
Finite quotient of the regular hyperbolic tilings {#section:siran_graphs}
=================================================
We are unaware of any method for constructing the required finite versions of $G(m)$ that does not involve a fair amount of algebra. In this section, we briefly recall Širáň’s method to construct such finite versions of the regular hyperbolic tiling $G(m)$. The first step is to construct $G(m)$ from a group of matrices over a ring of algebraic integers. Then this group is reduced modulo a prime number to yield the desired finite graph.
Denote by $P_k(X) = 2\cos(k\arccos(X/2))$ the $k$-th normalized Chebychev polynomial and let $\xi = 2\cos(\pi/m^2)$. Let $m \geq 5$ and consider the group $T(m)$ generated by the two following matrices of $SL_3({\mathbb{Z}}[\xi])$. $$a =
\left(
\begin{array}{ccc}
P_m(\xi)^2-1 & 0 & P_m(\xi)\\
P_m(\xi) & 1 & 0\\
-P_m(\ | } \left(p^{|E(C)| } (1 - p)^{|\partial(C)| } - (1 - p)^{|E(C)| } p^{|\partial(C)| } \right) \right),$$ where $ C$ ranges over all connected subgraphs of $ G(m)$ containing a given vertex, where $ V(C),E(C)$ announce the vertex and boundary stage set of $ C$, and where $ \partial (C)$ denotes the set of boundary with at least one endpoint in $ C$, which are not in $ E(C)$. As mentioned, this construction for $ D(p)$ does not necessitate the graphs $ G_t(m)$ anymore, but its validation crucially trust on their existence.
Section \[section: approximation\ ] proves that substitute $ D(p)$ in by a truncated series continues to succumb an upper bound on the critical probability $ p_c$ of $ G(m)$ (Theorem \[theo: bound\ ]). This allows us to compute explicit numerical upper bounds on $ p_c$. ultimately, Section \[section: conclusion\ ] summarizes the results with Theorem \[theo: final\ ] and give some concluding comments.
Finite quotient of the regular hyperbolic tilings { # section: siran_graphs }
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
We are unaware of any method acting for constructing the want finite versions of $ G(m)$ that does not involve a fair sum of algebra. In this section, we briefly recall Širáň ’s method to construct such finite versions of the regular hyperbolic tiling $ G(m)$. The first step is to construct $ G(m)$ from a group of matrices over a ring of algebraic integers. Then this group is reduced modulo a prime number to give way the desired finite graph.
Denote by $ P_k(X) = 2\cos(k\arccos(X/2))$ the $ k$-th normalized Chebychev polynomial and lease $ \xi = 2\cos(\pi / m^2)$. Let $ m \geq 5 $ and consider the group $ T(m)$ generated by the two following matrix of $ SL_3({\mathbb{Z}}[\xi])$. $ $ a =
\left (
\begin{array}{ccc }
P_m(\xi)^2 - 1 & 0 & P_m(\xi)\\
P_m(\xi) & 1 & 0\\
-P_m(\ | } \levt(p^{|E(C)|} (1-p)^{|\partial(C)|} - (1-p)^{|E(C)|} p^{|\pavtial(C)|} \right) \rigkr),$$ wherx $C$ ranfes over all connected subgraphs of $J(m)$ cintaibing a given vertex, whdre $V(C),E(C)$ fenote tye vtrtex and edge sev of $C$, akb whedc $\parciel (C)$ denotes thg set of edgas with at leavt oue endpoint in $C$, which are not in $E(C)$. As menyilned, this exprgssiom for $D(p)$ does not involve the graphs $G_t(j)$ anymoge, but its proof vrucially relies on their fxishence.
Section \[sectiln:approximajjon\] proves that feplacing $D(p)$ in by a ttuncated series continues to yieud an upper boubd on jhe critical probwbility $p_c$ on $G(m)$ (Thaorem \[thro:bound\]). This akloxs uw to compute explicit numerical upper bjunds on $[_c$. Finally, Section \[wextion:woncnusiub\] sjmmzrmzea the gesnlts with Tgeorem \[theo:funal\] and gives some cjbcluding commsnts.
Figiee quotient of the regular hyperbolic tplinfs {#section:siran_graphs}
=================================================
We are unaware of any mgthod for sonstructing the required finite versions of $G(m)$ tvat dkds uit inxilge a fair amount of algebra. In this section, wq btivfly recall Širáň’s mcthod to construct skcn finite versiuns of ths regular hyperbollc tiligg $G(m)$. The firsu stel is to construct $G(m)$ from a group of manricws over a ring of clgebraic incegers. Then this group is reduced oodumo a prime jumber to hield the desirea fpnita graph.
Dtvote by $P_k(X) = 2\cos(k\wrccos(X/2))$ tie $k$-tk normalkzed Chebyshev polynlmial and let $\xi = 2\cos(\pi/m^2)$. Let $l \ceq 5$ and clnsider the group $T(m)$ generated ub the two folkofinc matriczs of $XL_3({\mathbb{Z}}[\xi])$. $$a =
\left(
\begin{arrai}{ccc}
P_m(\xi)^2-1 & 0 & P_m(\bi)\\
P_m(\xi) & 1 & 0\\
-I_m(\ | } \left(p^{|E(C)|} (1-p)^{|\partial(C)|} - (1-p)^{|E(C)|} p^{|\partial(C)|} \right) $C$ over all subgraphs of $G(m)$ $V(C),E(C)$ the vertex and set of $C$, where $\partial (C)$ denotes the set edges with at least one endpoint in $C$, which are not in $E(C)$. mentioned, this expression for $D(p)$ does not involve the graphs $G_t(m)$ anymore, but proof relies their Section \[section:approximation\] proves that replacing $D(p)$ in by a truncated series continues to yield an upper on the critical probability $p_c$ of $G(m)$ (Theorem This allows us to explicit numerical upper bounds on Finally, \[section:conclusion\] summarizes results Theorem and gives some comments. Finite quotient of the regular hyperbolic tilings {#section:siran_graphs} ================================================= We are unaware of any method for the required of $G(m)$ does involve fair amount of this section, we briefly recall Širáň’s such finite versions of the regular hyperbolic tiling The first is to construct $G(m)$ from a of matrices over a ring of algebraic integers. this group is reduced modulo a prime number to yield the desired finite graph. Denote = 2\cos(k\arccos(X/2))$ the $k$-th Chebychev polynomial and $\xi 2\cos(\pi/m^2)$. $m 5$ and the group $T(m)$ generated by the two following matrices of $SL_3({\mathbb{Z}}[\xi])$. = \left( \begin{array}{ccc} P_m(\xi)^2-1 & 0 & P_m(\xi)\\ P_m(\xi) & 0\\ | } \left(p^{|E(C)|} (1-p)^{|\partial(C)|} - (1-p)^{|E(C)|} p^{|\partiAl(C)|} \right) \riGht),$$ whEre $c$ raNgEs ovEr alL connected subgRAphs Of $G(m)$ containing a given veRtex, wHeRE $V(C),E(c)$ DeNote tHe verteX AnD EDge SeT oF $C$, aNd WHeRe $\parTiaL (C)$ denotEs the set of EdgEs With at least oNE eNdpoint in $C$, WhiCh are not in $E(C)$. as mEntionEd, ThiS ExpreSsiOn for $d(p)$ does NOt invoLve the graPhS $g_t(m)$ anyMOre, but iTS PrOof cRucially relies on tHEiR Existence.
SectiOn \[sectIoN:ApPROxiMatIon\] proves tHaT replACing $D(p)$ iN By A TRUncATed series contInues to yielD An uPper boUnD on THe critIcal pRoBAbiLity $p_c$ of $G(m)$ (THeorEm \[theo:bouNd\]). This ALlows us TO computE expliCit NumEricAL uPpEr bOuNDs oN $P_c$. finALly, section \[sEcTiOn:conClusION\] SUmmaRizEs thE resuLts with TheoreM \[thEo:fiNAl\] aNd givEs somE conClUding CommenTs.
FinItE quotient of the rEgulAr hyperboLic TiLinGs {#SectiON:siran_GraPhs}
=================================================
we are unAware of ANy mEtHOD FoR constructing the reQuIREd Finite veRsions OF $G(M)$ tHAt does noT iNvoLve a FAIr amoUnt oF AlGebra. In tHis secTIoN, wE brieflY rEcall ŠIrÁň’s MetHod to COnstRuct suCh finite VersiONs of the regular HYperbolic tiliNG $G(M)$. tHe FIrst SteP is to constrUct $G(M)$ From A groUP oF maTRices Over a RiNG oF Algebraic integers. ThEn This grOup is Reduced modulo A prime numbER TO yield thE desIReD Finite graph.
DenOte by $p_k(X) = 2\cos(k\arcCOs(X/2))$ the $k$-tH normAlized ChEbychev poLYNomial anD leT $\xi = 2\Cos(\Pi/m^2)$. lET $m \Geq 5$ and consideR THe grOuP $T(m)$ geneRatEd by the Two FolLowIng MaTrices of $Sl_3({\mathbb{Z}}[\Xi])$. $$A =
\lEfT(
\bEgiN{arraY}{Ccc}
P_m(\xi)^2-1 & 0 & P_M(\xI)\\
P_m(\Xi) & 1 & 0\\
-p_m(\ | } \left(p^{|E(C)|} (1-p)^{ |\partial( C)|}- ( 1-p )^ {|E( C)|} p^{|\partial( C )|}\right) \right),$$ whe re $C $r ange s o ver a ll conn e ct e d su bg ra phs o f $ G(m)$ co ntainin g a givenver te x, where $V( C ), E(C)$ deno tethe vertex a ndedge s et of $C$,and wher e $\pa r tial ( C)$ denot es the se t of edg e s w ithat least one endp o in t in $C$, which are n ot in $ E(C )$. As mentio ne d, th i s expre s si o n for $D(p)$ does n ot involvet hegraphs $ G_t ( m)$ an ymore ,b utits proof c ruci ally reli es ont heir ex i stence.
Sect ion \[ sect i on :a ppr ox i mat i on \]p rov es thatre pl acing $D( p ) $ in b y a tru ncate d series cont inu es t o yi eld a n upp er b ou nd on the c ritic al probability $p _c$of $G(m)$ (T he ore m\[the o :bound \]) . T his all ows ust o c om p u t eexplicit numerical u p p er boundson $p_ c $. F i nally, S ec tio n \[ s e ction :con c lu sion\] s ummari z es t he resu lt s with T heo rem \[th e o:fi nal\]and give s som e concluding co m ments.
Finit e q u o ti e nt o f t he regularhype r boli c ti l in gs{ #sect ion:s ir a n_ g raphs}
============ == ====== ===== ============= ========== = We are u nawa r eo f any method f or co nstructing the requ iredfinite v ersions o f $G(m)$ t hat do esnot i nv olve a fair a m o untof algebr a.In this se cti on, we b riefly re call Šir áň ’s m et hod to c o nstructsu chfi nit e ver s ions o f the reg ul ar hyp erbolic ti l i ng $ G( m) $. T hefi rst s tepi s t o const ruct $G(m )$f romagr oup ofmatrices over a ring of a lg ebr aic in t e gers. Th en this group is reduce d modulo aprime num ber to yi eld the d esi r ed fin ite gr aph.
D eno t e by $ P _ k( X)=2\cos(k\ar c c os( X/2)) $the$k$-thnormalized Chebych e v p olynomial and le t $\ x i = 2\ c os ( \pi /m ^ 2)$ . Let $m \geq 5$and consid er th e group $T ( m)$ g enerate d by th e two followi ng matric es of $SL _3 ({\m a t hbb {Z}}[\xi]) $. $$a =
\left(
\ b egin{ a rr ay}{c cc}
P_m(\ xi )^2 -1 &0 & P_ m (\x i)\\P_m(\x i) & 1 & 0\\-P _m(\ | } \left(p^{|E(C)|}_(1-p)^{|\partial(C)|} -_(1-p)^{|E(C)|} p^{|\partial(C)|} \right) \right),$$_where $C$_ranges_over all_connected_subgraphs of $G(m)$_containing a given_vertex, where $V(C),E(C)$ denote_the vertex and_edge_set of $C$, and where $\partial (C)$ denotes the set of edges with at_least_one endpoint_in_$C$,_which are not in $E(C)$._As mentioned, this expression for_$D(p)$ does_not involve the graphs $G_t(m)$ anymore, but its_proof_crucially relies on_their existence.
Section \[section:approximation\] proves that replacing $D(p)$ in by_a truncated series continues to yield_an upper bound_on_the_critical probability $p_c$ of_$G(m)$ (Theorem \[theo:bound\]). This allows us to_compute explicit numerical upper bounds on_$p_c$. Finally, Section \[section:conclusion\] summarizes the results_with Theorem \[theo:final\] and gives some concluding_comments.
Finite quotient of the regular_hyperbolic tilings_{#section:siran_graphs}
=================================================
We are unaware of any_method for constructing_the required_finite versions of_$G(m)$ that does not involve a_fair amount of_algebra. In this section, we briefly_recall_Širáň’s method to_construct_such_finite versions_of the regular_hyperbolic_tiling $G(m)$._The_first step is to construct $G(m)$_from_a group of matrices over a ring_of algebraic integers. Then_this_group is reduced modulo_a prime number to yield_the desired finite graph.
Denote by $P_k(X)_= 2\cos(k\arccos(X/2))$_the $k$-th_normalized Chebychev polynomial and let $\xi = 2\cos(\pi/m^2)$. Let $m \geq_5$ and consider the group $T(m)$_generated by the two_following matrices_of_$SL_3({\mathbb{Z}}[\xi])$. $$a =
\left(
\begin{array}{ccc}
P_m(\xi)^2-1_&_0 &_P_m(\xi)\\
P_m(\xi) & 1 & 0\\
-P_m(\ |
\label{FYAI2}\end{aligned}$$ The total wave function is then given by: $$\begin{aligned}
\Psi &=& \Psi_{1+3} + \Psi_{2+2} \cr
\Psi_{1+3} &=& \left[ 1+ \varepsilon(P_{13}+P_{23} ) \right]\;
\left[ 1+ \varepsilon(P_{14}+P_{24}+P_{34}) \right]K \label{psi13}\\
\Psi_{2+2} &=& \left[ 1+\varepsilon( P_{13}+ P_{23}+ P_{14}+ P_{24} ) + P_{13}P_{24}\right] \;H \label{psi22}\end{aligned}$$ Each amplitude $\Phi=K,H$ is considered as a function of its natural set of Jacobi coordinates $\vec{x}_{\Phi},\vec{y}_{\Phi},\vec{z}_{\Phi}$, defined respectively by equations (\[K\]) and (\[H\]) with $(ijkl)=(1234)$ and $m=m_i$: $$\begin{aligned}
\vec{x}_K&=& \vec{r}_2-\vec{r}_1 \cr
\vec{y}_K&=&\displaystyle \sqrt{4 \over3}\left(\vec{r}_3-{\vec{r}_1+\vec{r}_2\over2}\right)\cr
\vec{z}_K&=&\displaystyle \sqrt{3 \over2}\left(\vec{r}_4-{\vec{r}_1+\vec{r}_2+\vec{r}_3\over3}\right)\end{aligned}$$$$\begin{aligned}
\vec{x}_H&=& \vec{r}_2-\vec{r}_1 \cr
\vec{y}_H&=& \vec{r}_4-\vec{r}_3 \cr
\vec{z}_H&=&\displaystyle \sqrt{2}\left({\vec{r}_3+\vec{r}_4\over2}-{\vec{r}_1+\vec{r}_2\over2}\right)\end{aligned}$$
They are expanded in angular momentum variables for each coordinate according to $$\label{KPW}
<\vec{x}\vec{y}\vec{z}|{\Phi}>=
\sum_{\alpha} \; {\phi | \label{FYAI2}\end{aligned}$$ The total wave function is then given by: $ $ \begin{aligned }
\Psi & = & \Psi_{1 + 3 } + \Psi_{2 + 2 } \cr
\Psi_{1 + 3 } & = & \left [ 1 + \varepsilon(P_{13}+P_{23 } ) \right]\;
\left [ 1 + \varepsilon(P_{14}+P_{24}+P_{34 }) \right]K \label{psi13}\\
\Psi_{2 + 2 } & = & \left [ 1+\varepsilon (P_{13}+ P_{23}+ P_{14}+ P_{24 }) + P_{13}P_{24}\right ] \;H \label{psi22}\end{aligned}$$ Each amplitude $ \Phi = K, H$ is consider as a affair of its natural set of Jacobi coordinates $ \vec{x}_{\Phi},\vec{y}_{\Phi},\vec{z}_{\Phi}$, define respectively by equations (\[K\ ]) and (\[H\ ]) with $ (ijkl)=(1234)$ and $ m = m_i$: $ $ \begin{aligned }
\vec{x}_K&= & \vec{r}_2-\vec{r}_1 \cr
\vec{y}_K&=&\displaystyle \sqrt{4 \over3}\left(\vec{r}_3-{\vec{r}_1+\vec{r}_2\over2}\right)\cr
\vec{z}_K&=&\displaystyle \sqrt{3 \over2}\left(\vec{r}_4-{\vec{r}_1+\vec{r}_2+\vec{r}_3\over3}\right)\end{aligned}$$$$\begin{aligned }
\vec{x}_H&= & \vec{r}_2-\vec{r}_1 \cr
\vec{y}_H&= & \vec{r}_4-\vec{r}_3 \cr
\vec{z}_H&=&\displaystyle \sqrt{2}\left({\vec{r}_3+\vec{r}_4\over2}-{\vec{r}_1+\vec{r}_2\over2}\right)\end{aligned}$$
They are boom in angular momentum variable for each coordinate according to $ $ \label{KPW }
< \vec{x}\vec{y}\vec{z}|{\Phi}>=
\sum_{\alpha } \; { \phi | \lwbel{FYAI2}\end{aligned}$$ The tutal wave functnin is vhen giben by: $$\bdgin{aligned}
\Psi &=& \Psi_{1+3} + \'si_{2+2} \xr
\Pwi_{1+3} &=& \left[ 1+ \varepsilon(P_{13}+O_{23} ) \gight]\;
\lefv[ 1+ \varepsilon(P_{14}+L_{24}+I_{34}) \rigkt]J \label{psi13}\\
\Psi_{2+2} &=& \left[ 1+\vasepsilon( P_{13}+ P_{23}+ P_{14}+ P_{24} ) + P_{13}P_{24}\right] \;H \label{psi22}\end{aligned}$$ Each amplitidf $\Phi=K,H$ is confidegeq as a function of its natural set of Nacobi boordinates $\vec{x}_{\Pni},\vec{y}_{\Phi},\vec{z}_{\Phi}$, defined rfspeftively by equatiojs (\[K\]) and (\[H\]) qith $(ujkl)=(1234)$ and $m=m_i$: $$\begin{aligned}
\vec{x}_K&=& \vec{t}_2-\vec{r}_1 \cr
\vec{y}_K&=&\displaystyle \sqrt{4 \oxer3}\leyt(\vec{r}_3-{\vec{r}_1+\vge{e}_2\ovft2}\right)\cr
\vec{z}_J&=&\dispjaystyle \sqrb{3 \over2}\laft(\vec{r}_4-{\fec{r}_1+\vec{r}_2+\vec{r}_3\ovcr3}\rigit)\ene{aligned}$$$$\begin{aligned}
\vxc{x}_H&=& \vec{r}_2-\vec{r}_1 \cr
\vec{y}_R&=& \vec{r}_4-\vec{s}_3 \er
\vec{z}_H&=&\displaystyle \swrr{2}\left({\eec{r}_3+\eec{r}_4\ucer2}-{\xec{d}_1+\vxc{r}_2\kver2}\rihht)\xnd{aligned}$$
Tgey are expqnded in angular mokegnim variables for ewcr coordinate according to $$\label{KPW}
<\vec{x}\vtc{y}\ved{z}|{\Phi}>=
\sum_{\alpha} \; {\phi | \label{FYAI2}\end{aligned}$$ The total wave function is then $$\begin{aligned} &=& \Psi_{1+3} \Psi_{2+2} \cr \Psi_{1+3} \right]\; 1+ \varepsilon(P_{14}+P_{24}+P_{34}) \right]K \Psi_{2+2} &=& \left[ P_{13}+ P_{23}+ P_{14}+ P_{24} ) + \;H \label{psi22}\end{aligned}$$ Each amplitude $\Phi=K,H$ is considered as a function of its natural of Jacobi coordinates $\vec{x}_{\Phi},\vec{y}_{\Phi},\vec{z}_{\Phi}$, defined respectively by equations (\[K\]) and (\[H\]) with $(ijkl)=(1234)$ $m=m_i$: \vec{x}_K&=& \cr \sqrt{4 \over3}\left(\vec{r}_3-{\vec{r}_1+\vec{r}_2\over2}\right)\cr \vec{z}_K&=&\displaystyle \sqrt{3 \over2}\left(\vec{r}_4-{\vec{r}_1+\vec{r}_2+\vec{r}_3\over3}\right)\end{aligned}$$$$\begin{aligned} \vec{x}_H&=& \vec{r}_2-\vec{r}_1 \cr \vec{y}_H&=& \vec{r}_4-\vec{r}_3 \cr \vec{z}_H&=&\displaystyle \sqrt{2}\left({\vec{r}_3+\vec{r}_4\over2}-{\vec{r}_1+\vec{r}_2\over2}\right)\end{aligned}$$ They are expanded angular momentum variables for each coordinate according to <\vec{x}\vec{y}\vec{z}|{\Phi}>= \sum_{\alpha} \; {\phi | \label{FYAI2}\end{aligned}$$ The totAl wave funcTion iS thEn gIvEn by: $$\BegiN{aligned}
\Psi &=& \Psi_{1+3} + \pSi_{2+2} \cr
\psi_{1+3} &=& \left[ 1+ \varepsilon(P_{13}+P_{23} ) \rigHt]\;
\lefT[ 1+ \vARepsILoN(P_{14}+P_{24}+P_{34}) \rIght]K \laBEl{PSI13}\\
\PsI_{2+2} &=& \lEfT[ 1+\vaRePSiLon( P_{13}+ P_{23}+ p_{14}+ P_{24} ) + P_{13}p_{24}\right] \;H \Label{psi22}\enD{alIgNed}$$ Each ampliTUdE $\Phi=K,H$ is coNsiDered as a funcTioN of its NaTurAL set oF JaCobi cOordinATes $\vec{X}_{\Phi},\vec{y}_{\PHi},\VEc{z}_{\Phi}$, DEfined rESPeCtivEly by equations (\[K\]) anD (\[h\]) wITh $(ijkl)=(1234)$ and $m=m_i$: $$\beGin{aliGnED}
\vEC{X}_K&=& \vEc{r}_2-\Vec{r}_1 \cr
\vec{y}_k&=&\dIsplaYStyle \sqRT{4 \oVER3}\LefT(\Vec{r}_3-{\vec{r}_1+\vec{r}_2\oVer2}\right)\cr
\vEC{z}_K&=&\DisplaYsTylE \Sqrt{3 \ovEr2}\lefT(\vEC{r}_4-{\vEc{r}_1+\vec{r}_2+\vec{r}_3\Over3}\Right)\end{aLigned}$$$$\BEgin{aliGNed}
\vec{x}_h&=& \vec{r}_2-\vEc{r}_1 \Cr
\vEc{y}_H&=& \VEc{R}_4-\vEc{r}_3 \Cr
\VEc{z}_h&=&\DiSplAYstYle \sqrt{2}\lEfT({\vEc{r}_3+\veC{r}_4\ovER2}-{\VEC{r}_1+\veC{r}_2\oVer2}\rIght)\eNd{aligned}$$
They Are ExpaNDed In angUlar mOmenTuM variAbles fOr eacH cOordinate accordIng tO $$\label{KPW}
<\Vec{X}\vEc{y}\VeC{z}|{\Phi}>=
\SUm_{\alphA} \; {\phI | \label{FYAI2}\end{aligne d}$$ The t otalwav e f un ctio n is then given by : $$\ begin{aligned}
\P si & =& \Ps i _{ 1+3}+ \Psi_ { 2+ 2 } \c r \Ps i_ { 1+ 3} &= & \left[1+ \vareps ilo n( P_{13}+P_{23 } ) \ri ght ]\;
\ le ft[ 1+ \v are psilo n(P_{1 4 }+P_{2 4}+P_{34} )\ right] K \labe l { ps i13} \\
\Psi_{2+2} &=& \ l eft[ 1+\vareps ilon(P_ { 13 } + P_ {23 }+ P_{14}+ P _{24} ) + P_{ 1 3} P _ { 24} \ right] \;H \ label{psi22 } \en d{alig ne d}$ $ Eachampli tu d e $ \Phi=K,H$ i s co nsideredas a f u nctiono f its n atural se t o f Ja c ob icoo rd i nat e s$\v e c{x }_{\Phi} ,\ ve c{y}_ {\Ph i } , \ vec{ z}_ {\Ph i}$,defined respe cti vely byequat ions(\[K \] ) and (\[H\ ]) wi th $(ijkl)=(1234) $ an d $m=m_i$ : $ $\ beg in {alig n ed}
\v ec{ x}_ K&=& \v ec{r}_2 - \ve c{ r } _ 1\cr
\vec{y}_K&=&\d is p l ay style \s qrt{4\ ov er 3 }\left(\ ve c{r }_3- { \ vec{r }_1+ \ ve c{r}_2\o ver2}\ r ig ht )\cr
\v ec {z}_K& =& \di spl aysty l e \s qrt{3\over2}\ left( \ vec{r}_4-{\vec { r}_1+\vec{r}_ 2 +\ v e c{ r }_3\ ove r3}\right)\ end{ a lign ed}$ $ $$ \be g in{al igned }\ ve c {x}_H&=& \vec{r}_2- \v ec{r}_ 1 \cr
\vec{y}_H&=& \vec{r}_4 - \ v ec{r}_3\cr\ ve c {z}_H&=&\displ aysty le \sqrt{2 } \left({\ vec{r }_3+\vec {r}_4\ove r 2 }-{\vec{ r}_ 1+\ vec {r} _ 2 \o ver2}\right)\ e n d{al ig ned}$$
Th ey areexp and edinan gular mom entum va ri ab le sfor each coordina te ac co rdi ng to $$\lab el{KP W}
< \v ec { x}\ vec{y}\ v ec { z }|{\ Ph i} >=
\ sum _{ \alph a} \ ; {\ phi | _\label{FYAI2}\end{aligned}$$ The_total wave function is_then given_by:_$$\begin{aligned}
__ \Psi_ &=& _\Psi_{1+3} + \Psi_{2+2} \cr_
\Psi_{1+3} &=&__\left[ 1+ \varepsilon(P_{13}+P_{23} ) \right]\;
__ ___ _ \left[ 1+ \varepsilon(P_{14}+P_{24}+P_{34})_\right]K _\label{psi13}\\
\Psi_{2+2} &=& \left[ 1+\varepsilon( P_{13}+ P_{23}+ P_{14}+_P_{24}_) + P_{13}P_{24}\right]_\;H \label{psi22}\end{aligned}$$ Each amplitude $\Phi=K,H$ is considered as_a function of its natural set_of Jacobi coordinates_$\vec{x}_{\Phi},\vec{y}_{\Phi},\vec{z}_{\Phi}$,_defined_respectively by equations (\[K\])_and (\[H\]) with $(ijkl)=(1234)$ and $m=m_i$:_$$\begin{aligned}
\vec{x}_K&=& \vec{r}_2-\vec{r}_1 \cr
\vec{y}_K&=&\displaystyle \sqrt{4 \over3}\left(\vec{r}_3-{\vec{r}_1+\vec{r}_2\over2}\right)\cr
\vec{z}_K&=&\displaystyle \sqrt{3_\over2}\left(\vec{r}_4-{\vec{r}_1+\vec{r}_2+\vec{r}_3\over3}\right)\end{aligned}$$$$\begin{aligned}
\vec{x}_H&=& \vec{r}_2-\vec{r}_1 \cr
\vec{y}_H&=& \vec{r}_4-\vec{r}_3 \cr
\vec{z}_H&=&\displaystyle \sqrt{2}\left({\vec{r}_3+\vec{r}_4\over2}-{\vec{r}_1+\vec{r}_2\over2}\right)\end{aligned}$$
They are_expanded in angular momentum variables for_each coordinate according to $$\label{KPW}
<\vec{x}\vec{y}\vec{z}|{\Phi}>=
\sum_{\alpha}_\; {\phi |
,j=1,\ldots,l$, the vectors $\tilde{x_1},\ldots,\tilde{x_k},\tilde{y_1},\ldots,\tilde{y_l},\tilde{u_1},\ldots,\tilde{u_l}\in {{\mathbf R}}^{N+1}$ are linearly independent.]{}
[**Proof of Proposition \[tensor\].**]{}First note that a tangent vector to $f(x,y)$ is of the form $(u\otimes y+x\otimes v,u \otimes 1, 1 \otimes v)$ where $u\in T_x M$ and $v\in T_y N$. Fix arbitrary distinct points $(x_i,y_i)\in M\times N, i =1,\ldots,k+l$ and arbitrary non-zero vectors $(u_j,v_j)\in T_{(x_j,y_j)}M\times N, j=k+1,\ldots,k+l.$ Because of remark \[operative\], it suffices to show, after the identification $$(U \otimes V)\oplus(U\otimes{{\mathbf R}})\oplus(V\otimes {{\mathbf R}})\oplus {{\mathbf R}}\cong (U\oplus {{\mathbf R}})\otimes (V \oplus {{\mathbf R}})$$ that the vectors in the set $$\mathcal S=\{\tilde{x_i}\otimes \tilde{y_i}| i=1,\ldots,k+l\} \cup \{\tilde u_j \otimes \tilde y_j + \tilde x_j \otimes \tilde v_j| j=k+1,\ldots,k+l\}$$ are linearly independent.
Since $M$ is $(k,l)$-regular, we know that the vectors in $\mathcal M=\{\tilde {x_i}, \tilde{u_j}|u_j\neq0\}$ are linearly independent and similarly for $\mathcal N=\{\tilde {y_i}, \tilde {v_j}|v_j\neq0\}.$ But now we observe that every nontrivial linear combination of vectors in $\mathcal S$ would also be a nontrivial linear combination of vectors in $$\{e\otimes f| e\in \mathcal M,f \in \mathcal N\},$$ and that is not possible. [$\Box$]{}
[**Proof of Theorem \[main\].**]{} First we prove the lower bound.We will make use of the following theorem, affine | , j=1,\ldots, l$, the vectors $ \tilde{x_1},\ldots,\tilde{x_k},\tilde{y_1},\ldots,\tilde{y_l},\tilde{u_1},\ldots,\tilde{u_l}\in { { \mathbf R}}^{N+1}$ are linearly independent. ] { }
[ * * Proof of Proposition \[tensor\].**]{}First note that a tangent vector to $ f(x, y)$ is of the human body $ (u\otimes y+x\otimes five, u \otimes 1, 1 \otimes v)$ where $ u\in T_x M$ and $ v\in T_y N$. Fix arbitrary distinct points $ (x_i, y_i)\in M\times N, i = 1,\ldots, k+l$ and arbitrary non - zero vector $ (u_j, v_j)\in T_{(x_j, y_j)}M\times N, j = k+1,\ldots, k+l.$ Because of remark \[operative\ ], it suffices to express, after the recognition $ $ (U \otimes V)\oplus(U\otimes{{\mathbf R}})\oplus(V\otimes { { \mathbf R}})\oplus { { \mathbf R}}\cong (U\oplus { { \mathbf R}})\otimes (V \oplus { { \mathbf R}})$$ that the vectors in the fit $ $ \mathcal S=\{\tilde{x_i}\otimes \tilde{y_i}| i=1,\ldots, k+l\ } \cup \{\tilde u_j \otimes \tilde y_j + \tilde x_j \otimes \tilde v_j| j = k+1,\ldots, k+l\}$$ are linearly independent.
Since $ M$ is $ (k, l)$-regular, we sleep together that the vectors in $ \mathcal M=\{\tilde { x_i }, \tilde{u_j}|u_j\neq0\}$ are linearly independent and similarly for $ \mathcal N=\{\tilde { y_i }, \tilde { v_j}|v_j\neq0\}.$ But immediately we respect that every nontrivial linear combination of vectors in $ \mathcal S$ would besides be a nontrivial linear combination of vectors in $ $ \{e\otimes f| e\in \mathcal M, f \in \mathcal N\},$$ and that is not potential. [ $ \Box$ ] { }
[ * * validation of Theorem \[main\ ]. * * ] { } First we prove the lower bound. We will make manipulation of the following theorem, affine | ,j=1,\ldlts,l$, the vectors $\tilde{x_1},\laots,\tilde{x_k},\tildg{y_1},\odots,\tmlde{y_l},\tjlde{u_1},\ldogs,\tilde{u_l}\in {{\mathbf R}}^{N+1}$ are liiearoy ineependent.]{}
[**Proof of Propusition \[tvnsor\].**]{}Firsr noue that a tangent vector bj $f(x,g)$ is mh the form $(u\otikes y+x\otimas v,u \otimes 1, 1 \ogiles v)$ where $u\in T_x M$ and $v\in T_y N$. Fyx arbiyrwry distinct pjintx $(x_i,y_j)\pn M\times N, i =1,\ldots,k+l$ and arbitdary noi-zero vectors $(u_k,v_j)\in T_{(x_j,y_j)}M\times N, j=k+1,\ldotd,k+l.$ Hecause of remark \[lperative\], ij susdices to shod, after tht ndentificatjon $$(U \otimes V)\oplus(U\otimes{{\mathbw R}})\opkus(V\otimes {{\mqthhx R}})\oplus {{\mavhbf R}}\bong (U\oplus {{\mathbf R}})\odimes (V \oplus {{\mathbf V}})$$ thav thw vectors in the set $$\kathcal S=\{\tilde{x_i}\ojimes \tilda{y_n}| i=1,\ldots,k+l\} \cup \{\tilde y_j \otimgs \tinde h_h + \gilse x_n \otimfs \vilde v_j| j=k+1,\mdots,k+l\}$$ are linearly independemt.
Fpmce $M$ is $(k,l)$-rsgular, wq know that the vectors in $\mathcal M=\{\tilge {s_i}, \tilde{u_j}|u_j\neq0\}$ are linwarly independent and similarlr for $\mathcal N=\{\tilde {y_i}, \tilde {v_j}|v_j\neq0\}.$ But now we mbserte thcb dcegy nontrivial linear combination of vectors ig $\mstmcal S$ would alsj be a nonttigisj linear combknatiou or vectors in $$\{e\otimfs f| e\ig \matycal M,f \ig \mayhcal N\},$$ and that is not poswible. [$\Box$]{}
[**Projd of Theorem \[main\].**]{} Yirst we prore the lowet bound.We will make use of fhe followijg theorej, affine | ,j=1,\ldots,l$, the vectors $\tilde{x_1},\ldots,\tilde{x_k},\tilde{y_1},\ldots,\tilde{y_l},\tilde{u_1},\ldots,\tilde{u_l}\in {{\mathbf R}}^{N+1}$ are [**Proof Proposition \[tensor\].**]{}First that a tangent the $(u\otimes y+x\otimes v,u 1, 1 \otimes where $u\in T_x M$ and $v\in N$. Fix arbitrary distinct points $(x_i,y_i)\in M\times N, i =1,\ldots,k+l$ and arbitrary non-zero $(u_j,v_j)\in T_{(x_j,y_j)}M\times N, j=k+1,\ldots,k+l.$ Because of remark \[operative\], it suffices to show, after identification \otimes R}})\oplus(V\otimes R}})\oplus {{\mathbf R}}\cong (U\oplus {{\mathbf R}})\otimes (V \oplus {{\mathbf R}})$$ that the vectors in the set S=\{\tilde{x_i}\otimes \tilde{y_i}| i=1,\ldots,k+l\} \cup \{\tilde u_j \otimes \tilde + \tilde x_j \otimes v_j| j=k+1,\ldots,k+l\}$$ are linearly independent. $M$ $(k,l)$-regular, we that vectors $\mathcal M=\{\tilde {x_i}, are linearly independent and similarly for $\mathcal N=\{\tilde {y_i}, \tilde {v_j}|v_j\neq0\}.$ But now we observe that every linear combination in $\mathcal would be nontrivial linear combination in $$\{e\otimes f| e\in \mathcal M,f and that is not possible. [$\Box$]{} [**Proof of \[main\].**]{} First prove the lower bound.We will make of the following theorem, affine | ,j=1,\ldots,l$, the vectors $\tilde{x_1},\ldOts,\tilde{x_k},\Tilde{Y_1},\ldOts,\TiLde{y_L},\tilDe{u_1},\ldots,\tilde{u_L}\In {{\maThbf R}}^{N+1}$ are linearly indepEndenT.]{}
[**PROof oF prOposiTion \[tenSOr\].**]{}fIRst NoTe ThaT a TAnGent vEctOr to $f(x,y)$ Is of the forM $(u\oTiMes y+x\otimes v,U \OtImes 1, 1 \otimes V)$ whEre $u\in T_x M$ and $V\in t_y N$. Fix ArBitRAry diStiNct poInts $(x_i,Y_I)\in M\tiMes N, i =1,\ldotS,k+L$ And arbITrary noN-ZErO vecTors $(u_j,v_j)\in T_{(x_j,y_j)}M\tIMeS n, j=k+1,\ldots,k+l.$ BecaUse of rEmARk \[OPEraTivE\], it sufficeS tO show, AFter the IDeNTIFicATion $$(U \otimes V)\oPlus(U\otimes{{\MAthBf R}})\oplUs(v\otIMes {{\matHbf R}})\oPlUS {{\maThbf R}}\cong (U\oPlus {{\Mathbf R}})\otImes (V \oPLus {{\mathBF R}})$$ that tHe vectOrs In tHe seT $$\MaThCal s=\{\tILde{X_I}\oTimES \tiLde{y_i}| i=1,\ldOtS,k+L\} \cup \{\tIlde U_J \OTImes \TilDe y_j + \Tilde X_j \otimes \tilde V_j| j=K+1,\ldoTS,k+l\}$$ Are liNearlY indEpEndenT.
Since $m$ is $(k,l)$-ReGular, we know that The vEctors in $\mAthCaL M=\{\tIlDe {x_i}, \tILde{u_j}|u_J\neQ0\}$ arE linearLy indepENdeNt AND SiMilarly for $\mathcal N=\{\TiLDE {y_I}, \tilde {v_j}|V_j\neq0\}.$ BUT nOw WE observe ThAt eVery NONtrivIal lINeAr combinAtion oF VeCtOrs in $\maThCal S$ woUlD alSo bE a nonTRiviAl lineAr combinAtion OF vectors in $$\{e\otiMEs f| e\in \mathcal m,F \iN \MAtHCal N\},$$ And That is not poSsibLE. [$\Box$]{}
[**prooF Of theORem \[maIn\].**]{} FirSt WE pROve the lower bound.We wIlL make uSe of tHe following thEorem, affinE | ,j=1,\ldots,l$, the vector s $\tilde{ x_1}, \ld ots ,\ tild e{x_ k},\tilde{y_1} , \ldo ts,\tilde{y_l},\tilde{ u_1}, \l d ots, \ ti lde{u _l}\in{ {\ m a thb fR} }^{ N+ 1 }$ arelin early i ndependent .]{ }
[**Proof of Pr oposition\[t ensor\].**]{ }Fi rst no te th a t a t ang ent v ectort o $f(x ,y)$ is o ft he for m $(u\ot i m es y+x \otimes v,u \otim e s1 , 1 \otimes v) $ wher e$ u\ i n T_ x M $ and $v\i nT_y N $ . Fix a r bi t r a ryd istinct point s $(x_i,y_i ) \in M\tim es N, i =1,\ ldots ,k + l$and arbitra ry n on-zero v ectors $(u_j,v _ j)\in T _{(x_j ,y_ j)} M\ti m es N , j =k + 1,\ l do ts, k +l. $ Becaus eof rema rk \ [ o p e rati ve\ ], i t suf fices to show , a fter the iden tific atio n$$(U\otime s V)\ op lus(U\otimes{{\ math bf R}})\o plu s( V\o ti mes { { \mathb f R }}) \oplus{{\math b f R }} \ c o ng (U\oplus {{\mathb fR } }) \otimes(V \op l us { { \mathbfR} })$ $ th a t thevect o rs in theset $$ \ ma th cal S= \{ \tilde {x _i} \ot imes\ tild e{y_i} | i=1,\l dots, k +l\} \cup \{\t i lde u_j \otim e s\ t il d e y_ j + \tilde x_j \ot i mes\til d ev_j | j=k+ 1,\ld ot s ,k + l\}$$ are linearlyin depend ent.
Since $M$ is $(k,l)$-r e g u lar, weknow th a t the vectorsin $\ mathcal M= \ {\tilde{x_i} , \tilde {u_j}|u_j \ n eq0\}$ a relin ear lyi n de pendent and s i m ilar ly for $\ mat hcal N= \{\ til de{y_ i} , \tilde{v_j}|v_ j\ ne q0 \} .$But n o w we obs er veth atevery nontri vialline ar c o mbi nationo fv e ctor sin $\m ath ca l S$woul d al so be a nontrivi all inea rco mbinati on of vectors i n $$\{e\ot im esf| e\i n \mathcal M,f \in \mathcal N\},$ $ and th atis no t po ssible. [ $\B ox$]{}
[ * *Proof of Th eorem \ [ma i n \].** ] { }Fir st we provet h e l owerbo und. We will make use of the f o llo wing theorem, af fine | ,j=1,\ldots,l$, the_vectors $\tilde{x_1},\ldots,\tilde{x_k},\tilde{y_1},\ldots,\tilde{y_l},\tilde{u_1},\ldots,\tilde{u_l}\in_{{\mathbf R}}^{N+1}$ are linearly_independent.]{}
[**Proof of_Proposition_\[tensor\].**]{}First note_that_a tangent vector_to $f(x,y)$ is_of the form $(u\otimes_y+x\otimes v,u \otimes_1,_1 \otimes v)$ where $u\in T_x M$ and $v\in T_y N$. Fix arbitrary distinct_points_$(x_i,y_i)\in M\times_N,_i_=1,\ldots,k+l$ and arbitrary non-zero vectors_$(u_j,v_j)\in T_{(x_j,y_j)}M\times N, j=k+1,\ldots,k+l.$ Because_of remark_\[operative\], it suffices to show, after the identification_$$(U_\otimes V)\oplus(U\otimes{{\mathbf R}})\oplus(V\otimes_{{\mathbf R}})\oplus {{\mathbf R}}\cong (U\oplus {{\mathbf R}})\otimes (V \oplus_{{\mathbf R}})$$ that the vectors in_the set $$\mathcal__S=\{\tilde{x_i}\otimes_\tilde{y_i}| i=1,\ldots,k+l\} \cup \{\tilde_u_j \otimes \tilde y_j + \tilde_x_j \otimes \tilde v_j| j=k+1,\ldots,k+l\}$$ are_linearly independent.
Since $M$ is $(k,l)$-regular, we know_that the vectors in $\mathcal M=\{\tilde_{x_i}, \tilde{u_j}|u_j\neq0\}$ are linearly independent_and similarly_for $\mathcal N=\{\tilde {y_i}, \tilde_{v_j}|v_j\neq0\}.$ But now_we observe_that every nontrivial_linear combination of vectors in $\mathcal_S$ would also_be a nontrivial linear combination of_vectors_in $$\{e\otimes f|_e\in_\mathcal_M,f \in_\mathcal N\},$$ and_that_is not_possible._[$\Box$]{}
[**Proof of Theorem \[main\].**]{} First we_prove_the lower bound.We will make use of_the following theorem, affine |
j=1]{}\^[N]{} (+łV\_[1]{}) g\_[ij]{}(Q) where the metric $g_{ij}$ is defined in (\[eq:samolst\]). As a consequence of the exponential decay of static solutions, the metric quickly becomes flat as the vortex separation increases, and $T(Q,\dot{Q})$ quickly approaches (\[eq:tapprox\]).
In summary, the evolution of a configuration of vortices in a scattering problem is expected to be well approximated by a an evolution of particles (the zeros of the Higgs field) according to the action \[eq:ourlag\] S = dt (T(Q,) - łV(Q)) where $T(Q,\dot{Q})$ is defined in (\[eq:kindef\]) and $V(Q)$ is defined in (\[eq:potdef\]).
In a recent paper [@stuart], Stuart has established some rigorous results on vortex dynamics. An existence theorem is proved for an initial value problem in configuration space, where the initial data is $O(|\l|)$ close to the moduli space, the initial field derivatives are $O(|\l|^{\half})$, and $O(|\l|)$ close to the translational zero modes (at critical coupling) on the moduli space. For a subsequent time $T^{*} = |\l|^{-\half}$ the evolution continues to satisfy the same conditions as the initial data. The point on the moduli space of closest approach to the field configuration is $O(|\l|^{\half})$ close to a Lagrangian motion given by the metric on the moduli space [*at critical coupling*]{} (\[eq:samolst\]) and the potential $V(Q)$ from (\[eq:potdef\]). This applies for all vortex configurations, not just for well separated vortices. However, the metric used is incorrect asymptotically, as the inertial mass of the vortices changes with $\l$. This contributes to the $O(|\l|^{\half})$ drift from the Hamiltonian evolution on the moduli space (this can be seen immediately by considering the motion of a single boosted near-critical vortex in the context of this problem). We anticipate that Stuart’s results would still hold if our metric (\[eq:kindef\]) was used together with the potential to give the Lagrangian evolution on the moduli space. Furthermore, in a scattering problem where the vortices are well separated except for a time of $O(T^{*})$, we conjecture that Stuart’s results may be improved to hold for all times | j=1]{}\^[N ] { } (+ łV\_[1 ] { }) g\_[ij]{}(Q) where the metric $ g_{ij}$ is defined in (\[eq: samolst\ ]). As a consequence of the exponential decay of static solutions, the system of measurement promptly becomes flat as the vortex legal separation increases, and $ T(Q,\dot{Q})$ quickly approach (\[eq: tapprox\ ]).
In summary, the evolution of a configuration of whirl in a scattering problem is expect to be well approximated by a an evolution of particle (the zero of the Higgs field) according to the action \[eq: ourlag\ ] S = dt (T(Q ,) - łV(Q) ) where $ T(Q,\dot{Q})$ is defined in (\[eq: kindef\ ]) and $ V(Q)$ is defined in (\[eq: potdef\ ]).
In a late paper [ @stuart ], Stuart has established some rigorous results on whirlpool dynamics. An existence theorem is prove for an initial value trouble in configuration space, where the initial data is $ O(|\l|)$ close to the moduli space, the initial discipline derivatives are $ O(|\l|^{\half})$, and $ O(|\l|)$ close to the translational zero modes (at critical coupling) on the moduli quad. For a subsequent time $ T^ { * } = |\l|^{-\half}$ the evolution continues to satisfy the same conditions as the initial data. The point on the moduli space of close approach to the field shape is $ O(|\l|^{\half})$ cheeseparing to a Lagrangian apparent motion given by the metric on the moduli space [ * at critical coupling * ] { } (\[eq: samolst\ ]) and the potential $ V(Q)$ from (\[eq: potdef\ ]). This applies for all whirl configurations, not just for well separated vortices. However, the metric used is faulty asymptotically, as the inertial mass of the vortices changes with $ \l$. This contributes to the $ O(|\l|^{\half})$ drift from the Hamiltonian development on the moduli outer space (this can be seen immediately by considering the motion of a single boosted about - critical vortex in the context of this problem). We anticipate that Stuart ’s results would still hold if our metric (\[eq: kindef\ ]) was used together with the electric potential to give the Lagrangian evolution on the moduli space. Furthermore, in a scattering trouble where the vortices are well separated except for a time of $ O(T^{*})$, we conjecture that Stuart ’s consequence may be improved to accommodate for all times | j=1]{}\^[N]{} (+łG\_[1]{}) g\_[ij]{}(Q) where the metric $n_{ij}$ is defined iu (\[eq:sammlst\]). Aa a consdquence of the exponential dxcay of suctic solutions, the mdtric quibkly becones hlat as the vortxs separation jkcreavxs, and $T(Q,\dot{Q})$ qoickly approdches (\[eq:tapprof\]).
Iv dummary, the evolution of a configurwtion og gortices in a fcatueryng lgonlem is expected to be well apprkximatev by a an evoluyion of particles (the zerod of the Higgs field) afcording to the qction \[eq:ouruag\] S = dt (U(Q,) - łV(Q)) where $T(Q,\dot{Q})$ is defined in (\[eq:kindef\]) xnd $V(A)$ is defingb in (\[gq:potdef\]).
In a recegt paper [@stuart], Stuast has rstablished soke cigoeous results on vortee dynamics. An existegce theoram is proved for an ibitian vanue oeobuem ii cknfigugatmon space, wgere the inutial data is $O(|\l|)$ clpsq to the modulj spacq, ehe initial field derivatives are $O(|\l|^{\halx})$, ahd $O(|\l|)$ close to the tranwlational zero modes (wt criticwl coupling) on the moduli space. For a subsequent dime $V^{*} = |\l|^{-\hcof}$ thd egolution continues to satisfy the same condityknx ss the initial data. The ppijt jn the moduli space of closest approach ho the sield configurwtiom is $O(|\l|^{\half})$ close to a Lagrqngian motiou guven by the metric on the modbli spsce [*ay critical coupling*]{} (\[eq:scmolst\]) and the pohential $V(S)$ from (\[eq:potdef\]). Tfis a[plies for all vortex confygurationw, noc just fur wgll sepwrated vorhices. However, the metric used iv incorrech asymptotically, as the inertial mass of the fostibes changzs witm $\l$. This contrifutes to the $O(|\k|^{\half})$ dxift ffom the Hajiltonien evolution on the modulh space (this ran be seqn inmeduately ch considering yhe motiou of q single boosted ncar-crkfical vortex in tht xontext of thix pfobjel). Xe anehcipate that Stuxrt’r resuuts would sbilu hokd if our metric (\[eq:khnder\]) was used togethet rith the potentiwl to give thr Lagrangian evoluuion oi the oodulo skace. Furthermore, in a scatterinf problem whcre the vortises qre well sepcrated except for a time of $O(T^{*})$, we conjecvure that Stuart’s resulrs may be improved jo mold for all timef | j=1]{}\^[N]{} (+łV\_[1]{}) g\_[ij]{}(Q) where the metric $g_{ij}$ in As a of the exponential metric becomes flat as vortex separation increases, $T(Q,\dot{Q})$ quickly approaches (\[eq:tapprox\]). In summary, evolution of a configuration of vortices in a scattering problem is expected to well approximated by a an evolution of particles (the zeros of the Higgs according the \[eq:ourlag\] = dt (T(Q,) - łV(Q)) where $T(Q,\dot{Q})$ is defined in (\[eq:kindef\]) and $V(Q)$ is defined in In a recent paper [@stuart], Stuart has established rigorous results on vortex An existence theorem is proved an value problem configuration where initial data is close to the moduli space, the initial field derivatives are $O(|\l|^{\half})$, and $O(|\l|)$ close to the translational modes (at on the space. a time $T^{*} = evolution continues to satisfy the same initial data. The point on the moduli space closest approach the field configuration is $O(|\l|^{\half})$ close a Lagrangian motion given by the metric on moduli space [*at critical coupling*]{} (\[eq:samolst\]) and the potential $V(Q)$ from (\[eq:potdef\]). This applies for configurations, not just for separated vortices. However, metric is asymptotically, the inertial of the vortices changes with $\l$. This contributes to the $O(|\l|^{\half})$ from the Hamiltonian evolution on the moduli space (this can immediately considering the motion a single boosted near-critical in context of this problem). that results if metric was used together with potential to give the Lagrangian on the moduli space. where the vortices are well separated except for time of $O(T^{*})$, we conjecture that Stuart’s may be improved to hold for all times | j=1]{}\^[N]{} (+łV\_[1]{}) g\_[ij]{}(Q) where the metric $g_{ij}$ Is defined iN (\[eq:saMolSt\]). AS a ConsEqueNce of the exponeNTial Decay of static solutions, The meTrIC quiCKlY becoMes flat AS tHE VorTeX sEpaRaTIoN incrEasEs, and $T(Q,\Dot{Q})$ quicklY apPrOaches (\[eq:tappROx\]).
in summary, tHe eVolution of a cOnfIguratIoN of VOrticEs iN a scaTterinG ProbleM is expectEd TO be welL ApproxiMATeD by a An evolution of partIClES (the zeros of the higgs fIeLD) aCCOrdIng To the actioN \[eQ:ourlAG\] S = dt (T(Q,) - łv(q)) wHERE $T(Q,\DOt{Q})$ is defined iN (\[eq:kindef\]) anD $v(Q)$ iS definEd In (\[eQ:Potdef\]).
in a reCeNT paPer [@stuart], StUart Has establIshed sOMe rigorOUs resulTs on voRteX dyNamiCS. AN eXisTeNCe tHEoRem IS prOved for aN iNiTial vAlue PROBLem iN coNfigUratiOn space, where tHe iNitiAL daTa is $O(|\L|)$ closE to tHe ModulI space, The inItIal field derivatIves Are $O(|\l|^{\half})$, And $o(|\l|)$ CloSe To the TRanslaTioNal Zero modEs (at criTIcaL cOUPLiNg) on the moduli space. foR A SuBsequent Time $T^{*} = |\l|^{-\HAlF}$ tHE evolutiOn ConTinuES To satIsfy THe Same condItions AS tHe Initial DaTa. The pOiNt oN thE moduLI spaCe of clOsest appRoach TO the field confiGUration is $O(|\l|^{\haLF})$ cLOSe TO a LaGraNgian motion GiveN By thE metRIc On tHE moduLi spaCe [*AT cRItical coupling*]{} (\[eq:samOlSt\]) and tHe potEntial $V(Q)$ from (\[eQ:potdef\]). ThiS APPlies for All vORtEX configurationS, not jUst for well SEparated VortiCes. HowevEr, the metrIC Used is inCorRecT asYmpTOTiCally, as the ineRTIal mAsS of the vOrtIces chaNgeS wiTh $\l$. thiS cOntributeS to the $O(|\l|^{\HaLf})$ DrIfT frOm the hAmiltoniAn EvoLuTioN on thE Moduli Space (This CaN bE SeeN immediATeLY By coNsIdErinG thE mOtion Of a sINglE boosteD near-critIcaL VortEx In The contExt of this probLeM). We anticipAtE thAt StuaRT’S results Would still hold if our metrIC (\[eq:kindEf\]) wAs useD togEther with The PotentIal TO give tHe LagrAngiaN eVolUTIon on THE mOduLi Space. FurthERMorE, in a sCaTterIng probLem where the vorticeS Are Well separated ExcEpt fOR A tIme OF $O(t^{*})$, We cOnJEctURE that Stuart’s resUlts may be iMpROvEd to hold foR All TiMes | j=1]{}\^[N]{} (+łV\_[1]{}) g\_[ij]{} (Q) w her e t he met ric$g_{ij}$ is de f ined in (\[eq:samolst\]).As aco n sequ e nc e ofthe exp o ne n t ial d ec ayof st aticsol utions, the metri c q ui ckly becomes fl at as thevor tex separati onincrea se s,a nd $T (Q, \dot{ Q})$ q u icklyapproache s( \[eq:t a pprox\] ) .
In s ummary, the evolu t io n of a configur ationof vo r t ice s i n a scatte ri ng pr o blem is ex p e c ted to be well ap proximatedb y a an ev ol uti o n of p artic le s (t he zeros of the Higgs fi eld) a c cording to theaction \[ eq: ourl a g\ ]S = d t (T ( Q, ) - łV( Q)) wher e$T (Q,\d ot{Q } ) $ is d efi nedin (\ [eq:kindef\]) an d $V ( Q)$ is d efine d in ( \[eq: potdef \]).
I n a recent pape r [@ stuart],Stu ar t h as esta b lished so merigorou s resul t s o nv o r te x dynamics. An exi st e n ce theorem is pr o ve df or an in it ial val u e prob lemi nconfigur ations pa ce , where t he ini ti aldat a is$ O(|\ l|)$ c lose tothe m o duli space, th e initial fiel d d e r iv a tive s a re $O(|\l|^ {\ha l f})$ , an d $ O(| \ l|)$close t o t h e translational zer omodes(at c ritical coupl ing) on th e m oduli sp ace. Fo r a subsequenttime$T^{*} = | \ l|^{-\ha lf}$the evol ution con t i nues tosat isf y t hes a me conditions a s thein itial d ata . The p oin t o n t hemo duli spac e of clo se st a pp roa ch to the fiel dcon fi gur ation is $O( |\l|^ {\ha lf }) $ cl ose toa L a g rang ia nmoti ongi ven b y th e me tric on the modu lis pace [ *a t criti cal coupling* ]{ } (\[eq:sa mo lst \]) an d the pote ntial $V(Q)$ from (\[eq : potdef\ ]). This app lies forall vorte x c o nfigur ations , not j ust f or we l l s epa ra ted vortic e s . H oweve r, the metric used is incorrect asy mptotically,asthei n er tia l m a ssof the v ortices changes with $\l$ .T hi s contribu t esto the $O (|\l|^{ \half } )$ drif t from th e Hamilto ni an e v o lut ion on the modulispace (th i s can be seen im mediat el y b y con sideri n g t he mo tion o fa sing le bo os ted near -critical vortex in the conte xt of th is proble m). Weanticipat e th at Stuart’ s r esu lts w oul d stil l ho l difo ur me tric (\[eq:kin d ef \]) w as used toget h e r wi th th e p o tentia l to give the Lagrang i an evolution o n th e mod uli spac e. Furthermore,inas c attering p roblem wher e the vo rt i ces a re wel l sepa rated e x c ep t for a tim e o f $O(T^{* })$ ,w e conje ct ur e thatStua rt ’s res ults m a y be i mproved to holdfor a l l time s | j=1]{}\^[N]{} (+łV\_[1]{})_g\_[ij]{}(Q) where_the metric $g_{ij}$ is_defined in_(\[eq:samolst\])._As a_consequence_of the exponential_decay of static_solutions, the metric quickly_becomes flat as_the_vortex separation increases, and $T(Q,\dot{Q})$ quickly approaches (\[eq:tapprox\]).
In summary, the evolution of a configuration_of_vortices in_a_scattering_problem is expected to be_well approximated by a an_evolution of_particles (the zeros of the Higgs field) according_to_the action \[eq:ourlag\]_S = dt (T(Q,) - łV(Q)) where $T(Q,\dot{Q})$ is_defined in (\[eq:kindef\]) and $V(Q)$ is_defined in (\[eq:potdef\]).
In_a_recent_paper [@stuart], Stuart has_established some rigorous results on vortex_dynamics. An existence theorem is proved_for an initial value problem in configuration_space, where the initial data is_$O(|\l|)$ close to the moduli_space, the_initial field derivatives are $O(|\l|^{\half})$,_and $O(|\l|)$ close_to the_translational zero modes_(at critical coupling) on the moduli_space. For a_subsequent time $T^{*} = |\l|^{-\half}$ the_evolution_continues to satisfy_the_same_conditions as_the initial data._The_point on_the_moduli space of closest approach to_the_field configuration is $O(|\l|^{\half})$ close to a_Lagrangian motion given by_the_metric on the moduli_space [*at critical coupling*]{} (\[eq:samolst\])_and the potential $V(Q)$ from (\[eq:potdef\])._This applies_for all_vortex configurations, not just for well separated vortices. However, the metric_used is incorrect asymptotically, as the_inertial mass of the_vortices changes_with_$\l$. This contributes_to_the $O(|\l|^{\half})$_drift from the Hamiltonian evolution on the_moduli space_(this can be seen immediately by_considering the motion of_a_single boosted near-critical vortex in the_context of this problem). We anticipate_that Stuart’s results would still_hold_if_our metric (\[eq:kindef\]) was used_together with the potential to give_the Lagrangian evolution_on the moduli space. Furthermore, in a_scattering_problem where the vortices are well_separated_except for a time of $O(T^{*})$,_we_conjecture_that Stuart’s results may be_improved to hold for all times |
2$ $10.2$ $0.505$ $0.0330$ $0.0210$ $0.0346$ $0$ $4.60$
\[\] $(12)$ $4.56$ $13.9$ $17.7$ $10.1$ $4.48$ $6.28$ $3.57$ $4.56$ $4.54$ $4.56$ $4.32$ $0$
\[\]
--------------------------- ------------ ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
: Average values of $\sqrt{|\Delta \chi^2|} \; {\rm sign}(\Delta \chi^2)$ from a Monte Carlo simulation at $N_{\rm side} = 64$, with full-sky coverage, adopting $E_{\rm for} = 10^{-2}$ and $E_{\rm cal} = 10^{-4}$, and considering each of the 19 frequency channels.[]{data-label="MC_corr_avg"}
[width=1]{}
--------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
[**$\sigma$ level**]{}
\[\] [**significance**]{} blackbody (units $10^{-5}$) $8 \times 10^{-8}$ $10^{-5}$ $10^{-6}$ $10^{-9}$ $2 \times 10^{-8}$ $-2 \times 10^{-9}$ $4 \times 10^{-7}$ $8 \times 10^{-6}$
$I_0^{\rm bf}$ $+ 1\,\sigma$ $ - 1\,\sigma$
\[\] $1.3$ $1.7$ $0.9$ $1.12 \times 10^{-7}$ $1.4 \times 10^{-5}$ $1.4 \times 10^{-6}$ $1.4 \times 10^{-9}$ $2.8 \times 10^{-8}$ $-2.8 \times 10^{-9}$ $10^{-7}$ $2 \times 10^{-6}$
Case (1) (2) | 2 $ $ 10.2 $ $ 0.505 $ $ 0.0330 $ $ 0.0210 $ $ 0.0346 $ $ 0 $ $ 4.60 $
\[\ ] $ (12)$ $ 4.56 $ $ 13.9 $ $ 17.7 $ $ 10.1 $ $ 4.48 $ $ 6.28 $ $ 3.57 $ $ 4.56 $ $ 4.54 $ $ 4.56 $ $ 4.32 $ $ 0 $
\[\ ]
--------------------------- ------------ ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
: Average values of $ \sqrt{|\Delta \chi^2| } \; { \rm sign}(\Delta \chi^2)$ from a Monte Carlo simulation at $ N_{\rm side } = 64 $, with wide - sky coverage, assume $ E_{\rm for } = 10^{-2}$ and $ E_{\rm cal } = 10^{-4}$, and considering each of the 19 frequency channels.[]{data - label="MC_corr_avg " }
[ width=1 ] { }
--------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
[ * * $ \sigma$ level * * ] { }
\[\ ] [ * * significance * * ] { } black body (units $ 10^{-5}$) $ 8 \times 10^{-8}$ $ 10^{-5}$ $ 10^{-6}$ $ 10^{-9}$ $ 2 \times 10^{-8}$ $ -2 \times 10^{-9}$ $ 4 \times 10^{-7}$ $ 8 \times 10^{-6}$
$ I_0^{\rm bf}$ $ + 1\,\sigma$ $ - 1\,\sigma$
\[\ ] $ 1.3 $ $ 1.7 $ $ 0.9 $ $ 1.12 \times 10^{-7}$ $ 1.4 \times 10^{-5}$ $ 1.4 \times 10^{-6}$ $ 1.4 \times 10^{-9}$ $ 2.8 \times 10^{-8}$ $ -2.8 \times 10^{-9}$ $ 10^{-7}$ $ 2 \times 10^{-6}$
Case (1) (2) | 2$ $10.2$ $0.505$ $0.0330$ $0.0210$ $0.0346$ $0$ $4.60$
\[\] $(12)$ $4.56$ $13.9$ $17.7$ $10.1$ $4.48$ $6.28$ $3.57$ $4.56$ $4.54$ $4.56$ $4.32$ $0$
\[\]
--------------------------- ------------ ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
: Average values of $\sqrt{|\Dalca \chi^2|} \; {\rm sign}(\Delta \xhu^2)$ frok a Kontd Caflo smmumation at $N_{\rm side} = 64$, with full-wky coverage, adoptimg $V_{\tm for} = 10^{-2}$ and $E_{\rm cwl} = 10^{-4}$, and considering each of the 19 frequenby cgannels.[]{data-label="MC_corr_acg"}
[width=1]{}
--------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
[**$\sigma$ level**]{}
\[\] [**signifibancw**]{} blackbody (onits $10^{-5}$) $8 \tnmes 10^{-8}$ $10^{-5}$ $10^{-6}$ $10^{-9}$ $2 \times 10^{-8}$ $-2 \times 10^{-9}$ $4 \times 10^{-7}$ $8 \cimes 10^{-6}$
$I_0^{\rm bf}$ $+ 1\,\sigma$ $ - 1\,\sigma$
\[\] $1.3$ $1.7$ $0.9$ $1.12 \tines 10^{-7}$ $1.4 \times 10^{-5}$ $1.4 \gimqs 10^{-6}$ $1.4 \dimes 10^{-9}$ $2.8 \tikes 10^{-8}$ $-2.8 \yimes 10^{-9}$ $10^{-7}$ $2 \times 10^{-6}$
Wase (1) (2) | 2$ $10.2$ $0.505$ $0.0330$ $0.0210$ $0.0346$ $0$ $(12)$ $13.9$ $17.7$ $4.48$ $6.28$ $3.57$ \[\] ------------ ---------------- ------------------- ----------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- -------------------- : Average of $\sqrt{|\Delta \chi^2|} \; {\rm sign}(\Delta \chi^2)$ from a Monte Carlo simulation at side} = 64$, with full-sky coverage, adopting $E_{\rm for} = 10^{-2}$ and $E_{\rm = and each the 19 frequency channels.[]{data-label="MC_corr_avg"} [width=1]{} --------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- [**$\sigma$ level**]{} \[\] [**significance**]{} blackbody (units $10^{-5}$) \times 10^{-8}$ $10^{-5}$ $10^{-6}$ $2 \times 10^{-8}$ $-2 \times $4 10^{-7}$ $8 10^{-6}$ bf}$ 1\,\sigma$ $ - \[\] $1.3$ $1.7$ $0.9$ $1.12 \times 10^{-7}$ $1.4 \times 10^{-5}$ $1.4 \times 10^{-6}$ $1.4 \times 10^{-9}$ $2.8 10^{-8}$ $-2.8 $10^{-7}$ $2 10^{-6}$ (1) | 2$ $10.2$ $0.505$ $0.0330$ $0.0210$ $0.0346$ $0$ $4.60$
\[\] $(12)$ $4.56$ $13.9$ $17.7$ $10.1$ $4.48$ $6.28$ $3.57$ $4.56$ $4.54$ $4.56$ $4.32$ $0$
\[\]
--------------------------- ------------ ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
: Average values of $\sqrt{|\Delta \cHi^2|} \; {\rm sign}(\DeLta \chI^2)$ frOm a moNte CArlo Simulation at $N_{\rM Side} = 64$, With full-sky coverage, adoPting $e_{\rM For} = 10^{-2}$ aND $E_{\Rm cal} = 10^{-4}$, And consIDeRINg eAcH oF thE 19 fREqUency ChaNnels.[]{daTa-label="MC_cOrr_AvG"}
[width=1]{}
--------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ---------------------- ---------------------- ----------------------- -------------------- --------------------
[**$\sigma$ lEVeL**]{}
\[\] [**significaNce**]{} Blackbody (uniTs $10^{-5}$) $8 \tImes 10^{-8}$ $10^{-5}$ $10^{-6}$ $10^{-9}$ $2 \tiMeS 10^{-8}$ $-2 \tiMEs 10^{-9}$ $4 \timEs 10^{-7}$ $8 \tImes 10^{-6}$
$I_0^{\Rm bf}$ $+ 1\,\siGMa$ $ - 1\,\sigmA$
\[\] $1.3$ $1.7$ $0.9$ $1.12 \times 10^{-7}$ $1.4 \timEs 10^{-5}$ $1.4 \TImes 10^{-6}$ $1.4 \tiMEs 10^{-9}$ $2.8 \times 10^{-8}$ $-2.8 \TIMeS 10^{-9}$ $10^{-7}$ $2 \timEs 10^{-6}$
Case (1) (2) | 2$ $10.2$ $ 0.5 05$ $0.0330$ $0.0210$ $ 0 .0 346$ $ 0$ $4.60$
\[ \] $(12)$ $4.56$ $13.9$ $1 7 .7$ $10 . 1$ $ 4 .48$ $6.28$ $3.57$ $ 4 .56 $ $4 . 54$ $4.56$ $4 .32$ $0$
\[\ ]
--- -- - - - -- ------------------- - - -- ------ - ------ - -- -- - --- ---- -- --- ---- - - -------- - -- -------- - ---- - -- -- ------- -- ------- --- --- ----- - ---- ---- - -------- ----- - ------- ------ - ------------- - -- - -- - ---- --- -------------- - ---- ---- - -- --- - --- - ----- -- - -- - -------- ---------- -- ------ --
: Average va lues of $\ s q r t{|\Delt a \c h i^ 2 |} \; {\rm sig n}(\D elta \chi^ 2 )$ froma Mon te Carlo simulati o n at $N_{ \rm si de} =6 4 $, with full-sk y cove ra ge, ado pti ng $E_{ \rm fo r}= 1 0^ {-2}$ and $E_{\rm c al }=10^ {-4}$ , and con si der in g e ach o f the 1 9 fre quen cy c h ann els.[]{ d at a - labe l= "M C_co rr_ av g"}
[wid t h=1 ]{}
--------- --- - ---- -- -- ----------------- - -- ---------- -- - - ------ - - -------- -- ---------------- --- - ------- --- ----- ---- -------- --- ------ --- - - ---- ------ ----- -- --- - - ---- - - -- --- -- ---------- - --- ----- -- ---- ------------------------ - --- - ----------- --- ---- - - - --- - -- - --- -- - --- - - -
[**$\sigm a$ level** ]{ }
\[\][**s ignificanc e** ]{} bl ack b ody (u n its $10^{ - 5} $) $ 8 \ti mes 10^{-8 }$ $10^{-5}$ $ 10^{ - 6 }$ $10^{-9}$ $2 \tim es 10^{-8}$ $-2 \ ti m es 10 ^{-9}$ $4 \times 1 0^ { -7}$ $8\ti mes 10^{- 6}$
$ I _0^{\rm bf}$ $+ 1 \ , \sigm a $ $-1\,\sig m a$
\[\ ] $1.3$ $1.7$ $0.9$ $1.1 2 \time s 10 ^{ -7}$ $1. 4 \times 10 ^{-5}$ $ 1.4 \ time s 10^{-6 }$ $ 1. 4 \ ti mes 10^{-9}$ $2.8 \ti mes 10^{ - 8 }$ $ -2. 8 \ti m es 10^ {- 9}$ $1 0^{-7 }$ $ 2 \t imes 10 ^ {-6 }$ C a se (1) ( 2) | 2$ _ _ _ __ __ $10.2$_ _ _ __ $0.505$ __ ___ $0.0330$ _ _ _ $0.0210$ __ _ $0.0346$ _ _ ___ _$0$ _ _ $4.60$
_ _ \[\] $(12)$ _ _ _ _$4.56$ _ _ $13.9$ _ _ __$17.7$ ___ _ __ _$10.1$_ __ $4.48$_ __ _ _ $6.28$ _ _ _ $3.57$_ _ _ __ _$4.56$_ _ _ _ $4.54$_ __ _ $4.56$_ ___ _ $4.32$ _ _ __ $0$
__ ___ \[\] _ _ _ _ ___ _ _ _ _ _ _ _ _ __ ____ _ _ _ _ _ _ __ _ _ __ _ _ _ __ _ _ _ _ _ _ _ _ __ ____ _ _ _ __ _ _
_ --------------------------- ------------ ---------------- ------------------- ---------------- ----------------------- ---------------------- ---------------------- ----------------------_---------------------- ----------------------- -------------------- --------------------
: Average values of $\sqrt{|\Delta \chi^2|} \; {\rm sign}(\Delta \chi^2)$ from a_Monte Carlo_simulation at $N_{\rm side}_= 64$,_with_full-sky coverage, adopting $E_{\rm for} =_10^{-2}$ and_$E_{\rm cal} = 10^{-4}$, and considering each of_the_19 frequency channels.[]{data-label="MC_corr_avg"}
[width=1]{}
--------------------------- ----------- ---------------- ------------------- ---------------- ----------------------- ---------------------- ----------------------_----------------------_----------------------_----------------------- -------------------- --------------------
__ [**$\sigma$ level**]{} _____ _ __ _ __ _ _ __ ____ _ __ _ __ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ __ _
_\[\] [**significance**]{} _blackbody __ _ _ _ _ (units $10^{-5}$) _ _ _ $8 \times 10^{-8}$ $10^{-5}$ _ __ $10^{-6}$ _ _ $10^{-9}$ _ $2 \times 10^{-8}$ ___ $-2 \times 10^{-9}$__ $4 \times 10^{-7}$_ $8 \times_10^{-6}$
_ _ _ _ _ __ _ __ _ __ $I_0^{\rm_bf}$ _ __$+ 1\,\sigma$_ __ $_- 1\,\sigma$__ _ _ _ __ _ _ _ _ _ _ _ __ _ _ __ _ _ _ _ _ _ ___ _ _ _ __ _ _ _ __ __ _
_ _ _ \[\] ___ _ __ _ $1.3$_ _ $1.7$ ___ __ $0.9$ _ $1.12 \times_10^{-7}$_ $1.4 \times 10^{-5}$ $1.4 \times_10^{-6}$ $1.4 \times 10^{-9}$ $2.8 \times 10^{-8}$ $-2.8 \times 10^{-9}$ __ $10^{-7}$ ___ _ $2 \times_10^{-6}$
_ _ _ _Case (1) _ _ (2) _ _ __ _ _ |
over the set of initial conditions to find $\lambda$. If this asymptotic value is positive, the system is defined as chaotic. To put a stricter criterion on the degree of chaos, we choose a threshold value of the exponent $\lambda_c$ beyond which the system is in the regime of hard chaos. We set $\lambda_c$ arbitrarily to the value 1, but as a check we generated Poincaré phase portraits to confirm chaos by looking for featureless ([*i.e.,*]{} no invariant tori) phase portraits. In this way, by varying the parameters $(\epsilon,{\tilde\omega_c})$ in a continuous fashion over the whole plane, running the map repeatedly and obtaining the resulting $\lambda$’s, we obtained the ‘phase diagram’ for this system, including a distinct ‘phase boundary’ separating the quasi-integrable and hard chaos regions. Of course, this phase boundary depends on the precise value of the cutoff $\lambda_c$ we choose. Nevertheless, we checked that on varying the cutoff $\lambda_c$, the phase boundary shifts only slightly and furthermore, the shape of the boundary remains qualitatively the same. Indeed, to a high degree of precision, the phase boundary can be fitted by $$\label{eq:eq18}
{\tilde\omega_c} = C(\lambda_c)/\epsilon,$$ where $C(\lambda_c)$ is a constant which depends on the value of the cutoff. Figure 1 shows the phase diagram for a cutoff $\lambda_c=1$.
We observe from the classical Poincaré sections that there is a symmetry line in the $(r,p_r)$ plane. This arises from the time-reversal invariance present in the problem as follows. Consider a particle kicked at $\tau=0$. The position $r_0$ remains unchanged, while the momentum changes : $p_r^{(+)} = p_r^{(-)} + \eta\ r_0$. Denoting $p_r^{+}$ by $p_0$, then at time $0^{(-)}$ the particle had momentum $p_r^{(-)} = p_0 -\eta\
r_0$. Taking into account the fact that the angular momentum is conserved, we see that propagating a particle [*forward*]{} in time from $(r_0,p_0)$ is the same as propagating it [*backward*]{} from $(r_0,\eta\ r_0-p_0)$. Thus, the motion is symmetric about the line $p_ | over the set of initial conditions to find $ \lambda$. If this asymptotic value is positive, the arrangement is define as chaotic. To put a stricter criterion on the academic degree of chaos, we choose a threshold value of the exponent $ \lambda_c$ beyond which the system is in the government of hard chaos. We set $ \lambda_c$ randomly to the value 1, but as a confirmation we generated Poincaré phase portrait to confirm chaos by looking for featureless ([ * i.e., * ] { } no invariant torus) phase portrayal. In this way, by varying the parameters $ (\epsilon,{\tilde\omega_c})$ in a continuous fashion over the whole airplane, running the map repeatedly and obtaining the leave $ \lambda$ ’s, we obtained the ‘ phase diagram ’ for this system, include a distinct ‘ phase boundary ’ separate the quasi - integrable and hard chaos regions. Of course, this phase boundary depends on the accurate value of the cutoff $ \lambda_c$ we choose. Nevertheless, we checked that on varying the cutoff $ \lambda_c$, the phase boundary shift only slightly and furthermore, the shape of the boundary remains qualitatively the same. Indeed, to a high degree of precision, the phase boundary can be match by $ $ \label{eq: eq18 }
{ \tilde\omega_c } = C(\lambda_c)/\epsilon,$$ where $ C(\lambda_c)$ is a constant which depends on the value of the cutoff. name 1 show the phase diagram for a cutoff $ \lambda_c=1$.
We observe from the authoritative Poincaré sections that there constitute a symmetry line in the $ (radius, p_r)$ plane. This arises from the time - reversal invariance present in the problem as succeed. Consider a particle kicked at $ \tau=0$. The position $ r_0 $ remain unchanged, while the momentum change: $ p_r^{(+) } = p_r^{(-) } + \eta\ r_0$. announce $ p_r^{+}$ by $ p_0 $, then at time $ 0^{(-)}$ the particle had momentum $ p_r^{(-) } = p_0 -\eta\
r_0$. Taking into report the fact that the angular momentum is conserved, we see that propagating a particle [ * forward * ] { } in time from $ (r_0,p_0)$ is the same as propagating it [ * backward * ] { } from $ (r_0,\eta\ r_0 - p_0)$. Thus, the movement is symmetric about the line $ phosphorus _ | ovfr the set of initial cokditions to find $\lambda$. If thjs asympgotic value is positive, the dywtem us defined as chaotic. Go put a dtricter criuerion on the degcse of cmcos, ws chomwe a threshold value of dhe exponent $\ldmcdc_c$ beyond which the system is in the regime ov hard chaos. Wg set $\jambsa_c$ arbitrarily to the value 1, but zs a chtck we generated Ppincaré phase portraits to fonflrm chaos by lookijg for featodelqws ([*i.e.,*]{} no invxriant torp) phase porjraits. In this way, by varying thd parcmeters $(\epsulin,{\tlnde\omega_c})$ ii a cogtinuous fasmpon oves the wnole plane, runking vhe nap repeatedly and obvaining the resultind $\lambda$’s, wz obtained the ‘phase eiqgram’ for thir syrtej, mncmuding a vistinct ‘phzse boundart’ separating the qussy-pmtegrable ans hard craos regions. Of course, this phase bounddry depends on the precise value of the cutoff $\pambda_c$ wq choose. Nevertheless, we checked that on varying dhe cntuff $\lwovdw_c$, the phase boundary shifts only slightly anq firnhermore, the shapc of the boundary telaogs qualitativgly the saje. Indeed, to a higj degreg of peecision, uhe pnase boundary can be fitted by $$\label{eq:ez18}
{\rilde\omega_c} = C(\lambba_c)/\epsilon,$$ wkere $C(\kambds_c)$ is a constant which bependa on the vapue of ths cutoff. Figure 1 rhoes the phase diagram for a cttoff $\lamuda_c=1$.
Wz observd frpm the classical Poingdré sections that tjere ns a vymmetry llne in the $(r,p_r)$ plane. This arises from the timg-reeerval invaxiance present in tre problem as yollows. Eonsiddr a partible kickev at $\tau=0$. The position $r_0$ ralains unchanjed, while the momwntum cfxnges : $p_r^{(+)} = p_r^{(-)} + \eta\ r_0$. Deujtint $p_r^{+}$ by $p_0$, then at bime $0^{(-)}$ fhe particle hab nomentum $p_r^{(-)} = p_0 -\rta\
r_0$. Eaniig inem account tha fazt gnat tfe angular numenyum is conserved, we vee fhat propagating a pwrticle [*dorward*]{} yn time from $(t_0,p_0)$ is the same as kropageting mt [*baclwatd*]{} from $(r_0,\eta\ r_0-p_0)$. Thus, the motion is symmehrig about the lyne $i_ | over the set of initial conditions to If asymptotic value positive, the system put stricter criterion on degree of chaos, choose a threshold value of the $\lambda_c$ beyond which the system is in the regime of hard chaos. We $\lambda_c$ arbitrarily to the value 1, but as a check we generated Poincaré portraits confirm by for featureless ([*i.e.,*]{} no invariant tori) phase portraits. In this way, by varying the parameters $(\epsilon,{\tilde\omega_c})$ a continuous fashion over the whole plane, running map repeatedly and obtaining resulting $\lambda$’s, we obtained the diagram’ this system, a ‘phase separating the quasi-integrable hard chaos regions. Of course, this phase boundary depends on the precise value of the cutoff $\lambda_c$ choose. Nevertheless, that on the $\lambda_c$, phase boundary shifts and furthermore, the shape of the the same. Indeed, to a high degree of the phase can be fitted by $$\label{eq:eq18} {\tilde\omega_c} C(\lambda_c)/\epsilon,$$ where $C(\lambda_c)$ is a constant which depends the value of the cutoff. Figure 1 shows the phase diagram for a cutoff $\lambda_c=1$. from the classical Poincaré that there is symmetry in $(r,p_r)$ This arises the time-reversal invariance present in the problem as follows. Consider a kicked at $\tau=0$. The position $r_0$ remains unchanged, while the : = p_r^{(-)} + r_0$. Denoting $p_r^{+}$ by then time $0^{(-)}$ the particle $p_r^{(-)} p_0 into the that the angular momentum conserved, we see that propagating particle [*forward*]{} in time as propagating it [*backward*]{} from $(r_0,\eta\ r_0-p_0)$. Thus, motion is symmetric about the line $p_ | over the set of initial conditIons to find $\LambdA$. If ThiS aSympTotiC value is positiVE, the System is defined as chaotIc. To pUt A StriCTeR critErion on THe DEGreE oF cHaoS, wE ChOose a ThrEshold vAlue of the eXpoNeNt $\lambda_c$ beyONd Which the sySteM is in the regiMe oF hard cHaOs. WE Set $\laMbdA_c$ arbItrariLY to the Value 1, but aS a CHeck we GEnerateD pOiNcarÉ phase portraits to COnFIrm chaos by lookIng for FeATuRELesS ([*i.e.,*]{} No invarianT tOri) phASe portrAItS. iN ThiS Way, by varying tHe parameterS $(\EpsIlon,{\tiLdE\omEGa_c})$ in a ContiNuOUs fAshion over tHe whOle plane, rUnning THe map rePEatedly And obtAinIng The rESuLtIng $\LaMBda$’S, We ObtAIneD the ‘phasE dIaGram’ fOr thIS SYStem, IncLudiNg a diStinct ‘phase boUndAry’ sEParAting The quAsi-iNtEgrabLe and hArd chAoS regions. Of coursE, thiS phase bouNdaRy DepEnDs on tHE preciSe vAluE of the cUtoff $\laMBda_C$ wE CHOoSe. Nevertheless, we chEcKED tHat on varYing thE CuToFF $\lambda_c$, ThE phAse bOUNdary ShifTS oNly slighTly and FUrThErmore, tHe Shape oF tHe bOunDary rEMainS qualiTatively The saME. Indeed, to a high DEgree of precisIOn, THE pHAse bOunDary can be fiTted BY $$\labEl{eq:EQ18}
{\tIldE\Omega_C} = C(\lamBdA_C)/\ePSilon,$$ where $C(\lambda_c)$ iS a ConstaNt whiCh depends on thE value of thE CUToff. FiguRe 1 shOWs THe phase diagram For a cUtoff $\lambdA_C=1$.
We obserVe froM the clasSical PoinCARé sectioNs tHat TheRe iS A SyMmetry line in tHE $(R,p_r)$ pLaNe. This aRisEs from tHe tIme-RevErsAl InvariancE present In ThE pRoBleM as foLLows. ConsIdEr a PaRtiCle kiCKed at $\tAu=0$. The PosiTiOn $R_0$ RemAins uncHAnGED, whiLe ThE momEntUm ChangEs : $p_r^{(+)} = P_R^{(-)} + \etA\ r_0$. DenotIng $p_r^{+}$ by $p_0$, tHen AT timE $0^{(-)}$ tHe ParticlE had momentum $p_R^{(-)} = p_0 -\Eta\
r_0$. Taking InTo aCcount THE fact thaT the angular momentum is coNServed, wE seE that PropAgating a pArtIcle [*foRwaRD*]{} in timE from $(r_0,P_0)$ is thE sAme AS PropaGATiNg iT [*bAckward*]{} froM $(R_0,\Eta\ R_0-p_0)$. ThuS, tHe moTion is sYmmetric about the liNE $p_ | over the set of initial c onditionsto fi nd$\l am bda$ . If this asymptot i c va lue is positive, the s ystem i s def i ne d aschaotic . T o put a s tri ct e rcrite rio n on th e degree o f c ha os, we choos e a threshold va lue of the e xpo nent $ \l amb d a_c$bey ond w hich t h e syst em is inth e regim e of har d ch aos. We set $\lambda_ c $a rbitrarily tothe va lu e 1 , but as a check w egener a ted Poi n ca r é pha s e portraits t o confirm c h aos by lo ok ing for fe ature le s s ( [*i.e.,*]{} noinvariant tori) phase p o rtraits . In t his wa y, b y v ar yin gt hep ar ame t ers $(\epsi lo n, {\til de\o m e g a _c}) $ i n aconti nuous fashion ov er t h e w holeplane , ru nn ing t he map repe at edly and obtain ingthe resul tin g$\l am bda$’ s , we o bta ine d the ‘ phase d i agr am ’ f or this system, incl ud i n ga distin ct ‘ph a se b o undary’se par atin g the q uasi - in tegrable and h a rd c haos re gi ons. O fcou rse , thi s pha se bou ndary de pends on the precise value of thec ut o f f$ \lam bda _c$ we choo se.N ever thel e ss , w e chec ked t ha t o n varying the cutoff $ \lambd a_c$, the phase bo undary shi f t s only sl ight l ya nd furthermore , the shape oft he bound ary r emains q ualitativ e l y the sa me. In dee d,t o a high degreeo f pre ci sion, t hephase b oun dar y c anbe fitted b y $$\lab el {e q: eq 18}
{\ti l de\omega _c } = C (\l ambda _ c)/\ep silon ,$$wh er e $C (\lambd a _c ) $ isaco nsta ntwh ich d epen d s o n the v alue of t hec utof f. F igure 1 shows the ph as e diagramfo r a cutof f $\lambda _c=1$.
We observe from the cla ssi cal P oinc aré secti ons thatthe r e is a symme try l in e i n the $ ( r ,p _r) $plane. Thi s ari ses f ro m th e time- reversal invarianc e pr esent in thepro blem a sfol l ow s . C on s ide r a particle kick ed at $\ta u= 0 $. The posit i on$r _0$ rem ains un chang e d, whil e the mom entum cha ng es : $ p_r ^{(+)} = p _r^{(-)} + \eta\r _0$.D en oting $p _r^{+} $by$p_0$ , then attime$0^{(- )} $ theparti cl e had mo mentum $p_r^{(-)} = p_0 -\eta \
r_ 0$. Taking i nto acc ount thefact that theang ula r mom ent u m iscons e rv ed, we se e th a t propaga t in g a p ar ticle [*for w a r d*] {} in ti m e from $(r _0,p_0)$ is the s a me as propagat ingi t [* bac k ward *] {} from $(r_0, \et a\ r _0-p_0)$ .Thus, the m otion is s y mmetr ic abo ut the line $ p _ | over_the set_of initial conditions to_find $\lambda$._If_this asymptotic_value_is positive, the_system is defined_as chaotic. To put_a stricter criterion_on_the degree of chaos, we choose a threshold value of the exponent $\lambda_c$ beyond_which_the system_is_in_the regime of hard chaos._We set $\lambda_c$ arbitrarily to_the value_1, but as a check we generated Poincaré_phase_portraits to confirm_chaos by looking for featureless ([*i.e.,*]{} no invariant tori)_phase portraits. In this way, by_varying the parameters_$(\epsilon,{\tilde\omega_c})$_in_a continuous fashion over_the whole plane, running the map_repeatedly and obtaining the resulting $\lambda$’s,_we obtained the ‘phase diagram’ for this_system, including a distinct ‘phase boundary’_separating the quasi-integrable and hard_chaos regions._Of course, this phase boundary_depends on the_precise value_of the cutoff_$\lambda_c$ we choose. Nevertheless, we checked_that on varying_the cutoff $\lambda_c$, the phase boundary_shifts_only slightly and_furthermore,_the_shape of_the boundary remains_qualitatively_the same._Indeed,_to a high degree of precision,_the_phase boundary can be fitted by $$\label{eq:eq18}
{\tilde\omega_c}_= C(\lambda_c)/\epsilon,$$ where $C(\lambda_c)$_is_a constant which depends_on the value of the_cutoff. Figure 1 shows the phase_diagram for_a cutoff_$\lambda_c=1$.
We observe from the classical Poincaré sections that there is a_symmetry line in the $(r,p_r)$ plane._This arises from the_time-reversal invariance_present_in the problem_as_follows. Consider_a particle kicked at $\tau=0$. The position_$r_0$ remains_unchanged, while the momentum changes :_$p_r^{(+)} = p_r^{(-)} +_\eta\_r_0$. Denoting $p_r^{+}$ by $p_0$, then_at time $0^{(-)}$ the particle had_momentum $p_r^{(-)} = p_0 -\eta\_
r_0$._Taking_into account the fact that_the angular momentum is conserved, we_see that propagating_a particle [*forward*]{} in time from $(r_0,p_0)$_is_the same as propagating it [*backward*]{}_from_$(r_0,\eta\ r_0-p_0)$. Thus, the motion is_symmetric_about_the line $p_ |
been possible thanks to generous support from the Italian Space Agency via contract I/038/080/0. Data presented in this paper were also analyzed using The Herschel interactive processing environment (HIPE), a joint development by the Herschel Science Ground Segment Consortium, consisting of ESA, the NASA Herschel Science Center, and the HIFI, PACS, and SPIRE consortia. This material is based upon work supported by the National Science Foundation under Award No. 1602583 and by NASA through an award issued by JPL/Caltech via NASA Grant \#1350780.
[^1]: For an O6 star, the main sequence lifetime is about 2.4 $\times$ 10$^{6}$ years [@mae87], so 15% is 3.6 $\times$ 10$^{5}$ years. Note that a less massive mid-B star would have a lifetime about 5$\times$ longer, changing our absolute lifetime estimates by that factor. This variation gives a sense of the uncertainties involved in deriving absolute lifetimes.
[^2]: @rob08, the Red MSX Catalog from @urq08, and the EGO catalog from @cyg08.
---
abstract: 'We study supersymmetric black holes in $AdS_4$ in the framework of four dimensional gauged $\N=2$ supergravity coupled to hypermultiplets. We derive the flow equations for a general electrically gauged theory where the gauge group is Abelian and, restricting them to the fixed points, we derive the gauged supergravity analogue of the attractor equations for theories coupled to hypermultiplets. The particular models we analyze are consistent truncations of M-theory on certain Sasaki-Einstein seven-manifolds. We study the space of horizon solutions of the form $AdS_2\times \Sig_g$ with both electric and magnetic charges and find a four-dimensional solution space when the theory arises from a reduction on $Q^{111}$. For other $SE_7$ reductions, the solutions space is a subspace of this. We construct explicit examples of spherically symmetric black holes numerically.'
---
1.5cm Nick Halmagyi$^*$, Michela Petrini$^*$, Alberto Zaffaroni$^{\dagger}$\
0.5cm $^{*}$ Laboratoire de Physique Théorique et Hautes Energies,\
Université Pierre et Marie Curie, CNRS UMR 7589,\
| been possible thanks to generous support from the Italian Space Agency via contract I/038/080/0. Data presented in this newspaper were besides analyzed using The Herschel synergistic processing environment (HIPE), a joint exploitation by the Herschel Science Ground Segment Consortium, consisting of ESA, the NASA Herschel Science Center, and the HIFI, PACS, and SPIRE consortia. This fabric is based upon work supported by the National Science Foundation under Award No. 1602583 and by NASA through an prize issue by JPL / Caltech via NASA Grant \#1350780.
[ ^1 ]: For an O6 star, the independent sequence lifetime is approximately 2.4 $ \times$ 10$^{6}$ years [ @mae87 ], so 15% is 3.6 $ \times$ 10$^{5}$ years. Note that a less massive mid - B star would have a lifetime approximately 5$\times$ longer, changing our absolute lifetime estimates by that divisor. This variation gives a common sense of the uncertainties involve in deriving absolute life.
[ ^2 ]: @rob08, the Red MSX Catalog from @urq08, and the EGO catalog from @cyg08.
---
abstract:' We study supersymmetric black holes in $ AdS_4 $ in the framework of four dimensional gauge $ \N=2 $ supergravity coupled to hypermultiplets. We derive the flow equations for a general electrically gauge hypothesis where the gauge group is Abelian and, restricting them to the fixed points, we derive the gauged supergravity analogue of the attractor equations for theory match to hypermultiplets. The particular models we analyze are coherent truncation of M - theory on certain Sasaki - Einstein seven - manifold. We study the space of horizon solutions of the form $ AdS_2\times \Sig_g$ with both electric and charismatic charges and find a four - dimensional solution space when the theory arises from a reduction on $ Q^{111}$. For other $ SE_7 $ reductions, the solutions space is a subspace of this. We construct explicit model of spherically symmetrical bootleg fix numerically.'
---
1.5 cm Nick Halmagyi$^*$, Michela Petrini$^*$, Alberto Zaffaroni$^{\dagger}$\
0.5 cm $ ^{*}$ Laboratoire de Physique Théorique et Hautes Energies,\
Université Pierre et Marie Curie, CNRS UMR 7589,\ | befn possible thanks to gekerous support fton the Mtalian Space Aeency via contract I/038/080/0. Data prxsenred ib this paper were also analyzed using Tye Htrschel interactits processing skviroumxnt (HIPE), a joinj developmend by the Herscveu Dcience Ground Segment Consortium, cjnsistimg of ESA, the NAFA Htrsshel Science Center, and the HIFI, PACS, znd SPIGE consortia. This material is based upon wogk skpported by the Nahional Sciebce Siundation unaer Award No. 1602583 and by NZSA through an award issued by GPL/Caktech via BAWA Htant \#1350780.
[^1]: For an O6 stwr, the main sequence lifetike is about 2.4 $\tlmes$ 10$^{6}$ yeqrs [@mae87], so 15% is 3.6 $\times$ 10$^{5}$ years. Note that w less mavsnve mid-B star would hqvw a lhfethme xvoug 5$\tjmxs$ monger, chenging our zbsolute lidetime estimates by trqt factor. Thia variwtyon gives a sense of the uncertainties pnvomved in deriving absolure lifetimes.
[^2]: @rob08, the Ted MSX Caealog from @urq08, and the EGO catalog from @cyg08.
---
abstrdct: 'Wx rtubn rypfrsymmetric black holes in $AdS_4$ in the framewowi pf four dimensiokal gauged $\N=2$ superbrwvojy coupled to fypermbmtjplets. We derive tje flow equarions for a grneral electrically gauged rheory where rhe gauge group is Abelian anb, resttictinb them to the fixed poiuts, we derive the gauged shoergravity analoeue ox the atufactor equations sor theormes cpupled go hipermuleiplets. Thf parbhcular models we ajalyzg are wonsistent truncations of M-theory on certemn Sasaki-Einsjeit svven-manifjlds. Ee study the fpace of horizpn solucions uf the forj $AdS_2\tikes \Sig_g$ wieh both electslc and magnevic chargqs abd fund a fujr-dimensional xolution space when the theory arises fruj a reduction ou $W^{111}$. For other $SE_7$ tedjctyojs, tra solutions vpacd ir a sucspace of tmis. We vonstruct explicit efampmes of spherically snmmetric vlack hojes numericalky.'
---
1.5cm Nick Halmagyi$^*$, Michxla Pevrini$^*$, Slbgrto Zaffaroni$^{\dagger}$\
0.5cm $^{*}$ Laboratkire de Pjyslque Théorique et Mautgs Energies,\
Bniversité Pierre et Marie Curie, CNRS UMR 7589,\
| been possible thanks to generous support from Space via contract Data presented in using Herschel interactive processing (HIPE), a joint by the Herschel Science Ground Segment consisting of ESA, the NASA Herschel Science Center, and the HIFI, PACS, and consortia. This material is based upon work supported by the National Science Foundation Award 1602583 by through an award issued by JPL/Caltech via NASA Grant \#1350780. [^1]: For an O6 star, the sequence lifetime is about 2.4 $\times$ 10$^{6}$ years so 15% is 3.6 10$^{5}$ years. Note that a massive star would a about longer, changing our lifetime estimates by that factor. This variation gives a sense of the uncertainties involved in deriving absolute [^2]: @rob08, MSX Catalog @urq08, the catalog from @cyg08. 'We study supersymmetric black holes in framework of four dimensional gauged $\N=2$ supergravity coupled hypermultiplets. We the flow equations for a general gauged theory where the gauge group is Abelian restricting them to the fixed points, we derive the gauged supergravity analogue of the attractor theories coupled to hypermultiplets. particular models we are truncations M-theory certain Sasaki-Einstein We study the space of horizon solutions of the form $AdS_2\times with both electric and magnetic charges and find a four-dimensional when theory arises from reduction on $Q^{111}$. For $SE_7$ the solutions space is of We of symmetric holes numerically.' --- 1.5cm Halmagyi$^*$, Michela Petrini$^*$, Alberto Zaffaroni$^{\dagger}$\ $^{*}$ Laboratoire de Physique Pierre et Marie Curie, CNRS UMR 7589,\ | been possible thanks to generOus support From tHe ITalIaN SpaCe AgEncy via contracT i/038/080/0. DatA presented in this paper wEre alSo ANalyZEd Using the HersCHeL INteRaCtIve PrOCeSsing EnvIronmenT (HIPE), a joinT deVeLopment by the hErSchel ScienCe GRound Segment conSortiuM, cOnsISting Of EsA, the nASA HeRSchel SCience CenTeR, And the hiFI, PACS, AND SpIRE Consortia. This mateRIaL Is based upon worK suppoRtED bY THe NAtiOnal SciencE FOundaTIon undeR awARD no. 1602583 aND by NASA througH an award issUEd bY JPL/CaLtEch VIa NASA grant \#1350780.
[^1]: foR An O6 Star, the main SequEnce lifetIme is aBOut 2.4 $\timeS$ 10$^{6}$ Years [@maE87], so 15% is 3.6 $\tImeS$ 10$^{5}$ yeArs. NOTe ThAt a LeSS maSSiVe mID-B sTar would HaVe A lifeTime ABOUT 5$\timEs$ lOngeR, chanGing our absoluTe lIfetIMe eStimaTes by That FaCtor. THis varIatioN gIves a sense of the UnceRtainties InvOlVed In DerivINg absoLutE liFetimes.
[^2]: @Rob08, the RED MSx CATALoG from @urq08, and the EGO cAtALOg From @cyg08.
---
aBstracT: 'we StUDy supersYmMetRic bLACk holEs in $aDS_4$ In the fraMework OF fOuR dimensIoNal gauGeD $\N=2$ sUpeRgravITy coUpled tO hypermuLtiplETs. We derive the fLOw equations foR A gENErAL eleCtrIcally gaugeD theORy whEre tHE gAugE Group Is AbeLiAN aND, restricting them to tHe Fixed pOints, We derive the gaUged supergRAVIty analoGue oF ThE Attractor equatIons fOr theories COupled to HyperMultipleTs. The partICUlar modeLs wE anAlyZe aRE CoNsistent truncATIons Of m-theory On cErtain SAsaKi-EInsTeiN sEven-manifOlds. We stUdY tHe SpAce Of horIZon solutIoNs oF tHe fOrm $Ads_2\Times \SIg_g$ wiTh boTh ElECtrIc and maGNeTIC chaRgEs And fInd A fOur-diMensIOnaL solutiOn space whEn tHE theOrY aRises frOm a reduction oN $Q^{111}$. for other $SE_7$ ReDucTions, tHE SolutionS space is a subspace of this. wE constrUct ExpliCit eXamples of SphEricalLy sYMmetriC black Holes NuMerICAlly.'
---
1.5cM nIcK HaLmAgyi$^*$, MichelA pEtrIni$^*$, AlBeRto ZAffaronI$^{\dagger}$\
0.5cm $^{*}$ LaboratoiRE de physique ThéorIquE et HAUTeS EnERgIEs,\
UNiVErsITÉ Pierre et Marie CUrie, CNRS UMr 7589,\
| been possible thanks to g enerous su pport fr omth e It alia n Space Agency viacontract I/038/080/0.Datapr e sent e din th is pape r w e r e a ls oana ly z ed usin g T he Hers chel inter act iv e processing en vironment(HI PE), a joint de velopm en t b y theHer schel Scien c e Grou nd Segmen tC onsort i um, con s i st ingof ESA, the NASAH er s chel Science C enter, a n dt h e H IFI , PACS, an dSPIRE consort i a. T h ism aterial is ba sed upon wo r k s upport ed by the Na tiona lS cie nce Foundat ionunder Awa rd No. 1602583 and byNASA t hro ugh ana wa rd is su e d b y J PL/ C alt ech viaNA SA Gran t \# 1 3 5 0 780.
[ ^1]: Foran O6 star, t hemain seq uence life time i s abo ut 2.4 $\ti me s$ 10$^{6}$ yea rs [ @mae87],so15 % i s3.6 $ \ times$ 10 $^{ 5}$ yea rs. Not e th at a le ss massive mid-B s ta r wo uld have a lif e ti me about 5$ \t ime s$ l o n ger,chan g in g our ab solute li fe time es ti matesby th atfacto r . Th is var iation g ivesa sense of theu ncertaintiesi nv o l ve d inder iving absol utel ifet imes .
[^2 ] : @ro b08,th e R e d MSX Catalog from@u rq08,and t he EGO catalo g from @cy g 0 8 .
---
a bstr a ct : 'We study sup ersym metric bla c k holesin $A dS_4$ in the fram e w ork of f our di men sio n a lgauged $\N=2$ s uper gr avity c oup led tohyp erm ult ipl et s. We der ive thefl ow e qu ati ons f o r a gene ra l e le ctr icall y gauge d the orywh er e th e gauge gr o u p is A be lian an d, rest rict i ngthem to the fixe d p o ints ,we derive the gauged s up ergravityan alo gue of t he attra ctor equations for theo r ies cou ple d tohype rmultiple ts. The p art i cularmodels we a na lyz e are c o n si ste nt truncatio n s of M-th eo ry o n certa in Sasaki-Einstein sev en-manifolds. We stu d y t hes pa c e o fh ori z o n solutions ofthe form $ Ad S _2 \times \Si g _g$ w ith bot h elect ric a n d magne tic charg es and fi nd a f o u r-d imensional solutio n space w h en th e t heory ar ises f ro m a redu ctiono n $ Q^{11 1}$. F or other $SE_ 7$ reducti ons, the solutions spac e is a subs pac e of this . W e co nstruct e xpli cit exampl esofspher ica l ly sy mmet r ic bl a ck ho lesn umericall y .'
-- -
1 .5cm Nick H a l m agy i$^*$ , M i chelaPetr ini$^*$, AlbertoZ affaroni$^{\da gger } $ \
0 .5c m $^{ *} $ LaboratoiredePh y s ique Thé or ique et Hau tes Ener gi e s,\
U nivers ité Pi erre et M ar i e Curi e, C NRS UMR 7589 ,\ | been_possible thanks_to generous support from_the Italian_Space_Agency via_contract_I/038/080/0. Data presented_in this paper_were also analyzed using_The Herschel interactive_processing_environment (HIPE), a joint development by the Herschel Science Ground Segment Consortium, consisting of_ESA,_the NASA_Herschel_Science_Center, and the HIFI, PACS,_and SPIRE consortia. This material_is based_upon work supported by the National Science Foundation_under_Award No. 1602583_and by NASA through an award issued by JPL/Caltech_via NASA Grant \#1350780.
[^1]: For an_O6 star, the_main_sequence_lifetime is about 2.4_$\times$ 10$^{6}$ years [@mae87], so 15%_is 3.6 $\times$ 10$^{5}$ years. Note_that a less massive mid-B star would_have a lifetime about 5$\times$ longer,_changing our absolute lifetime estimates_by that_factor. This variation gives a_sense of the_uncertainties involved_in deriving absolute_lifetimes.
[^2]: @rob08, the Red MSX Catalog_from @urq08, and_the EGO catalog from @cyg08.
---
abstract:_'We_study supersymmetric black_holes_in_$AdS_4$ in_the framework of_four_dimensional gauged_$\N=2$_supergravity coupled to hypermultiplets. We derive_the_flow equations for a general electrically gauged_theory where the gauge_group_is Abelian and, restricting_them to the fixed points,_we derive the gauged supergravity analogue_of the_attractor equations_for theories coupled to hypermultiplets. The particular models we analyze are_consistent truncations of M-theory on certain_Sasaki-Einstein seven-manifolds. We study_the space_of_horizon solutions of_the_form $AdS_2\times_\Sig_g$ with both electric and magnetic charges_and find_a four-dimensional solution space when the_theory arises from a_reduction_on $Q^{111}$. For other $SE_7$ reductions,_the solutions space is a subspace_of this. We construct explicit_examples_of_spherically symmetric black holes numerically.'
---
1.5cm_Nick Halmagyi$^*$, Michela Petrini$^*$, Alberto Zaffaroni$^{\dagger}$\
0.5cm_$^{*}$ Laboratoire de_Physique Théorique et Hautes Energies,\
Université Pierre et_Marie_Curie, CNRS UMR 7589,\
|
quark masses, at first for the pseudoscalar case, and investigate the resulting set of problems and solutions. We attempt to find patterns in – as well as to quantify corrections to – the rainbow-ladder truncation. In addition, we open this approach to phenomenological predictions of the heavy quark symmetry.'
author:
- 'María Gómez-Rocha'
- Thomas Hilger
- Andreas Krassnigg
date: 'Received: '
title: 'First look at heavy-light mesons with a dressed quark-gluon vertex [^1] '
---
Motivation {#sec:motivation}
==========
In modern studies of Quantum Chromodynamics (QCD) meson states made of one light and one heavy quark have attracted particular attention for a number of years. The reasons are manifold, e.g., the possibility to construct a theoretical bridge between effective field theories valid in opposite regimes like chiral perturbation theory as well as perturbative nonrelativistic QCD and heavy-quark effective theory, respectively. Another challenging reason is the need to understand a system determined by different scales of the order of $\Lambda_{\rm{QCD}}$ on one hand and the bottom-quark mass on the other.
Among the most promising approaches that are able to deal with all of these requirements as well as able to provide a solid and modern theoretical framework for bound-state studies are lattice-regularized QCD studies and also methods of continuum quantum field theories. Our method of choice herein is the Dyson-Schwinger-Bethe-Salpeter-equation (DSBSE) approach, which has been applied to various problems of theoretical hadron physics in the past decades with increasing success and comprehension, see, e.g., [@Bashir:2012fs; @Cloet:2013jya] and references therein.
Typically, sophisticated hadron studies in the DSBSE approach are numerical in nature and make use of a truncation of the infinite tower of coupled integral equations involved; see, e.g., [@Blank:2010bp; @Krassnigg:2009zh; @Fischer:2014xha] for the details of such a setup. A popular variant is the so-called rainbow-ladder (RL) truncation, where the dressed quark-gluon vertex [@Alkofer:2008tt] assumes a strongly simplified structure; it models the gluon-quark interaction and reduces the effort of solving relevant equations to the Dyson | quark masses, at first for the pseudoscalar case, and investigate the resulting hardening of trouble and solutions. We attempt to find practice in – as well as to quantify corrections to – the rainbow - run truncation. In addition, we open this access to phenomenological predictions of the heavy quark cheese symmetry.'
author:
-' María Gómez - Rocha'
- Thomas Hilger
- Andreas Krassnigg
date:' Received:'
title:' foremost look at heavy - inner light mesons with a dressed quark cheese - gluon vertex [ ^1 ]'
---
Motivation { # sec: motivation }
= = = = = = = = = =
In modern studies of Quantum Chromodynamics (QCD) meson states make of one light and one heavy quark cheese have attracted finical attention for a issue of years. The reasons are manifold, e.g., the hypothesis to construct a theoretical bridge between effective field theories valid in opposite regimes like chiral perturbation hypothesis equally well as perturbative nonrelativistic QCD and big - quark effective theory, respectively. Another challenging reason is the need to understand a system determined by different plate of the order of $ \Lambda_{\rm{QCD}}$ on one hand and the bottom - quark bulk on the other.
Among the most promising approaches that are able to share with all of these requirements as well as able to provide a solid and mod theoretical framework for bound - state studies are lattice - regularized QCD report and also methods of continuum quantum field theories. Our method of choice herein is the Dyson - Schwinger - Bethe - Salpeter - equality (DSBSE) overture, which has been applied to various problems of theoretical hadron physics in the past decade with increasing success and comprehension, examine, e.g., [ @Bashir:2012fs; @Cloet:2013jya ] and references therein.
Typically, sophisticated hadron studies in the DSBSE approach path are numerical in nature and make function of a truncation of the countless tower of coupled integral equations imply; see, e.g., [ @Blank:2010bp; @Krassnigg:2009zh; @Fischer:2014xha ] for the detail of such a setup. A democratic variant is the so - called rainbow - run (RL) truncation, where the dressed quark - gluon vertex [ @Alkofer:2008tt ] assumes a strongly simplified social organization; it models the gluon - quark interaction and reduces the effort of solve relevant equations to the Dyson | quagk masses, at first for tme pseudoscalar eqse, anv invesfigate tfe resulting set of problems abd sooutions. We attempt to wind pattvrns in – qs wtll as to quantifb correcbnons fl – tke rainbow-ladder truncatiot. In addition, fe o'en this approach to phenomenologicaj predivtlons of the hewvy atark symmetry.'
author:
- 'María Gómez-Rocha'
- Thojas Hiljer
- Andreas Kraxsnigg
date: 'Received: '
title: 'Virsh look at heavy-ligjt mesons wuth w dressed quafk-gluon vegcex [^1] '
---
Motivajion {#sec:motivation}
==========
In modern studkes oy Quantum Cyrimofinamics (QCD) neson states made of one night amd one heavy qmark iave attracted particular attention for a nomber of yaaxs. The reasons are mabidold, g.g., tha porwibkliuy vo donstrkct a theoretjcal bridge between effective giqod theories vzlid ig jpposite regimes like chiral perturbatimn fheory as well as pertuebative nonrelativistlc QCD anq heavy-quark effective theory, respectively. Anothes chamuenylkg rdqsln is the need to understand a system determigsd bj different scalef of the orceg ps $\Lambda_{\rm{QCD}}$ on onz hznd and the bottom-euark mwss ob the othtr.
Amomg the most promising approqches that age avle to deal with apl of these reqoirememts as well as able to 'rovids a solid ajd modern gheoretical framdwogk fmr bound-state studies are jattice-rejularnzed QCD stucies agd also mehhods of continuum quantkm figld thaories. Our method of choice herein is the Dyson-Schwinget-Bedhe-Valpeter-zquatipn (DSBSE) apprjach, which has been a'plied to varioua problxms of theorqtical hadron khysics in thx past desadew wirh incrdxsing success snd comprvhtnsion, see, e.g., [@Bashir:2012fs; @Cloet:2013mya] avs references thzxeun.
Typically, sopnisgicwtvd iadrog studies in dhe ASBRR appfoach are nmmeficak in nature and make use of a truncation og bhe infinute towew of coupled ontegral equations invonvev; see, r.g., [@Flank:2010bp; @Krassnigg:2009zh; @Fischer:2014xha] for the fetwils of such a sctup. A popular variant is the so-called rainbow-ladder (CL) truncation, where thg dressed quark-gluon revtex [@Alkofer:2008vt] asstmes a stsongly simplified steucture; it models the gluon-quark interzction and geduces the effort of solving relevant equations to the Dyson | quark masses, at first for the pseudoscalar investigate resulting set problems and solutions. in as well as quantify corrections to the rainbow-ladder truncation. In addition, we this approach to phenomenological predictions of the heavy quark symmetry.' author: - 'María - Thomas Hilger - Andreas Krassnigg date: 'Received: ' title: 'First look at mesons a quark-gluon [^1] ' --- Motivation {#sec:motivation} ========== In modern studies of Quantum Chromodynamics (QCD) meson states made one light and one heavy quark have attracted attention for a number years. The reasons are manifold, the to construct theoretical between field theories valid opposite regimes like chiral perturbation theory as well as perturbative nonrelativistic QCD and heavy-quark effective theory, respectively. challenging reason need to a determined different scales of of $\Lambda_{\rm{QCD}}$ on one hand and on the other. Among the most promising approaches are able deal with all of these requirements well as able to provide a solid and theoretical framework for bound-state studies are lattice-regularized QCD studies and also methods of continuum quantum Our method of choice is the Dyson-Schwinger-Bethe-Salpeter-equation approach, has applied various problems theoretical hadron physics in the past decades with increasing success and see, e.g., [@Bashir:2012fs; @Cloet:2013jya] and references therein. Typically, sophisticated hadron the approach are numerical nature and make use a of the infinite tower integral involved; @Krassnigg:2009zh; for details of such a A popular variant is the rainbow-ladder (RL) truncation, where assumes a strongly simplified structure; it models the interaction and reduces the effort of solving equations to the Dyson | quark masses, at first for the pSeudoscalaR case, And InvEsTigaTe thE resulting set oF ProbLems and solutions. We atteMpt to FiND patTErNs in – aS well as TO qUANtiFy CoRreCtIOnS to – thE raInbow-laDder truncaTioN. IN addition, we oPEn This approaCh tO phenomenoloGicAl predIcTioNS of thE heAvy quArk symMEtry.'
auThor:
- 'María góMEz-RochA'
- thomas HILGeR
- AndReas Krassnigg
date: 'rEcEIved: '
title: 'First Look at HeAVy-LIGht MesOns with a drEsSed quARk-gluon VErTEX [^1] '
---
motIVation {#sec:motiVation}
==========
In modERn sTudies Of quaNTum ChrOmodyNaMIcs (qCD) meson staTes mAde of one lIght anD One heavY Quark haVe attrActEd pArtiCUlAr AttEnTIon FOr A nuMBer Of years. THe ReAsons Are mANIFOld, e.G., thE posSibilIty to construcT a tHeorETicAl briDge beTweeN eFfectIve fieLd theOrIes valid in opposIte rEgimes likE chIrAl pErTurbaTIon theOry As wEll as peRturbatIVe nOnRELAtIvistic QCD and heavy-QuARK eFfective Theory, REsPeCTively. AnOtHer ChalLENging ReasON iS the need To undeRStAnD a systeM dEtermiNeD by DifFerenT ScalEs of thE order of $\lambdA_{\Rm{QCD}}$ on one hand ANd the bottom-quARk MASs ON the OthEr.
Among the mOst pROmisIng aPPrOacHEs thaT are aBlE To DEal with all of these reQuIremenTs as wEll as able to prOvide a soliD AND modern tHeorETiCAl framework for Bound-State studiES are lattIce-reGularizeD QCD studiES And also mEthOds Of cOntINUuM quantum field THEoriEs. our methOd oF choice HerEin Is tHe DYsOn-SchwingEr-Bethe-SAlPeTeR-eQuaTion (DsbSE) approAcH, whIcH haS been APplied To varIous PrObLEms Of theorETiCAL hadRoN pHysiCs iN tHe pasT decADes With incReasing suCceSS and CoMpRehensiOn, see, e.g., [@Bashir:2012Fs; @cloet:2013jya] anD rEfeRences THErein.
TypIcally, sophisticated hadrON studieS in The DSbSE aPproach arE nuMericaL in NAture aNd make Use of A tRunCATion oF THe InfInIte tower of COUplEd intEgRal eQuationS involved; see, e.g., [@BlanK:2010Bp; @KRassnigg:2009zh; @FisCheR:2014xha] FOR tHe dETaILs oF sUCh a SETup. A popular variAnt is the so-CaLLeD rainbow-laDDer (rL) TruncatIon, wherE the dREssed quArk-gluon vErtex [@AlkoFeR:2008tt] aSSUmeS a strongly SimplifiEd structuRE; it moDElS the gLuoN-quark InTerActioN and reDUceS the eFfort oF sOlving RelevAnT equatioNs to the Dyson | quark masses, at first for the pseud oscal arcas e, and inv estigate the r e sult ing set of problems an d sol ut i ons. We atte mpt tof in d pat te rn s i n– a s wel l a s to qu antify cor rec ti ons to – the ra inbow-ladd ertruncation.Inadditi on , w e open th is ap proach to phe nomenolog ic a l pred i ctionso f t he h eavy quark symmet r y. '
author:
- 'Ma ría Gó me z -R o c ha'
-Thomas Hil ge r
- A n dreas K r as s n i ggd ate: 'Receive d: '
title: 'Fi rst lo ok at heavy- light m e son s with a dr esse d quark-g luon v e rtex [^ 1 ] '
---
Moti vat ion {#s e c: mo tiv at i on} == === = === =
In mo de rn stud ieso f Q uant umChro modyn amics (QCD) m eso n st a tes made of o ne l ig ht an d oneheavy q uark have attra cted particul arat ten ti on fo r a num ber of years. The re a son sa r e m anifold, e.g., the p o s si bility t o cons t ru ct a theore ti cal bri d g e bet ween ef fectivefieldt he or ies val id in op po sit e r egime s lik e chir al pertu rbati o n theory as we l l as perturba t iv e no n rela tiv istic QCD a nd h e avy- quar k e ffe c tivetheor y, re s pectively. Anotherch alleng ing r eason is theneed to un d e r stand asyst e md etermined by d iffer ent scales of the o rderof $\Lam bda_{\rm{ Q C D}}$ onone ha ndand t he bottom-quark m asson the ot her .
Amon g t hemos t p ro mising ap proaches t ha tar e a ble t o deal wi th al lofthese requir ement s as w el l as able t o p r o vide a s olid an dmoder n th e ore tical f rameworkfor boun d- st ate stu dies are latt ic e-regulari ze d Q CD stu d i es and a lso methods of continuu m quantu m f ieldtheo ries. Our me thod o f c h oice h ereinis th eDys o n -Schw i n ge r-B et he-Salpete r - equ ation ( DSBS E) appr oach, which has be e n a pplied to var iou s pr o b le mso ft heo re t ica l hadron physicsin the pas td ec ades withi ncr ea sing su ccess a nd co m prehens ion, see, e.g., [@ Ba shir : 2 012 fs; @Cloet :2013jya ] and ref e rence s t herei n.
Typic al ly, soph istica t edhadro n stud ie s in t he DS BS E approa ch are numerical in nat ure an d mak e u se of a t run c ati on of the inf inite towe r o f c ouple d i n tegra l eq u at ion s invo lved ; see, e.g . ,[@B l a nk :2010bp; @K r a s sni gg:20 09z h ; @Fis cher :2014xha] for the details of suc h as e tup . A popu la r variant is t heso - c alled ra in bow-ladder(RL) tru nc a tion, where the d ressedq u ar k -gluon ver tex [@Alkofe r:2 00 8 tt] ass um es a stro ngly s implif ied st r uctu r e ; it models thegluon - q uarki nte racti on and re d uces the effor t of solvin g rele vant equa tions t othe Dy son | quark masses,_at first_for the pseudoscalar case,_and investigate_the_resulting set_of_problems and solutions._We attempt to_find patterns in –_as well as_to_quantify corrections to – the rainbow-ladder truncation. In addition, we open this approach to_phenomenological_predictions of_the_heavy_quark symmetry.'
author:
- 'María Gómez-Rocha'
- Thomas_Hilger
- Andreas Krassnigg
date: 'Received: '
title:_'First look_at heavy-light mesons with a dressed quark-gluon vertex_[^1]_'
---
Motivation {#sec:motivation}
==========
In modern_studies of Quantum Chromodynamics (QCD) meson states made of_one light and one heavy quark_have attracted particular_attention_for_a number of years._The reasons are manifold, e.g., the_possibility to construct a theoretical bridge_between effective field theories valid in opposite_regimes like chiral perturbation theory as_well as perturbative nonrelativistic QCD_and heavy-quark_effective theory, respectively. Another challenging_reason is the_need to_understand a system_determined by different scales of the_order of $\Lambda_{\rm{QCD}}$_on one hand and the bottom-quark_mass_on the other.
Among_the_most_promising approaches_that are able_to_deal with_all_of these requirements as well as_able_to provide a solid and modern theoretical_framework for bound-state studies_are_lattice-regularized QCD studies and_also methods of continuum quantum_field theories. Our method of choice_herein is_the Dyson-Schwinger-Bethe-Salpeter-equation_(DSBSE) approach, which has been applied to various problems of theoretical_hadron physics in the past decades_with increasing success and_comprehension, see,_e.g.,_[@Bashir:2012fs; @Cloet:2013jya] and_references_therein.
Typically, sophisticated_hadron studies in the DSBSE approach are_numerical in_nature and make use of a_truncation of the infinite_tower_of coupled integral equations involved; see,_e.g., [@Blank:2010bp; @Krassnigg:2009zh; @Fischer:2014xha] for the_details of such a setup._A_popular_variant is the so-called rainbow-ladder_(RL) truncation, where the dressed quark-gluon_vertex [@Alkofer:2008tt] assumes_a strongly simplified structure; it models the_gluon-quark_interaction and reduces the effort of_solving_relevant equations to the Dyson |
, with the magnetic term included with a coupling constant $\kappa$ relative to the electric term. If we expand in the manner of Eq. (\[Div\]), and integrate over the divergence, it only contributes a velocity-independent interaction term to the Hamiltonian. We can however retain its full form, and find that the conjugate momenta are now $$P^{\mu}(x) \equiv F^{\mu 0}(x) + \kappa \tilde{F}^{\mu 0},
\label{NewP}$$ i.e. the sum of the electric and magnetic fields. The new Hamiltonian then becomes $$H^{\prime} = \int d^3x \Big( \frac12 {\langle}P^i,P^i {\rangle}+ \frac14 (1 - \kappa^2) {\langle}F_{ij}, F_{ij} {\rangle}+ \frac12 \kappa \epsilon^{ijk} {\langle}P^i, F_{jk} {\rangle}- {\langle}A_0, \partial_i P^i + [P^i,A_i] {\rangle}\Big).$$ Inspection of this Hamiltonian reveals that the dynamics can be simplified by the choice of the coupling parameter, so henceforth we will put $\kappa = 1$, as we did initially in our [*ad hoc*]{} construction.
The constraint structure is similar to the electric case, and results in the final constraint $\Lambda^{\prime} = 3 \epsilon^{ijk} [P^i,A_j,A_k] + 6 [A_0,P^i,A_i]$, with $$\begin{aligned}
\frac16 \frac{d\Lambda^{\prime}}{dt} & = &
\epsilon^{ijk} \Big( \frac12 [[A_0,P^i],A_j,A_k] + [P^i,\partial_j A_0, A_k]
+ [P^i,[A_0,A_j],A_k] \nonumber \\
& & + [A_0,\partial_j P^i,A_k] + [A_0,[P^i,A_j],A_k] + \frac12 [A_0,P^i,F_{jk}]
\Big) \nonumber \\
& & + [\partial_j P^i,A_i,A_j] + [[P^i,A | , with the magnetic term included with a coupling changeless $ \kappa$ relative to the electric condition. If we expand in the manner of Eq. (\[Div\ ]), and integrate over the divergence, it entirely contributes a velocity - autonomous interaction term to the Hamiltonian. We can however retain its full form, and recover that the conjugate momenta are now $ $ P^{\mu}(x) \equiv F^{\mu 0}(x) + \kappa \tilde{F}^{\mu 0 },
\label{NewP}$$ i.e. the union of the electric and magnetic fields. The raw Hamiltonian then become $ $ H^{\prime } = \int d^3x \Big (\frac12 { \langle}P^i, P^i { \rangle}+ \frac14 (1 - \kappa^2) { \langle}F_{ij }, F_{ij } { \rangle}+ \frac12 \kappa \epsilon^{ijk } { \langle}P^i, F_{jk } { \rangle}- { \langle}A_0, \partial_i P^i + [ P^i, A_i ] { \rangle}\Big).$$ Inspection of this Hamiltonian reveals that the dynamics can be simplified by the choice of the coupling argument, so henceforth we will put $ \kappa = 1 $, as we did initially in our [ * ad hoc * ] { } construction.
The restraint structure is similar to the electric case, and results in the final restraint $ \Lambda^{\prime } = 3 \epsilon^{ijk } [ P^i, A_j, A_k ] + 6 [ A_0,P^i, A_i]$, with $ $ \begin{aligned }
\frac16 \frac{d\Lambda^{\prime}}{dt } & = &
\epsilon^{ijk } \Big (\frac12 [ [ A_0,P^i],A_j, A_k ] + [ P^i,\partial_j A_0, A_k ]
+ [ P^i,[A_0,A_j],A_k ] \nonumber \\
& & + [ A_0,\partial_j P^i, A_k ] + [ A_0,[P^i, A_j],A_k ] + \frac12 [ A_0,P^i, F_{jk } ]
\Big) \nonumber \\
& & + [ \partial_j P^i, A_i, A_j ] + [ [ P^i, A | , wihh the magnetic term incuuded with a coopoing cmnstanf $\kappa$ felative to the electric terl. Uf we expand in the manner uf Eq. (\[Div\]), wnd intetratt over the divergxhce, it only ckktribbtxs a velocity-incependent hnteraction tesm tl the Hamiltonian. We can however reeain itx vull form, and sind ehat nht conjugate momenta are now $$P^{\mu}(x) \squiv F^{\ku 0}(x) + \kappa \tolde{F}^{\mu 0},
\label{NewP}$$ i.e. the skm ov the electric and magnetic fueldf. The new Hamkltonian then becomes $$G^{\prime} = \int d^3x \Big( \frac12 {\langle}P^k,P^i {\rcngle}+ \frac14 (1 - \japod^2) {\langle}F_{ij}, F_{ij} {\gangle}+ \frac12 \kappa \epshlon^{ijk} {\langle}P^i, F_{jk} {\vanglx}- {\labgle}A_0, \partial_i P^i + [P^i,E_i] {\rangle}\Big).$$ Inspectyon of thhs Hamiltonian reveqlw thaj the dynxnicr czn bs simppifmed by the dhoice of tye coupling paramettr, fi henceforth se wilj [ut $\kappa = 1$, as we did initially in our [*dd goc*]{} construction.
The conwtraint structure is dimilar tj the electric case, and results in the final consdrainv $\Uamyea^{\priow} = 3 \epsilon^{ijk} [P^i,A_j,A_k] + 6 [A_0,P^i,A_i]$, with $$\begin{aligned}
\sdav16 \nrac{d\Lambda^{\prime}}{bt} & = &
\epsilon^{ijk} \Nih( \gtac12 [[A_0,P^i],A_j,A_k] + [P^k,\particm_j A_0, A_k]
+ [P^i,[A_0,A_j],A_k] \nonulber \\
& & + [A_0,\parrial_j P^i,A_h] + [A_0,[L^i,A_j],A_k] + \frac12 [A_0,P^i,F_{jk}]
\Big) \nonunber \\
& & + [\partpal_j P^i,A_i,A_j] + [[P^i,A | , with the magnetic term included with constant relative to electric term. If of (\[Div\]), and integrate the divergence, it contributes a velocity-independent interaction term to Hamiltonian. We can however retain its full form, and find that the conjugate are now $$P^{\mu}(x) \equiv F^{\mu 0}(x) + \kappa \tilde{F}^{\mu 0}, \label{NewP}$$ i.e. the of electric magnetic The new Hamiltonian then becomes $$H^{\prime} = \int d^3x \Big( \frac12 {\langle}P^i,P^i {\rangle}+ \frac14 (1 - {\langle}F_{ij}, F_{ij} {\rangle}+ \frac12 \kappa \epsilon^{ijk} {\langle}P^i, F_{jk} {\langle}A_0, \partial_i P^i + {\rangle}\Big).$$ Inspection of this Hamiltonian that dynamics can simplified the of the coupling so henceforth we will put $\kappa = 1$, as we did initially in our [*ad hoc*]{} construction. constraint structure to the case, results the final constraint 3 \epsilon^{ijk} [P^i,A_j,A_k] + 6 [A_0,P^i,A_i]$, \frac{d\Lambda^{\prime}}{dt} & = & \epsilon^{ijk} \Big( \frac12 [[A_0,P^i],A_j,A_k] [P^i,\partial_j A_0, + [P^i,[A_0,A_j],A_k] \nonumber \\ & & [A_0,\partial_j P^i,A_k] + [A_0,[P^i,A_j],A_k] + \frac12 [A_0,P^i,F_{jk}] \Big) \\ & & + [\partial_j P^i,A_i,A_j] + [[P^i,A | , with the magnetic term includEd with a couPling ConStaNt $\KappA$ relAtive to the elecTRic tErm. If we expand in the mannEr of EQ. (\[DIV\]), and INtEgratE over thE DiVERgeNcE, iT onLy COnTribuTes A velociTy-independEnt InTeraction terM To The HamiltoNiaN. We can howeveR reTain itS fUll FOrm, anD fiNd thaT the coNJugate Momenta arE nOW $$P^{\mu}(x) \eQUiv F^{\mu 0}(x) + \KAPpA \tilDe{F}^{\mu 0},
\label{NewP}$$ i.e. tHE sUM of the electric And magNeTIc FIEldS. ThE new HamiltOnIan thEN becomeS $$h^{\pRIME} = \inT D^3x \Big( \frac12 {\langLe}P^i,P^i {\ranglE}+ \FraC14 (1 - \kappa^2) {\LaNglE}f_{ij}, F_{ij} {\RanglE}+ \fRAc12 \kAppa \epsilon^{Ijk} {\lAngle}P^i, F_{jK} {\ranglE}- {\Langle}A_0, \PArtial_i p^i + [P^i,A_i] {\RanGle}\big).$$ INSpEcTioN oF ThiS haMilTOniAn revealS tHaT the dYnamICS CAn be SimPlifIed by The choice of thE coUpliNG paRametEr, so hEnceFoRth we Will puT $\kappA = 1$, aS we did initially In ouR [*ad hoc*]{} conStrUcTioN.
THe conSTraint StrUctUre is siMilar to THe eLeCTRIc Case, and results in thE fINAl ConstraiNt $\LambDA^{\pRiME} = 3 \epsilon^{IjK} [P^i,a_j,A_k] + 6 [a_0,p^I,A_i]$, wiTh $$\beGIn{Aligned}
\fRac16 \fraC{D\LAmBda^{\primE}}{dT} & = &
\epsilOn^{Ijk} \big( \Frac12 [[A_0,p^I],A_j,A_K] + [P^i,\parTial_j A_0, A_k]
+ [p^i,[A_0,A_j],a_K] \nonumber \\
& & + [A_0,\partIAl_j P^i,A_k] + [A_0,[P^i,A_j],A_K] + \FrAC12 [a_0,P^I,f_{jk}]
\BIg) \nOnumber \\
& & + [\partIal_j p^I,A_i,A_J] + [[P^i,A | , with the magnetic term i ncluded wi th acou pli ng con stan t $\kappa$ rel a tive to the electric term. If w ee xpan d i n the manner of E q.(\ [D iv\ ]) , a nd in teg rate ov er the div erg en ce, it onlyc on tributes a ve locity-indep end ent in te rac t ion t erm to t he Ham i ltonia n. We can h o weverr etain i t s f ullform, and find th a tt he conjugate m omenta a r en o w $ $P^ {\mu}(x) \ eq uiv F ^ {\mu 0} ( x) + \ka p pa \tilde{F}^ {\mu 0},
\l a bel {NewP} $$ i. e . thesum o ft heelectric an d ma gnetic fi elds.T he newH amilton ian th enbec omes $$ H^ {\p ri m e}= \ int d^3 x \Big(\f ra c12 { \lan g l e } P^i, P^i {\r angle }+ \frac14 (1 -\kap p a^2 ) {\l angle }F_{ ij }, F_ {ij} { \rang le }+ \frac12 \kap pa \ epsilon^{ ijk }{\l an gle}P ^ i, F_{ jk} {\ rangle} - {\lan g le} A_ 0 , \p artial_i P^i + [P^ i, A _ i] {\rangl e}\Big ) .$ $I nspectio nofthis H amilt onia n r eveals t hat th e d yn amics c an be si mp lif ied by t h e ch oice o f the co uplin g parameter, so henceforth we wi l l p u t $\ kap pa = 1$, as wed id i niti a ll y i n our[*adho c *] { } construction.
Th econstr aintstructure issimilar to t h e electr ic c a se , and results i n the final con s traint $ \Lamb da^{\pri me} = 3 \ e p silon^{i jk} [P ^i, A_j , A _k ] + 6 [A_0,P^ i , A_i] $, with $ $\b egin{al ign ed}
\f rac 16 \frac{d\ Lambda^{ \p ri me }} {dt } & = &
\epsil on ^{i jk } \ Big(\ frac12 [[A_ 0,P^ i] ,A _ j,A _k] + [ P ^i , \ part ia l_ j A_ 0,A_ k]
+[P^i , [A_ 0,A_j], A_k] \non umb e r \\
& & + [A_0 ,\partial_j P ^i ,A_k] + [A _0 ,[P ^i,A_j ] , A_k] + \ frac12 [A_0,P^i,F_{jk}] \Big) \ non umber \\& & + [\p art ial_jP^i , A_i,A_ j] + [ [P^i, A | , with_the magnetic_term included with a_coupling constant_$\kappa$_relative to_the_electric term. If_we expand in_the manner of Eq. (\[Div\]),_and integrate over_the_divergence, it only contributes a velocity-independent interaction term to the Hamiltonian. We can however_retain_its full_form,_and_find that the conjugate momenta_are now $$P^{\mu}(x) \equiv F^{\mu_0}(x) +_\kappa \tilde{F}^{\mu 0},
\label{NewP}$$ i.e. the sum of the_electric_and magnetic fields._The new Hamiltonian then becomes $$H^{\prime} = \int d^3x_\Big( \frac12 {\langle}P^i,P^i {\rangle}+ \frac14 (1_- \kappa^2) {\langle}F_{ij},_F_{ij}_{\rangle}+_\frac12 \kappa \epsilon^{ijk} {\langle}P^i,_F_{jk} {\rangle}- {\langle}A_0, \partial_i P^i +_[P^i,A_i] {\rangle}\Big).$$ Inspection of this Hamiltonian_reveals that the dynamics can be simplified_by the choice of the coupling_parameter, so henceforth we will_put $\kappa_= 1$, as we did_initially in our_[*ad hoc*]{}_construction.
The constraint structure_is similar to the electric case,_and results in_the final constraint $\Lambda^{\prime} = 3_\epsilon^{ijk}_[P^i,A_j,A_k] + 6_[A_0,P^i,A_i]$,_with_$$\begin{aligned}
\frac16 \frac{d\Lambda^{\prime}}{dt}_& = &
\epsilon^{ijk}_\Big(_\frac12 [[A_0,P^i],A_j,A_k]_+_[P^i,\partial_j A_0, A_k]
+ [P^i,[A_0,A_j],A_k] \nonumber \\
&_&_+ [A_0,\partial_j P^i,A_k] + [A_0,[P^i,A_j],A_k] + \frac12_[A_0,P^i,F_{jk}]
\Big) \nonumber \\
& &_+_[\partial_j P^i,A_i,A_j] + [[P^i,A |
solitons. As opposed to the bright soliton case explored in I, however, the real and imaginary parts of $p_j$ are not independent because of the constraints (3.2c). Actually, it may be parameterized either by the velocity of the $j$th soliton or by a single angular variable, as will see in section 4. An auxiliary variable $\tau$ introduced in (3.2a) will be used conveniently in performing the proof of (2.10). It can be set to zero after all the calculations have been completed.
[**Remark 3.1.**]{} The tau function $g$ given by (3.1b) is represented by the determinant of an $(N+1)\times (N+1)$ matrix. It can be rewritten by the determinant of an $N\times N$ matrix. To show this, we multiply the $(N+1)$th column of $g$ by $z_{k,t}^*/\rho^2$ and subtract it from the $k$th column for $k=1, 2,..., N$ to obtain $$g=\left|\left(\delta_{jk}-{\kappa+{\rm i}p_k^*\over p_j+p_k^*}\,{p_j\over p_k^*}\,z_jz_k^*\right)_{1\leq j,k\leq N}\right|. \eqno(3.3)$$ Although the tau function from (3.1b) is used in the proof of the dark $N$-soliton solution, an equivalent form (3.3) will be employed in section 4 to analyze the structure of the solution.
[**Remark 3.2.**]{} The complex parameters $p_j$ subjected to the constraints (3.2c) exist only if the condition $\kappa(1+\kappa\rho^2)>0$ is satisfied, as confirmed easily by putting $p_j=a_j+{\rm i}b_j$ with real $a_j$ and $b_j$. We will show in section 3.6 that this condition is closely related to the stability of the plane wave solution of the FL equation.
[*3.2. Notation and basic formulas for determinants*]{}
Before entering into the proof of the dark $N$-soliton solution, we first define the matrices associated with the dark $N$-soliton solution and then provide some basic formulas for determinants | solitons. As opposed to the bright soliton case explored in I, however, the veridical and fanciful parts of $ p_j$ are not independent because of the constraint (3.2c). Actually, it may be parameterized either by the speed of the $ j$th soliton or by a single angular variable, as will see in part 4. An auxiliary variable $ \tau$ precede in (3.2a) will be used conveniently in do the proof of (2.10). It can be set to zero after all the calculation have been complete.
[ * * Remark 3.1. * * ] { } The tau function $ g$ give by (3.1b) is constitute by the antigenic determinant of an $ (N+1)\times (N+1)$ matrix. It can be rewritten by the determinant of an $ N\times N$ matrix. To show this, we reproduce the $ (N+1)$th column of $ g$ by $ z_{k, t}^*/\rho^2 $ and subtract it from the $ k$th column for $ k=1, 2, ..., N$ to obtain $ $ g=\left|\left(\delta_{jk}-{\kappa+{\rm i}p_k^*\over p_j+p_k^*}\,{p_j\over p_k^*}\,z_jz_k^*\right)_{1\leq j, k\leq N}\right|. \eqno(3.3)$$ Although the tau function from (3.1b) is practice in the proof of the black $ N$-soliton solution, an equivalent form (3.3) will be employed in section 4 to analyze the structure of the solution.
[ * * Remark 3.2. * * ] { } The complex parameters $ p_j$ subjugate to the constraints (3.2c) exist only if the condition $ \kappa(1+\kappa\rho^2)>0 $ is satisfied, as confirmed easily by putting $ p_j = a_j+{\rm i}b_j$ with real $ a_j$ and $ b_j$. We will show in section 3.6 that this condition is closely related to the stability of the plane wave solution of the FL equation.
[ * 3.2. Notation and basic formula for determinants * ] { }
Before entering into the validation of the blue $ N$-soliton solution, we foremost define the matrices associated with the dark $ N$-soliton solution and then provide some basic formulas for determinants | sopitons. As opposed to the bright soliton case eeplored in I, hodever, the real and imaginary pqrts if $p_j$ are not independdnt becaude of thw coistraints (3.2c). Actuemly, it may be iaramztxrized either bi the velocidy of the $j$th vouicon or by a single angular variable, ws will sfe in section 4. An stxiljary variable $\tau$ introduced in (3.2a) sill be used convenirntly in performing the prlof lf (2.10). It can be set ho zero aftgd ajo the calculxtions havt yeen complejed.
[**Remark 3.1.**]{} The tau function $g$ gixen bv (3.1b) is reprgssnhgd by the deverminwnt of an $(N+1)\tlkes (N+1)$ katrix. Ot can be rewrltten by the determinant of ai $N\times N$ matrix. To show thiv, ce multiply the $(N+1)$th cilymn ox $g$ ty $z_{y,r}^*/\rhu^2$ ahd shbtrach iv from the $i$th column dor $k=1, 2,..., N$ to obtain $$b=\lqdt|\left(\delta_{jk}-{\iappa+{\ri y}p_k^*\over p_j+p_k^*}\,{p_j\over p_k^*}\,z_jz_k^*\right)_{1\leq j,k\leq N}\rjght|. \eqno(3.3)$$ Although the rau function from (3.1b) id used in the proof of the dark $N$-soliton solution, an equivdlent worn (3.3) dulp be employed in section 4 to analyze the strusfute of the solutijn.
[**Remark 3.2.**]{} Tne cpiplex parametgrs $p_j$ subnected to the conshraints (3.2c) exust only yf tne condition $\kappa(1+\kappa\rho^2)>0$ us satisfied, qs confirmed easilv by putting $p_j=a_k+{\rm i}n_j$ with real $a_j$ and $b_j$. Ce wilm show in sfction 3.6 tgxt this conditiov ix wlosely related to the stafility of the plane dave solutyon of the FL equation.
[*3.2. Notation anf basnc fosmulas for determinants*]{}
Before entering invp the proof pf thv dark $N$-sjlitok solution, we fyrst define thg matricef assuciated winh the dack $N$-soliton folution and djen provide vome basyc firmuoas for aeterminants | solitons. As opposed to the bright soliton in however, the and imaginary parts because the constraints (3.2c). it may be either by the velocity of the soliton or by a single angular variable, as will see in section 4. auxiliary variable $\tau$ introduced in (3.2a) will be used conveniently in performing the of It be to zero after all the calculations have been completed. [**Remark 3.1.**]{} The tau function $g$ given (3.1b) is represented by the determinant of an (N+1)$ matrix. It can rewritten by the determinant of $N\times matrix. To this, multiply $(N+1)$th column of by $z_{k,t}^*/\rho^2$ and subtract it from the $k$th column for $k=1, 2,..., N$ to obtain $$g=\left|\left(\delta_{jk}-{\kappa+{\rm i}p_k^*\over p_k^*}\,z_jz_k^*\right)_{1\leq j,k\leq Although the function (3.1b) used in the the dark $N$-soliton solution, an equivalent be employed in section 4 to analyze the of the [**Remark 3.2.**]{} The complex parameters $p_j$ to the constraints (3.2c) exist only if the $\kappa(1+\kappa\rho^2)>0$ is satisfied, as confirmed easily by putting $p_j=a_j+{\rm i}b_j$ with real $a_j$ and $b_j$. show in section 3.6 this condition is related the of plane wave of the FL equation. [*3.2. Notation and basic formulas for determinants*]{} entering into the proof of the dark $N$-soliton solution, we the associated with the $N$-soliton solution and then some formulas for determinants | solitons. As opposed to the briGht soliton Case eXplOreD iN I, hoWeveR, the real and imaGInarY parts of $p_j$ are not indepeNdent BeCAuse OF tHe conStraintS (3.2C). ACTUalLy, It May Be PArAmeteRizEd eitheR by the veloCitY oF the $j$th solitON oR by a single AngUlar variable, As wIll see In SecTIon 4. An AuxIliarY variaBLe $\tau$ iNtroduced In (3.2A) Will be USed convENIeNtly In performing the prOOf OF (2.10). It can be set to zEro aftEr ALl THE caLcuLations havE bEen coMPleted.
[**REMaRK 3.1.**]{} tHe tAU function $g$ givEn by (3.1b) is reprESenTed by tHe DetERminanT of an $(n+1)\tIMes (n+1)$ matrix. It caN be rEwritten bY the deTErminanT Of an $N\tiMes N$ maTriX. To Show THiS, wE muLtIPly THe $(n+1)$th COluMn of $g$ by $z_{K,t}^*/\RhO^2$ and sUbtrACT IT froM thE $k$th ColumN for $k=1, 2,..., N$ to obtaiN $$g=\lEft|\lEFt(\dElta_{jK}-{\kappA+{\rm i}P_k^*\Over p_J+p_k^*}\,{p_j\oVer p_k^*}\,Z_jZ_k^*\right)_{1\leq j,k\leq n}\rigHt|. \eqno(3.3)$$ AltHouGh The TaU funcTIon froM (3.1b) iS usEd in the Proof of THe dArK $n$-SOlIton solution, an equiVaLENt Form (3.3) will Be emplOYeD iN Section 4 tO aNalYze tHE StrucTure OF tHe solutiOn.
[**RemaRK 3.2.**]{} THe Complex PaRameteRs $P_j$ sUbjEcted TO the ConstrAints (3.2c) exIst onLY if the conditioN $\Kappa(1+\kappa\rho^2)>0$ IS sATIsFIed, aS coNfirmed easiLy by PUttiNg $p_j=A_J+{\rM i}b_J$ With rEal $a_j$ AnD $B_j$. wE will show in section 3.6 tHaT this cOnditIon is closely rElated to thE STAbility oF the PLaNE wave solution oF the Fl equation.
[*3.2. NOTation anD basiC formulaS for deterMINants*]{}
BefOre EntEriNg iNTO tHe proof of the dARK $N$-soLiTon soluTioN, we firsT deFinE thE maTrIces assocIated witH tHe DaRk $n$-soLiton SOlution aNd TheN pRovIde soME basic FormuLas fOr DeTErmInants | solitons. As opposed to t he brightsolit oncas eexpl ored in I, however , the real and imaginary pa rts o f$ p_j$ ar e not indepe n de n t be ca us e o ft he cons tra ints (3 .2c). Actu all y, it may be p a ra meterizedeit her by the v elo city o fthe $j$th so liton or by a sing le angula rv ariabl e , as wi l l s ee i n section 4. An a u xi l iary variable$\tau$ i n tr o d uce d i n (3.2a) w il l beu sed con v en i e n tly in performing the proofo f ( 2.10). I t c a n be s et to z e roafter all t he c alculatio ns hav e been c o mpleted .
[** Rem ark 3.1 . ** ]{ } T he tau fu nct i on$g$ give nby (3.1 b) i s r e pres ent ed b y the determinantofan $ ( N+1 )\tim es (N +1)$ m atrix . It c an be r ewritten by the det erminantofan $N \t imesN $ matr ix. To show t his, we mul ti p l y t he $(N+1)$th colum no f $ g$ by $z _{k,t} ^ */ \r h o^2$ and s ubt ract i t fro m th e $ k$th col umn fo r $ k= 1, 2,.. ., N$ to o bta in$$g=\ l eft| \left( \delta_{ jk}-{ \ kappa+{\rm i}p _ k^*\over p_j+ p _k ^ * }\ , {p_j \ov er p_k^*}\, z_jz _ k^*\ righ t )_ {1\ l eq j, k\leq N } \r i ght|. \eqno(3.3)$$Al though thetau functionfrom (3.1b ) i s used i n th e p r oof of the dar k $N$ -soliton s o lution,an eq uivalent form (3. 3 ) will be em plo yed in s ec tion 4 to ana l y ze t he struct ure of the so lut ion .
[* *Remark 3 .2.**]{} T he c om ple x par a meters $ p_ j$su bje ctedt o theconst rain ts ( 3 .2c ) exist on l y ifth econd iti on $\ka ppa( 1 +\k appa\rh o^2)>0$ i s s a tisf ie d, as con firmed easily b y putting$p _j= a_j+{\ r m i}b_j$with real $a_j$ and $b_ j $. We w ill show insection 3 .6that t his condit ion is clos el y r e l atedt o t hest ability of t heplane w avesolutio n of the FL equati o n.
[*3.2. Notat ion and b as icf or m ula sf ord e terminants*]{}
Before en te r in g into the pro of of the dark $ N$-so l iton so lution, w e first d ef inet h e m atrices as sociated with the dark$ N$ -soli ton solut io n a nd th en pro v ide some basic f ormula s for d etermina nts | solitons._As opposed_to the bright soliton_case explored_in_I, however,_the_real and imaginary_parts of $p_j$_are not independent because_of the constraints_(3.2c)._Actually, it may be parameterized either by the velocity of the $j$th soliton or_by_a single_angular_variable,_as will see in section_4. An auxiliary variable $\tau$_introduced in_(3.2a) will be used conveniently in performing the_proof_of (2.10). It_can be set to zero after all the calculations_have been completed.
[**Remark 3.1.**]{} The tau_function $g$ given_by_(3.1b)_is represented by the_determinant of an $(N+1)\times (N+1)$ matrix._It can be rewritten by the_determinant of an $N\times N$ matrix. To_show this, we multiply the $(N+1)$th_column of $g$ by $z_{k,t}^*/\rho^2$_and subtract_it from the $k$th column_for $k=1, 2,...,_N$ to_obtain $$g=\left|\left(\delta_{jk}-{\kappa+{\rm i}p_k^*\over_p_j+p_k^*}\,{p_j\over p_k^*}\,z_jz_k^*\right)_{1\leq j,k\leq N}\right|. \eqno(3.3)$$ Although_the tau function_from (3.1b) is used in the_proof_of the dark_$N$-soliton_solution,_an equivalent_form (3.3) will_be_employed in_section_4 to analyze the structure of_the_solution.
[**Remark 3.2.**]{} The complex parameters $p_j$ subjected_to the constraints (3.2c)_exist_only if the condition_$\kappa(1+\kappa\rho^2)>0$ is satisfied, as confirmed_easily by putting $p_j=a_j+{\rm i}b_j$ with_real $a_j$_and $b_j$._We will show in section 3.6 that this condition is closely_related to the stability of the_plane wave solution of_the FL_equation.
[*3.2._Notation and basic_formulas_for determinants*]{}
Before_entering into the proof of the dark_$N$-soliton solution,_we first define the matrices associated_with the dark $N$-soliton_solution_and then provide some basic formulas_for determinants |
\end{gathered}$$ and projectors $\Pi^{\theta_\perp}_{k}$, $\Pi^{\tilde{\theta}}_{p}$, $\Pi^{\tilde{\theta}_\perp}_{p}$ are defined analogously to $\Pi^{\theta}_{k}$. It should be noted that observables (\[seq:observables\]) commute $$[S^{\theta}_{k},S^{\tilde{\theta}}_{p}] = 0,
\label{eq:observables_commutator}$$ therefore the correlation function in the scalar state (\[eq:scalar\_state\_general\]) has the form $$\mathcal{C}^{\chi(k,p)}_{\textsf{polarization}}(k,\theta;p,\tilde{\theta}) =
\frac{{\langle\chi(k,p)|} S^{\tilde{\theta}}_{p} S^{\theta}_{k}
{|\chi(k,p)\rangle}}{{\langle\chi(k,p)|\chi(k,p)\rangle}}.
\label{eq:correl_function_polarization_general}$$ The numerator of the right side of Eq. (\[eq:correl\_function\_polarization\_general\]) can be written as \[see Eqs. (\[seq:observables\])\] $$\begin{aligned}
{\langle\chi(k,p)|} S^{\tilde{\theta}}_{p} S^{\theta}_{k}
{|\chi(k,p)\rangle} & = {\langle\chi(k,p)|}
\Pi^{\tilde{\theta}}_{p} \Pi^{\theta}_{k} {|\chi(k,p)\rangle} \nonumber \\
& + {\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p}
\Pi^{\theta_\perp}_{k} {|\chi(k,p)\rangle} \nonumber \\
& - {\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p}
\Pi^{\theta}_{k} {|\chi(k,p)\rangle} \nonumber \\
& - {\langle\chi(k,p)|} \Pi^{\tilde{\theta}}_{p}
\Pi^{\theta_\perp}_{k} {|\chi(k,p)\rangle}.
\label{eq:numerator_correl_func}
\end{aligned}$$ It is enough to calculate explicitly only ${\langle\chi(k,p)|}
| \end{gathered}$$ and projectors $ \Pi^{\theta_\perp}_{k}$, $ \Pi^{\tilde{\theta}}_{p}$, $ \Pi^{\tilde{\theta}_\perp}_{p}$ are defined analogously to $ \Pi^{\theta}_{k}$. It should be noted that observables (\[seq: observables\ ]) commute $ $ [ S^{\theta}_{k},S^{\tilde{\theta}}_{p } ] = 0,
\label{eq: observables_commutator}$$ therefore the correlation coefficient affair in the scalar state (\[eq: scalar\_state\_general\ ]) has the human body $ $ \mathcal{C}^{\chi(k, p)}_{\textsf{polarization}}(k,\theta;p,\tilde{\theta }) =
\frac{{\langle\chi(k, p)| } S^{\tilde{\theta}}_{p } S^{\theta}_{k }
{ |\chi(k, p)\rangle}}{{\langle\chi(k, p)|\chi(k, p)\rangle } }.
\label{eq: correl_function_polarization_general}$$ The numerator of the correct side of Eq. (\[eq: correl\_function\_polarization\_general\ ]) can be written as \[see Eqs. (\[seq: observables\])\ ] $ $ \begin{aligned }
{ \langle\chi(k, p)| } S^{\tilde{\theta}}_{p } S^{\theta}_{k }
{ |\chi(k, p)\rangle } & = { \langle\chi(k, p)| }
\Pi^{\tilde{\theta}}_{p } \Pi^{\theta}_{k } { |\chi(k, p)\rangle } \nonumber \\
& + { \langle\chi(k, p)| } \Pi^{\tilde{\theta}_\perp}_{p }
\Pi^{\theta_\perp}_{k } { |\chi(k, p)\rangle } \nonumber \\
& - { \langle\chi(k, p)| } \Pi^{\tilde{\theta}_\perp}_{p }
\Pi^{\theta}_{k } { |\chi(k, p)\rangle } \nonumber \\
& - { \langle\chi(k, p)| } \Pi^{\tilde{\theta}}_{p }
\Pi^{\theta_\perp}_{k } { |\chi(k, p)\rangle }.
\label{eq: numerator_correl_func }
\end{aligned}$$ It is adequate to calculate explicitly only $ { \langle\chi(k, p)| } | \enf{gathered}$$ and projectors $\Pi^{\theta_\perp}_{k}$, $\Pn^{\rilde{\tieta}}_{p}$, $\Pj^{\tilde{\thdta}_\perp}_{p}$ are defined analogonsly to $\Pu^{\theta}_{k}$. It should be nuted that observavles (\[weq:observaumes\]) commute $$[S^{\fmeta}_{k},V^{\vilde{\theta}}_{p}] = 0,
\lsbel{eq:obsesvables_commutadof}$$ cherefore the correlation function ig the svapar state (\[eq:scwlar\_xeate\_fvntral\]) has the form $$\mathcal{C}^{\chi(k,p)}_{\testsf{polerization}}(k,\theta;l,\tilde{\theta}) =
\frac{{\langle\chl(k,p)|} D^{\tilde{\theta}}_{p} S^{\thetw}_{k}
{|\chi(k,p)\rantle}}{{\lwbgle\chi(k,p)|\chi(y,p)\rangle}}.
\label{eq:correl_runction_polarization_general}$$ The numexator of thg dihvt side of Xq. (\[eq:cogrel\_function\_polarizathon\_genetal\]) can be wribten es \[swe Eqs. (\[seq:observables\])\] $$\uegin{aligned}
{\langle\cri(k,p)|} S^{\tilge{\cheta}}_{p} S^{\theta}_{k}
{|\chi(k,p)\rqntle} & = {\lancle\cfu(k,p)|}
\Pi^{\uilve{\tgeta}}_{p} \Oi^{\tieta}_{k} {|\chi(k,p)\dangle} \nonunber \\
& + {\langle\chi(k,k)|} \Py^{\nolde{\theta}_\perl}_{p}
\[i^{\eheta_\perp}_{k} {|\chi(k,p)\rangle} \nonumber \\
& - {\lancle\dhi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p}
\Pi^{\theta}_{k} {|\chi(k,p)\ranhle} \nonumfer \\
& - {\langle\chi(k,p)|} \Pi^{\tilde{\theta}}_{p}
\Pi^{\theta_\perp}_{k} {|\chi(k,')\rxngoe}.
\uqbfl{eq:numerator_correl_func}
\end{aligned}$$ It is enoudg uo balculate explicibly only ${\langle\chi(l,p)|}
| \end{gathered}$$ and projectors $\Pi^{\theta_\perp}_{k}$, $\Pi^{\tilde{\theta}}_{p}$, $\Pi^{\tilde{\theta}_\perp}_{p}$ are to It should noted that observables \label{eq:observables_commutator}$$ the correlation function the scalar state has the form $$\mathcal{C}^{\chi(k,p)}_{\textsf{polarization}}(k,\theta;p,\tilde{\theta}) = \frac{{\langle\chi(k,p)|} S^{\theta}_{k} {|\chi(k,p)\rangle}}{{\langle\chi(k,p)|\chi(k,p)\rangle}}. \label{eq:correl_function_polarization_general}$$ The numerator of the right side of Eq. (\[eq:correl\_function\_polarization\_general\]) can written as \[see Eqs. (\[seq:observables\])\] $$\begin{aligned} {\langle\chi(k,p)|} S^{\tilde{\theta}}_{p} S^{\theta}_{k} {|\chi(k,p)\rangle} & = {\langle\chi(k,p)|} \Pi^{\theta}_{k} \nonumber & {\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p} \Pi^{\theta_\perp}_{k} {|\chi(k,p)\rangle} \nonumber \\ & - {\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p} \Pi^{\theta}_{k} {|\chi(k,p)\rangle} \nonumber \\ & - \Pi^{\tilde{\theta}}_{p} \Pi^{\theta_\perp}_{k} {|\chi(k,p)\rangle}. \label{eq:numerator_correl_func} \end{aligned}$$ It is enough calculate explicitly only ${\langle\chi(k,p)|} | \end{gathered}$$ and projectors $\PI^{\theta_\perp}_{K}$, $\Pi^{\tiLde{\TheTa}}_{P}$, $\Pi^{\tIlde{\Theta}_\perp}_{p}$ are dEFineD analogously to $\Pi^{\theta}_{k}$. it shoUlD Be noTEd That oBservabLEs (\[SEQ:obSeRvAblEs\]) COmMute $$[S^{\TheTa}_{k},S^{\tilDe{\theta}}_{p}] = 0,
\laBel{Eq:Observables_cOMmUtator}$$ therEfoRe the correlaTioN functIoN in THe scaLar State (\[Eq:scalAR\_state\_General\]) haS tHE form $$\mAThcal{C}^{\cHI(K,p)}_{\TextSf{polarization}}(k,\thETa;P,\Tilde{\theta}) =
\frac{{\Langle\ChI(K,p)|} s^{\TIldE{\thEta}}_{p} S^{\theta}_{K}
{|\cHi(k,p)\rANgle}}{{\lanGLe\CHI(K,p)|\cHI(k,p)\rangle}}.
\labeL{eq:correl_fuNCtiOn_polaRiZatIOn_geneRal}$$ ThE nUMerAtor of the riGht sIde of Eq. (\[eq:Correl\_FUnction\_POlarizaTion\_geNerAl\]) cAn be WRiTtEn aS \[sEE EqS. (\[SeQ:obSErvAbles\])\] $$\begIn{AlIgned}
{\LangLE\CHI(k,p)|} S^{\TilDe{\thEta}}_{p} S^{\Theta}_{k}
{|\chi(k,p)\raNglE} & = {\lanGLe\cHi(k,p)|}
\PI^{\tildE{\theTa}}_{P} \Pi^{\thEta}_{k} {|\chI(k,p)\raNgLe} \nonumber \\
& + {\langlE\chi(K,p)|} \Pi^{\tilde{\TheTa}_\PerP}_{p}
\pi^{\theTA_\perp}_{k} {|\Chi(K,p)\rAngle} \noNumber \\
& - {\lANglE\cHI(K,P)|} \PI^{\tilde{\theta}_\perp}_{p}
\Pi^{\ThETA}_{k} {|\Chi(k,p)\ranGle} \nonUMbEr \\
& - {\LAngle\chi(K,p)|} \pi^{\tIlde{\THEta}}_{p}
\PI^{\theTA_\pErp}_{k} {|\chi(k,P)\ranglE}.
\LaBeL{eq:numeRaTor_corReL_fuNc}
\eNd{aliGNed}$$ IT is enoUgh to calCulatE Explicitly only ${\LAngle\chi(k,p)|}
| \end{gathered}$$ and proj ectors $\P i^{\t het a_\ pe rp}_ {k}$ , $\Pi^{\tilde { \the ta}}_{p}$, $\Pi^{\tild e{\th et a }_\p e rp }_{p} $ are d e fi n e d a na lo gou sl y t o $\P i^{ \theta} _{k}$. Itsho ul d be noted t h at observabl es(\[seq:obser vab les\]) c omm u te $$ [S^ {\the ta}_{k } ,S^{\t ilde{\the ta } }_{p}] = 0,
\ l a be l{eq :observables_comm u ta t or}$$ therefor e theco r re l a tio n f unction in t he sc a lar sta t e( \ [ eq: s calar\_state\ _general\]) has the f or m $ $ \mathc al{C} ^{ \ chi (k,p)}_{\te xtsf {polariza tion}} ( k,\thet a ;p,\til de{\th eta })=
\ fr ac {{\ la n gle \ ch i(k , p)| } S^{\ti ld e{ \thet a}}_ { p } S^{\ the ta}_ {k}
{|\chi(k,p)\r ang le}} { {\l angle \chi( k,p) |\ chi(k ,p)\ra ngle} }.
\label{eq:cor rel_ function_ pol ar iza ti on_ge n eral}$ $ T henumerat or of t h e r ig h t si de of Eq. (\[eq:co rr e l \_ function \_pola r iz at i on\_gene ra l\] ) ca n be wr itte n a s \[seeEqs. ( \ [s eq :observ ab les\]) \] $$ \be gin{a l igne d}
{\ langle\c hi(k, p )|} S^{\tilde{ \ theta}}_{p} S ^ {\ t h et a }_{k }
{|\chi(k,p) \ran g le}& ={ \l ang l e\chi (k,p) |} \ P i^{\tilde{\theta}}_ {p } \Pi^ {\the ta}_{k} {|\ch i(k,p)\ran g l e } \nonum ber\ \ & + {\langle\ chi(k ,p)|} \Pi^ { \tilde{\ theta }_\perp} _{p}
\ P i^{\thet a_\ per p}_ {k} { |\ chi(k,p)\rang l e } \n on umber \ \
& - {\ lan gle \ch i(k ,p )|} \Pi^{ \tilde{\ th et a} _\ per p}_{p }
\Pi ^{ \th et a}_ {k} { | \chi(k ,p)\r angl e} \ n onu mber \\ & - {\ la ng le\c hi( k, p)|}\Pi^ { \ti lde{\th eta}}_{p}
\Pi ^{ \t heta_\p erp}_{k} {|\c hi (k,p)\rang le }. \labe l { eq:numer ator_correl_func}
\end { aligned }$$ It i s en ough to c alc ulateexp l icitly only${\la ng le\ c h i(k,p ) | } | \end{gathered}$$_and projectors_$\Pi^{\theta_\perp}_{k}$, $\Pi^{\tilde{\theta}}_{p}$, $\Pi^{\tilde{\theta}_\perp}_{p}$ are_defined analogously_to_$\Pi^{\theta}_{k}$. It_should_be noted that_observables (\[seq:observables\]) commute_$$[S^{\theta}_{k},S^{\tilde{\theta}}_{p}] = 0,
\label{eq:observables_commutator}$$_therefore the correlation_function_in the scalar state (\[eq:scalar\_state\_general\]) has the form $$\mathcal{C}^{\chi(k,p)}_{\textsf{polarization}}(k,\theta;p,\tilde{\theta}) =
\frac{{\langle\chi(k,p)|} S^{\tilde{\theta}}_{p} S^{\theta}_{k}
_{|\chi(k,p)\rangle}}{{\langle\chi(k,p)|\chi(k,p)\rangle}}.
_\label{eq:correl_function_polarization_general}$$ The_numerator_of_the right side of Eq. (\[eq:correl\_function\_polarization\_general\])_can be written as \[see_Eqs. (\[seq:observables\])\] $$\begin{aligned}
_{\langle\chi(k,p)|} S^{\tilde{\theta}}_{p} S^{\theta}_{k}
{|\chi(k,p)\rangle} & = {\langle\chi(k,p)|}
\Pi^{\tilde{\theta}}_{p}_\Pi^{\theta}_{k}_{|\chi(k,p)\rangle} \nonumber \\
_ & + {\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p}
\Pi^{\theta_\perp}_{k}_{|\chi(k,p)\rangle} \nonumber \\
& -_{\langle\chi(k,p)|} \Pi^{\tilde{\theta}_\perp}_{p}
___\Pi^{\theta}_{k} {|\chi(k,p)\rangle} \nonumber \\
_ & - {\langle\chi(k,p)|} \Pi^{\tilde{\theta}}_{p}
_ \Pi^{\theta_\perp}_{k} {|\chi(k,p)\rangle}.
\label{eq:numerator_correl_func}
\end{aligned}$$_It is enough to calculate explicitly only_${\langle\chi(k,p)|}
|
efumi Mori. Quotients by groupoids., 145(1):193–213, 1997.
J[á]{}nos Koll[á]{}r. Projectivity of complete moduli., 32(1):235–268, 1990.
Serge Lang. Hyperbolic and [D]{}iophantine analysis., 14(2):159–205, 1986.
Michael McQuillan. Quelques compléments à une démonstration de [F]{}altings., 319(7):649–652, 1994.
D. Mumford, J. Fogarty, and F. Kirwan., volume 34 of [*Ergebnisse der Mathematik und ihrer Grenzgebiete (2) \[Results in Mathematics and Related Areas (2)\]*]{}. Springer-Verlag, Berlin, third edition, 1994.
David Mumford. An analytic construction of degenerating abelian varieties over complete rings., 24:239–272, 1972.
Iku Nakamura. Stability of degenerate abelian varieties. Preprint [MPI 98-78]{}, 1998.
Patricia L. Pacelli. Uniform boundedness for rational points., 88(1):77–102, 1997.
Patricia L. Pacelli. Uniform bounds for stably integral points on elliptic curves. Preprint, 1997.
Kenji Ueno. Classification of algebraic varieties. [I]{}., 27:277–342, 1973.
Paul Vojta. A higher-dimensional [M]{}ordell conjecture. In [*Arithmetic geometry (Storrs, Conn., 1984)*]{}, pages 341–353. Springer, New York, 1986.
[^1]: D.A. Partially supported by NSF grant DMS-9700520 and by an Alfred P. Sloan research fellowship
[^2]: K.M. partially supported by NSA grant MDA904-96-1-0008.
---
author:
- Biao Xiong
- Xun Li
- 'Shi-Lei Chao'
- Ling Zhou
title: 'Optomechanical quadrature squeezing in the non-Markovian regime'
---
Quantum physics exhibits many interesting non-calssical effects [@PhysRevA.82.045805; @6046093; @Berrada2013; @Eleuch2008; @Ele | efumi Mori. Quotients by groupoids. , 145(1):193–213, 1997.
J[á]{}nos Koll[á]{}r. Projectivity of complete moduli. , 32(1):235–268, 1990.
Serge Lang. Hyperbolic and [ D]{}iophantine analysis. , 14(2):159–205, 1986.
Michael McQuillan. Quelques compléments à une démonstration de [ F]{}altings. , 319(7):649–652, 1994.
D. Mumford, J. Fogarty, and F. Kirwan. , bulk 34 of [ * Ergebnisse der Mathematik und ihrer Grenzgebiete (2) \[Results in Mathematics and Related Areas (2)\ ] * ] { }. Springer - Verlag, Berlin, third version, 1994.
David Mumford. An analytic construction of degenerating abelian variety over arrant rings. , 24:239–272, 1972.
Iku Nakamura. Stability of debauched abelian varieties. Preprint [ MPI 98 - 78 ] { }, 1998.
Patricia L. Pacelli. Uniform boundedness for rational point. , 88(1):77–102, 1997.
Patricia L. Pacelli. Uniform bounds for stably integral compass point on elliptic curve. Preprint, 1997.
Kenji Ueno. Classification of algebraic varieties. [ I ] { }. , 27:277–342, 1973.
Paul Vojta. A higher - dimensional [ M]{}ordell conjecture. In [ * Arithmetic geometry (Storrs, Conn., 1984) * ] { }, pages 341–353. Springer, New York, 1986.
[ ^1 ]: D.A. Partially supported by NSF concession DMS-9700520 and by an Alfred P. Sloan research fellowship
[ ^2 ]: K.M. partially supported by NSA concession MDA904 - 96 - 1 - 0008.
---
author:
- Biao Xiong
- Xun Li
-' Shi - Lei Chao'
- Ling Zhou
title:' Optomechanical quadrature squeezing in the non - Markovian regime'
---
Quantum physics exhibit many interesting non - calssical effects [ @PhysRevA.82.045805; @6046093; @Berrada2013; @Eleuch2008; @Ele | efuli Mori. Quotients by grompoids., 145(1):193–213, 1997.
J[á]{}nos Koll[á]{}r. Projxctivitg of comolete moduli., 32(1):235–268, 1990.
Serge Lang. Hypxrbooic abd [D]{}iophantine analysir., 14(2):159–205, 1986.
Michaep McQuiloan. Wyelques compléments à une sélonscretion de [F]{}altinns., 319(7):649–652, 1994.
D. Mumforg, J. Fogarty, and F. Yixwan., volume 34 of [*Ergebnisse der Mathemwtik unc lhrer Grenzgebyete (2) \[Reshlts in Mathematics and Related Arsas (2)\]*]{}. Spginger-Verlag, Berlon, third edition, 1994.
David Mumvord. An analytic constguction of eegegwrating abelkan varietpzs over comklete rings., 24:239–272, 1972.
Iku Nakamura. Stabilkty oy degeneratg zbfnian varietmes. Prvprint [MPI 98-78]{}, 1998.
Patricia L. Pacello. Uniform bouncediess for rational points., 88(1):77–102, 1997.
Patricia L. Pacelli. Uniform tobnds for stably integeao poitts mn euoipgic cnrvss. Preoriit, 1997.
Kenji Ueho. Classifixation of algebraic vwgoeties. [I]{}., 27:277–342, 1973.
Pahl Vojea. A higher-dimensional [M]{}ordell conjecture. In [*Arithmetic geometry (Stirrs, Conn., 1984)*]{}, pages 341–353. Sprlnger, New York, 1986.
[^1]: D.A. Partially supported by NSF grant DMS-9700520 ang by en Alyvcd P. Wllan research fellowship
[^2]: K.M. partially supporteq bu KSA grant MDA904-96-1-0008.
---
aubhor:
- Biao Xiong
- Xum Pi
- 'Fhi-Lei Chao'
- Lkng Zhou
tjtle: 'Optomechanicap quadrwture squeezind in the non-Markovian regime'
---
Quabtum physics wxhibits many intexesting non-cclssicsl efgects [@PhysRevA.82.045805; @6046093; @Berrada2013; @Elehch2008; @Ele | efumi Mori. Quotients by groupoids., 145(1):193–213, 1997. Projectivity complete moduli., 1990. Serge Lang. 1986. McQuillan. Quelques compléments une démonstration de 319(7):649–652, 1994. D. Mumford, J. Fogarty, F. Kirwan., volume 34 of [*Ergebnisse der Mathematik und ihrer Grenzgebiete (2) \[Results Mathematics and Related Areas (2)\]*]{}. Springer-Verlag, Berlin, third edition, 1994. David Mumford. An construction degenerating varieties complete rings., 24:239–272, 1972. Iku Nakamura. Stability of degenerate abelian varieties. Preprint [MPI 98-78]{}, 1998. Patricia Pacelli. Uniform boundedness for rational points., 88(1):77–102, 1997. L. Pacelli. Uniform bounds stably integral points on elliptic Preprint, Kenji Ueno. of varieties. 27:277–342, 1973. Paul A higher-dimensional [M]{}ordell conjecture. In [*Arithmetic geometry (Storrs, Conn., 1984)*]{}, pages 341–353. Springer, New York, 1986. [^1]: Partially supported grant DMS-9700520 by Alfred Sloan research fellowship partially supported by NSA grant MDA904-96-1-0008. Biao Xiong - Xun Li - 'Shi-Lei Chao' Ling Zhou 'Optomechanical quadrature squeezing in the non-Markovian --- Quantum physics exhibits many interesting non-calssical effects @6046093; @Berrada2013; @Eleuch2008; @Ele | efumi Mori. Quotients by groupOids., 145(1):193–213, 1997.
J[á]{}nos KOll[á]{}r. proJecTiVity Of coMplete moduli., 32(1):235–268, 1990.
SeRGe LaNg. Hyperbolic and [D]{}iophanTine aNaLYsis., 14(2):159–205, 1986.
mIcHael MCQuillaN. quELQueS cOmPléMeNTs À une dÉmoNstratiOn de [F]{}altinGs., 319(7):649–652, 1994.
D. muMford, J. FogartY, AnD F. Kirwan., voLumE 34 of [*ErgebnissE deR MatheMaTik UNd ihrEr GRenzgEbiete (2) \[rEsults In MathemaTiCS and ReLAted AreAS (2)\]*]{}. spRingEr-Verlag, Berlin, thiRD eDItion, 1994.
David MumfOrd. An aNaLYtIC ConStrUction of deGeNeratINg abeliAN vARIEtiES over complete Rings., 24:239–272, 1972.
Iku NakAMurA. StabiLiTy oF DegeneRate aBeLIan Varieties. PrEpriNt [MPI 98-78]{}, 1998.
PatrIcia L. PACelli. UnIForm bouNdedneSs fOr rAtioNAl PoIntS., 88(1):77–102, 1997.
PATriCIa l. PaCEllI. Uniform BoUnDs for StabLY INTegrAl pOintS on elLiptic curves. PRepRint, 1997.
kEnjI Ueno. classIficAtIon of AlgebrAic vaRiEties. [I]{}., 27:277–342, 1973.
Paul Vojta. a higHer-dimensIonAl [m]{}orDeLl conJEcture. in [*ARitHmetic gEometry (sTorRs, cONN., 1984)*]{}, pAges 341–353. Springer, New YorK, 1986.
[^1]: D.a. pArTially suPporteD By nSf Grant DMS-9700520 AnD by An AlFREd P. SlOan rESeArch fellOwship
[^2]: k.m. pArTially sUpPorted By nSA GraNt MDA904-96-1-0008.
---
AUthoR:
- Biao XIong
- Xun LI
- 'Shi-LEI Chao'
- Ling Zhou
tITle: 'OptomechanICaL QUaDRatuRe sQueezing in tHe noN-markOviaN ReGimE'
---
quantUm phySiCS eXHibits many interestiNg Non-calSsicaL effects [@PhysREvA.82.045805; @6046093; @Berrada2013; @eLEUch2008; @Ele | efumi Mori. Quotients by g roupoids., 145( 1): 193 –2 13,1997 .
J[á]{}nos K o ll[á ]{}r. Projectivity ofcompl et e mod u li ., 32 (1):235 – 26 8 , 19 90 .
Se rg e L ang.Hyp erbolic and [D]{} iop ha ntine analys i s. , 14(2):15 9–2 05, 1986.
M ich ael Mc Qu ill a n. Qu elq ues c omplém e nts àune démon st r ationd e [F]{} a l ti ngs. , 319(7):649–652, 19 9 4.
D. Mumford , J. F og a rt y , an d F . Kirwan., v olume 34 of [ * Er g e b nis s e der Mathema tik und ihr e r G renzge bi ete (2) \[ Resul ts inMathematics and RelatedAreas( 2)\]*]{ } . Sprin ger-Ve rla g,Berl i n, t hir de dit i on , 1 9 94.
DavidMu mf ord.An a n a l y ticcon stru ction of degenerat ing abe l ian vari eties ove rcompl ete ri ngs., 2 4:239–272, 1972 .
I ku Nakamu ra. S tab il ity o f degen era teabelian variet i es. P r e p ri nt [MPI 98-78]{},19 9 8 .
Patrici a L. P a ce ll i . Unifor mbou nded n e ss fo r ra t io nal poin ts., 8 8 (1 ): 77–102, 1 997.
Pa tri cia L. P a cell i. Uni form bou nds f o r stably integ r al points one ll i p ti c cur ves . Preprint, 199 7 .
K enji Ue no. Class ifica ti o no f algebraic varieti es . [I]{ }., 2 7:277–342, 19 73.
PaulV o j ta. A hi gher - di m ensional [M]{} ordel l conjectu r e. In [* Arith metic ge ometry (S t o rrs, Con n., 19 84) *]{ } , p ages 341–353. S prin ge r, NewYor k, 1986 .
[^1 ]:D.A .Partially support ed b yNS F g rantD MS-97005 20 an dbyan Al f red P. Sloa n re se ar c h f ellowsh i p[ ^2]: K .M . pa rti al ly su ppor t edby NSAgrant MDA 904 - 96-1 -0 00 8.
---
author:
- Bi ao Xiong
- X un Li
- 'Sh i - Lei Chao '
- Ling Zhou
title: 'O p tomecha nic al qu adra ture sque ezi ng inthe non-Ma rkovia n reg im e'- - -
Qu a n tu m p hy sics exhib i t s m any i nt eres ting no n-calssical effect s [@ PhysRevA.82.0 458 05;@ 6 04 609 3 ;@ Ber ra d a20 1 3 ; @Eleuch2008;@Ele | efumi Mori._Quotients by_groupoids., 145(1):193–213, 1997.
J[á]{}nos Koll[á]{}r._Projectivity of_complete_moduli., 32(1):235–268,_1990.
Serge_Lang. Hyperbolic and_[D]{}iophantine analysis., 14(2):159–205,_1986.
Michael McQuillan. Quelques compléments_à une démonstration_de_[F]{}altings., 319(7):649–652, 1994.
D. Mumford, J. Fogarty, and F. Kirwan., volume 34 of [*Ergebnisse der Mathematik und ihrer Grenzgebiete_(2)_\[Results in_Mathematics_and_Related Areas (2)\]*]{}. Springer-Verlag, Berlin,_third edition, 1994.
David Mumford. An_analytic construction_of degenerating abelian varieties over complete rings., 24:239–272,_1972.
Iku_Nakamura. Stability of_degenerate abelian varieties. Preprint [MPI 98-78]{}, 1998.
Patricia L. Pacelli. Uniform_boundedness for rational points., 88(1):77–102, 1997.
Patricia L._Pacelli. Uniform bounds_for_stably_integral points on elliptic_curves. Preprint, 1997.
Kenji Ueno. Classification of_algebraic varieties. [I]{}., 27:277–342, 1973.
Paul Vojta._A higher-dimensional [M]{}ordell conjecture. In [*Arithmetic geometry_(Storrs, Conn., 1984)*]{}, pages 341–353. Springer,_New York, 1986.
[^1]: D.A. Partially_supported by_NSF grant DMS-9700520 and by_an Alfred P._Sloan research_fellowship
[^2]: K.M. partially_supported by NSA grant MDA904-96-1-0008.
---
author:
-_Biao Xiong
- Xun_Li
- 'Shi-Lei Chao'
- Ling Zhou
title: 'Optomechanical_quadrature_squeezing in the_non-Markovian_regime'
---
Quantum_physics exhibits_many interesting non-calssical_effects_[@PhysRevA.82.045805; @6046093;_@Berrada2013;_@Eleuch2008; @Ele |
a class of functions satisfying certain special entropy conditions that include the class of measurable sets. The behavior of various ratio-type empirical processes are also investigated. Applications of these sharp bounds to the models mentioned above are detailed in Section \[section:application\]. For clarity of presentation, proofs for some of the results in Section \[section:ep\_sharp\_bounds\] are deferred to the Appendix.
Notation {#section:notation}
--------
For a real-valued random variable $\xi$ and $1\leq p<\infty$, let ${\lVert\xi\rVert_{p}} := \big({\mathbb{E}}{\lvert\xi\rvert}^p\big)^{1/p} $ denote the ordinary $p$-norm.
For a real-valued measurable function $f$ defined on $(\mathcal{X},\mathcal{A},P)$, ${\lVertf\rVert_{L_p(P)}}\equiv {\lVertf\rVert_{P,p}}\equiv \big(P{\lvertf\rvert}^p)^{1/p}$ denotes the usual $L_p$-norm under $P$, and ${\lVertf\rVert_{\infty}}\equiv \sup_{x \in \mathcal{X}} {\lvertf(x)\rvert}$. $f$ is said to be $P$-centered if $Pf=0$. $L_p(g,B)$ denotes the $L_p(P)$-ball centered at $g$ with radius $B$. For simplicity we write $L_p(B)\equiv L_p(0,B)$. Throughout the article ${\varepsilon}_1,\ldots,{\varepsilon}_n$ will be i.i.d. Rademacher random variables independent of all other random variables. $C_{x}$ will denote a generic constant that depends only on $x$, whose numeric value may change from line to line unless otherwise specified. $a\lesssim_{x} b$ and $a\gtrsim_x b$ mean $a\leq C_x b$ and $a\geq C_x b$ respectively, and $a\asymp_x b$ means $a\lesssim_{x} b$ and $a\gtrsim_x b$ \[$a\lesssim b$ means $a\leq Cb$ for some absolute constant $C$\]. For two real numbers $a,b$, $a\vee b\equiv \max\{a,b\}$ and $a\wedge b\equiv\min | a class of functions satisfying certain limited information conditions that include the class of measurable set. The behavior of various proportion - type empirical processes are besides investigate. Applications of these astute bounds to the models note above are detail in Section \[section: application\ ]. For clarity of presentation, proofs for some of the results in part \[section: ep\_sharp\_bounds\ ] are deferred to the Appendix.
Notation { # section: notation }
--------
For a very - valued random variable $ \xi$ and $ 1\leq p<\infty$, permit $ { \lVert\xi\rVert_{p } }: = \big({\mathbb{E}}{\lvert\xi\rvert}^p\big)^{1 / p } $ denote the average $ p$-norm.
For a real - measure measurable function $ f$ defined on $ (\mathcal{X},\mathcal{A},P)$, $ { \lVertf\rVert_{L_p(P)}}\equiv { \lVertf\rVert_{P, p}}\equiv \big(P{\lvertf\rvert}^p)^{1 / p}$ denotes the common $ L_p$-norm under $ P$, and $ { \lVertf\rVert_{\infty}}\equiv \sup_{x \in \mathcal{X } } { \lvertf(x)\rvert}$. $ f$ is said to be $ P$-centered if $ Pf=0$. $ L_p(g, B)$ denotes the $ L_p(P)$-ball centered at $ g$ with radius $ B$. For simplicity we write $ L_p(B)\equiv L_p(0,B)$. Throughout the article $ { \varepsilon}_1,\ldots,{\varepsilon}_n$ will be i.i.d. Rademacher random variable star independent of all other random variables. $ C_{x}$ will denote a generic constant that depends only on $ x$, whose numeric value may change from line to line unless differently stipulate. $ a\lesssim_{x } b$ and $ a\gtrsim_x b$ mean $ a\leq C_x b$ and $ a\geq C_x b$ respectively, and $ a\asymp_x b$ mean $ a\lesssim_{x } b$ and $ a\gtrsim_x b$ \[$a\lesssim b$ means $ a\leq Cb$ for some absolute constant $ C$ \ ]. For two real numbers $ a, b$, $ a\vee b\equiv \max\{a, b\}$ and $ a\wedge b\equiv\min | a flass of functions satisnying certain spgcual envropy cknditionr that include the class of leqsuravle sets. The behavior uf varioud ratio-ttpe tmpirical processxa are also inbcstigctxd. Applications of these vharp bounds tm ghz models mentioned above are detaileq in Sevtlon \[section:appjicauiog\]. Fod clarity of presentation, proofs fkr some of the resulys in Section \[section:ep\_shagp\_boknds\] are deferred ho the Appebdix.
Gitation {#sectkon:notation}
--------
For a real-vzlued random variable $\xi$ and $1\led p<\inyty$, let ${\lVetc\zi\rGgrt_{p}} := \big({\matibb{E}}{\lvvrt\xi\rvert}^p\bin)^{1/i} $ denode the prdinary $p$-norm.
Nor a reql-valued measurable fnnction $f$ defined on $(\mathcal{X},\kachcal{A},P)$, ${\lVertf\rVert_{L_p(P)}}\ewuiv {\nVerdf\rVdet_{P,o}}\eqhit \bjg(P{\lvegtf\cvert}^p)^{1/p}$ denktes the usyal $L_p$-norm under $P$, snq ${\lVertf\rVert_{\ihfty}}\eqtid \sup_{x \in \mathcal{X}} {\lvertf(x)\rvert}$. $f$ is sapd tk be $P$-centered if $Pf=0$. $L_p(t,B)$ denotes the $L_p(P)$-balp centereq at $g$ with radius $B$. For simplicity we write $L_p(B)\exuiv M_o(0,B)$. Uhvjjthlut the article ${\varepsilon}_1,\ldots,{\varepsilon}_n$ wijm ne i.i.d. Rademachev random variables ijdrkendent of all other rahdom variables. $C_{x}$ aill degote q generic conxtant that depends only on $z$, whose numegic calue may change fxom line to uine unlexs otherwise specified. $c\lesssjm_{x} b$ and $a\htrsim_x b$ oean $a\leq C_x b$ avd $s\gaq C_x b$ respectively, and $a\wsymp_x b$ neanx $a\lessrim_{x} b$ and $a\gtrsim_x h$ \[$a\lesssim b$ means $a\leq Fb$ fot some absolute fonstant $C$\]. For two real numbers $a,b$, $a\vee b\equif \kax\{d,b\}$ and $a\cedge n\equiv\min | a class of functions satisfying certain special that the class measurable sets. The processes also investigated. Applications these sharp bounds the models mentioned above are detailed Section \[section:application\]. For clarity of presentation, proofs for some of the results in \[section:ep\_sharp\_bounds\] are deferred to the Appendix. Notation {#section:notation} -------- For a real-valued random $\xi$ $1\leq let := \big({\mathbb{E}}{\lvert\xi\rvert}^p\big)^{1/p} $ denote the ordinary $p$-norm. For a real-valued measurable function $f$ defined on $(\mathcal{X},\mathcal{A},P)$, {\lVertf\rVert_{P,p}}\equiv \big(P{\lvertf\rvert}^p)^{1/p}$ denotes the usual $L_p$-norm under $P$, ${\lVertf\rVert_{\infty}}\equiv \sup_{x \in \mathcal{X}} $f$ is said to be if $L_p(g,B)$ denotes $L_p(P)$-ball at with radius $B$. simplicity we write $L_p(B)\equiv L_p(0,B)$. Throughout the article ${\varepsilon}_1,\ldots,{\varepsilon}_n$ will be i.i.d. Rademacher random variables independent of other random will denote generic that only on $x$, value may change from line to specified. $a\lesssim_{x} b$ and $a\gtrsim_x b$ mean $a\leq b$ and C_x b$ respectively, and $a\asymp_x b$ $a\lesssim_{x} b$ and $a\gtrsim_x b$ \[$a\lesssim b$ means Cb$ for some absolute constant $C$\]. For two real numbers $a,b$, $a\vee b\equiv \max\{a,b\}$ and | a class of functions satisfyiNg certain sPeciaL enTroPy CondItioNs that include tHE claSs of measurable sets. The bEhaviOr OF varIOuS ratiO-type emPIrICAl pRoCeSseS aRE aLso inVesTigated. applicatioNs oF tHese sharp bouNDs To the modelS meNtioned above Are DetailEd In SECtion \[SecTion:aPplicaTIon\]. For Clarity of PrESentatIOn, proofS FOr Some Of the results in SecTIoN \[Section:ep\_sharp\_Bounds\] ArE DeFERreD to The AppendiX.
NOtatiON {#sectioN:NoTATIon}
--------
fOr a real-valued Random variaBLe $\xI$ and $1\leQ p<\InfTY$, let ${\lVErt\xi\RVERt_{p}} := \Big({\mathbb{E}}{\lVert\Xi\rvert}^p\bIg)^{1/p} $ denOTe the orDInary $p$-nOrm.
For A reAl-vAlueD MeAsUraBlE FunCTiOn $f$ DEfiNed on $(\matHcAl{x},\mathCal{A},p)$, ${\LvERtf\rverT_{L_p(P)}}\Equiv {\LVertf\rVert_{P,p}}\EquIv \biG(p{\lvErtf\rVert}^p)^{1/P}$ denOtEs the Usual $L_P$-norm UnDer $P$, and ${\lVertf\rVErt_{\iNfty}}\equiv \Sup_{X \iN \maThCal{X}} {\lVErtf(x)\rVerT}$. $f$ iS said to Be $P$-centERed If $pF=0$. $l_P(g,b)$ denotes the $L_p(P)$-ball CeNTErEd at $g$ witH radiuS $b$. FOr SImplicitY wE wrIte $L_P(b)\Equiv l_p(0,B)$. THRoUghout thE articLE ${\vArEpsilon}_1,\LdOts,{\varEpSilOn}_n$ Will bE I.i.d. RAdemacHer randoM variABles independenT Of all other ranDOm VARiABles. $c_{x}$ wIll denote a gEnerIC conStanT ThAt dEPends Only oN $x$, WHoSE numeric value may chaNgE from lIne to Line unless othErwise specIFIEd. $a\lesssIm_{x} b$ ANd $A\Gtrsim_x b$ mean $a\lEq C_x b$ And $a\geq C_x b$ REspectivEly, anD $a\asymp_x B$ means $a\leSSSim_{x} b$ and $A\gtRsiM_x b$ \[$A\leSSSiM b$ means $a\leq Cb$ FOR somE aBsolute ConStant $C$\]. FOr tWo rEal NumBeRs $a,b$, $a\vee b\Equiv \max\{A,b\}$ AnD $a\WeDge B\equiV\Min | a class of functions sati sfying cer tainspe cia lentr opyconditions tha t inc lude the class of meas urabl es ets. Th e beh avior o f v a r iou sra tio -t y pe empi ric al proc esses areals oinvestigated . A pplication s o f these shar p b oundsto th e mode lsmenti oned a b ove ar e detaile di n Sect i on \[se c t io n:ap plication\]. Forc la r ity of present ation, p r oo f s fo r s ome of the r esult s in Sec t io n \ [se c tion:ep\_shar p\_bounds\] are defer re d t o the A ppend ix .
N otation {#s ecti on:notati on}
-- - -----
F or a re al-val ued ra ndom va ri abl e$ \xi $ a nd$ 1\l eq p<\in ft y$ , let ${\ l V e r t\xi \rV ert_ {p}}:= \big({\mat hbb {E}} { \lv ert\x i\rve rt}^ p\ big)^ {1/p}$ den ot e the ordinary$p$- norm.
Fo r a r eal -v alued measur abl e f unction $f$ de f ine do n $( \mathcal{X},\mathc al { A }, P)$, ${\ lVertf \ rV er t _{L_p(P) }} \eq uiv{ \ lVert f\rV e rt _{P,p}}\ equiv\ bi g( P{\lver tf \rvert }^ p)^ {1/ p}$ d e note s theusual $L _p$-n o rm under $P$,a nd ${\lVertf\ r Ve r t _{ \ inft y}} \equiv \sup _{x\ in \ math c al {X} } {\lv ertf( x) \ rv e rt}$. $f$ is said t obe $P$ -cent ered if $Pf=0 $. $L_p(g, B ) $ denotes the $L _ p(P)$-ball cen tered at $g$ wi t h radius $B$. For sim plicity w e write $L _p( B)\ equ ivL _ p( 0,B)$. Throug h o ut t he articl e $ {\varep sil on} _1, \ld ot s,{\varep silon}_n $wi ll b e i .i.d. Rademach er ra nd omvaria b les in depen dent o fa llother r a nd o m var ia bl es.$C_ {x }$ wi ll d e not e a gen eric cons tan t tha tde pends o nly on $x$, w ho se numeric v alu e mayc h ange fro m line to line unless o t herwise sp ecifi ed.$a\lesssi m_{ x} b$and $a\gtr sim_xb$ me an $a \ l eq C_ x b$ an d$a\geq C_x b $ r espec ti vely , and $ a\asymp_x b$ means $a\ lesssim_{x} b $ a nd $ a \ gt rsi m _x b$\[ $ a\l e s ssim b$ means $ a\leq Cb$fo r s ome absolu t e c on stant $ C$\]. F or tw o real n umbers $a ,b$, $a\v ee b\e q u iv\max\{a,b\ }$ and $ a\wedge b \ equiv \ mi n | a_class of_functions satisfying certain special_entropy conditions_that_include the_class_of measurable sets._The behavior of_various ratio-type empirical processes_are also investigated._Applications_of these sharp bounds to the models mentioned above are detailed in Section \[section:application\]._For_clarity of_presentation,_proofs_for some of the results_in Section \[section:ep\_sharp\_bounds\] are deferred_to the_Appendix.
Notation {#section:notation}
--------
For a real-valued random variable $\xi$ and_$1\leq_p<\infty$, let ${\lVert\xi\rVert_{p}}_:= \big({\mathbb{E}}{\lvert\xi\rvert}^p\big)^{1/p} $ denote the ordinary $p$-norm.
For a real-valued_measurable function $f$ defined on $(\mathcal{X},\mathcal{A},P)$,_${\lVertf\rVert_{L_p(P)}}\equiv {\lVertf\rVert_{P,p}}\equiv \big(P{\lvertf\rvert}^p)^{1/p}$_denotes_the_usual $L_p$-norm under $P$,_and ${\lVertf\rVert_{\infty}}\equiv \sup_{x \in \mathcal{X}} {\lvertf(x)\rvert}$._$f$ is said to be $P$-centered_if $Pf=0$. $L_p(g,B)$ denotes the $L_p(P)$-ball centered_at $g$ with radius $B$. For_simplicity we write $L_p(B)\equiv L_p(0,B)$._Throughout the_article ${\varepsilon}_1,\ldots,{\varepsilon}_n$ will be i.i.d._Rademacher random variables_independent of_all other random_variables. $C_{x}$ will denote a generic_constant that depends_only on $x$, whose numeric value_may_change from line_to_line_unless otherwise_specified. $a\lesssim_{x} b$_and_$a\gtrsim_x b$_mean_$a\leq C_x b$ and $a\geq C_x_b$_respectively, and $a\asymp_x b$ means $a\lesssim_{x} b$_and $a\gtrsim_x b$ \[$a\lesssim_b$_means $a\leq Cb$ for_some absolute constant $C$\]. For_two real numbers $a,b$, $a\vee b\equiv_\max\{a,b\}$ and_$a\wedge b\equiv\min |
\infty$) give the corresponding values of the parameters of the zero dimensional energy balance model, $A$, $B$, $a$, $b$; and the effective inverse response times for the atmosphere and the surface. The first important thing to notice with model (\[model\]) is that it recovers the zero dimensional energy balance model. This can be seen by averaging Eq. (\[model\]) over the cells in the limit $N \to \infty$. The following reduced version of Eq. (\[model\]) is obtained: $$\begin{aligned}
\label{motion}
\dot{y}=-\lambda [y-(A+ax)]+\varepsilon(t), \\ \nonumber
\dot{x}=-\gamma [x-(B+by)].\end{aligned}$$
From the fact that Eq. (\[motion\]) represents an overdamped dynamics in a parabolic potential, the system converges to a stationary state with mean value given by the solution of the zero dimensional energy balance model. This result shows that model (\[model\]) is capable of representing with good approximation the mean behavior of energy in the coupled atmosphere – surface system. Now it will be argued that Eq. (\[model\]) can give realistic descriptions of temperature fluctuations as well. A large variability over the intrinsic characteristic times of the surface around the globe is expected to exist, as a consequence of the different response times present in the geosphere, hydrosphere, cryosphere and biosphere. At first instance, this situation is modeled by treating the $\gamma _n$’s like independent random variables taken from a uniform probability distribution. On the other hand, under the basis of the relative homogeneity of the atmosphere composition, it will be assumed by now that the response time to perturbations of the atmosphere is the same in all cells. In what follows, the value $\lambda_n = \lambda = 1$ is used. Under these assumptions, the time unit is defined as the mean atmospheric response time. The model represents the interaction between a hierarchy of time scales, ranging from minutes to days to geological times. From this point of view, it would be therefore reasonable to assume that $1/ \lambda$ lies in the intermediate scales, roughly in the range from weeks to months. This and other important aspects about the definition of the parameters in the model are intended to be refined by the author in the near future by close cooperation with climate experts. At this point the main goal is limited to explore the capabilities of the model (\[model\]), in order to give qualitatively realistic statistical descriptions of the temperature fluctuations | \infty$) give the corresponding values of the parameters of the zero dimensional department of energy libra model, $ A$, $ B$, $ a$, $ b$; and the effective inverse response time for the air and the surface. The first significant thing to notice with exemplar (\[model\ ]) is that it recovers the zero dimensional energy libra model. This can be seen by average Eq. (\[model\ ]) over the cell in the limit $ N \to \infty$. The following reduced version of Eq. (\[model\ ]) is obtained: $ $ \begin{aligned }
\label{motion }
\dot{y}=-\lambda [ y-(A+ax)]+\varepsilon(t), \\ \nonumber
\dot{x}=-\gamma [ x-(B+by)].\end{aligned}$$
From the fact that Eq. (\[motion\ ]) represent an overdamped dynamics in a parabolic potential, the system converge to a stationary state with mean value given by the solution of the zero dimensional energy proportion model. This result shows that model (\[model\ ]) is capable of map with good approximation the mean behavior of energy in the coupled atmosphere – surface system. Now it will be argued that Eq. (\[model\ ]) can give realistic descriptions of temperature fluctuations as well. A large unevenness over the intrinsic characteristic fourth dimension of the surface around the ball is expect to exist, as a consequence of the different response times present in the lithosphere, hydrosphere, cryosphere and biosphere. At first case, this situation is modeled by treat the $ \gamma _ n$ ’s like independent random variables taken from a uniform probability distribution. On the other hand, under the footing of the relative homogeneity of the atmosphere composition, it will be assumed by now that the response time to disruption of the atmosphere is the same in all cells. In what follows, the value $ \lambda_n = \lambda = 1 $ is used. Under these assumption, the time unit is defined as the base atmospheric response time. The model represents the interaction between a hierarchy of prison term scales, crop from minutes to days to geological fourth dimension. From this point of position, it would be therefore fair to assume that $ 1/ \lambda$ lies in the average scales, roughly in the range from weeks to months. This and early important aspects about the definition of the parameters in the mannequin are intended to be refined by the author in the near future by close cooperation with climate expert. At this point the main finish is limited to explore the capabilities of the mannequin (\[model\ ]), in order to give qualitatively realistic statistical description of the temperature fluctuations | \invty$) give the correspondikg values of the paramevers of the zeru dimensional energy balance midel, $Q$, $B$, $a$, $b$; and the effectkve inverde respobse uimes for the atmosphere and ths suryare. The first imkortant thinc to notice widh mldel (\[model\]) is that it recovers the sero dikejsional energy balsgce jodel. This can be seen by averaginf Eq. (\[movel\]) over the cekls in the limit $N \to \inftj$. Thf following reducef version od Eq. (\[nodel\]) is obtxined: $$\begin{aligned}
\labem{motion}
\dot{y}=-\lambda [y-(A+ax)]+\varepsilov(t), \\ \npnumber
\dot{z}=-\gqmmw [x-(B+by)].\end{alijned}$$
Frjm the fact bnat Eq. (\[motion\]) represents an ovxrdanped dynamics in a pacabolic potential, thg system cmnrerges to a stationart wtate witv mexb vxlut gmveh by tje aolution or the zero eimensional energy najqnce model. Thjs restle shows that model (\[model\]) is capable of geprssenting with good appriximation the mean bejavior of energy in the coupled atmosphere – surface system. Now mt wiol cw wrgued that Eq. (\[model\]) can give realistic descryltooks of temperaturc fluctuations as eepl. W large variacility ovsr the intrinsic cjaractetistic times of the surface around the globe iw expected tj exist, as a conseqbence of the difgereny response times presenc in tge geospherf, hydrospgdre, cryosphere avd niovphere. Au first instance, tris situavion ns modeldd bi treatyng the $\galma _n$’s like independent gandol eariables haken from a uniform probabilitb distributiom. Mn nhe other hand, under the bafis of the relctive hoiogendity of thv atmosphxre composityon, it will ba assumed by iow that ehe eespinse tiod to perturbatoons of tkt atmospheee is the same in ell zslls. In what fooloqs, the value $\lakbdx_n = \lwmuda = 1$ hs used. Undes thdse sssumotions, uke bimd unot is defined as the meah atmospheric resppnfe time. Jhe model represents tne interaction betaeen e hiererchy pf jime scales, ranging from minutea to days to geological eimew. From this 'oint of view, it would be therefore reasinable to assume jhat $1/ \lambda$ lies in jhe intermediave scajes, roughny in the range from weeks to months. Uhis and other importanf aspewts ahout the definition of the parameters in the model are intended to be refuned bb ehe author jn tne nedr yuture fy cpuse cooperation eith climate experts. At this pomnt the mahn goal is limited to explore yhd capabilitier of the model (\[model\]), in krder to give qualitatively realistic syatistical descriptions pf the twmperzture flucvuations | \infty$) give the corresponding values of the the dimensional energy model, $A$, $B$, inverse times for the and the surface. first important thing to notice with (\[model\]) is that it recovers the zero dimensional energy balance model. This can seen by averaging Eq. (\[model\]) over the cells in the limit $N \to The reduced of (\[model\]) is obtained: $$\begin{aligned} \label{motion} \dot{y}=-\lambda [y-(A+ax)]+\varepsilon(t), \\ \nonumber \dot{x}=-\gamma [x-(B+by)].\end{aligned}$$ From the fact that Eq. represents an overdamped dynamics in a parabolic potential, system converges to a state with mean value given the of the dimensional balance This result shows model (\[model\]) is capable of representing with good approximation the mean behavior of energy in the coupled – surface it will argued Eq. can give realistic temperature fluctuations as well. A large intrinsic characteristic times of the surface around the is expected exist, as a consequence of the response times present in the geosphere, hydrosphere, cryosphere biosphere. At first instance, this situation is modeled by treating the $\gamma _n$’s like independent taken from a uniform distribution. On the hand, the of relative homogeneity the atmosphere composition, it will be assumed by now that the time to perturbations of the atmosphere is the same in In follows, the value = \lambda = 1$ used. these assumptions, the time defined the time. model the interaction between a of time scales, ranging from to days to geological view, it would be therefore reasonable to assume $1/ \lambda$ lies in the intermediate scales, in the range from weeks to months. This and other important aspects the definition parameters in the model are intended to be by the author in near future by close cooperation with climate experts. At point main goal limited to explore capabilities of the (\[model\]), in order qualitatively realistic descriptions the | \infty$) give the corresponding Values of thE paraMetErs Of The zEro dImensional enerGY balAnce model, $A$, $B$, $a$, $b$; and the effEctivE iNVersE ReSponsE times fOR tHE AtmOsPhEre AnD ThE surfAce. the firsT important ThiNg To notice with MOdEl (\[model\]) is tHat It recovers thE zeRo dimeNsIonAL enerGy bAlancE model. tHis can Be seen by aVeRAging EQ. (\[Model\]) ovER ThE celLs in the limit $N \to \inFTy$. tHe following redUced veRsIOn OF eq. (\[mOdeL\]) is obtaineD: $$\bEgin{aLIgned}
\laBEl{MOTIon}
\DOt{y}=-\lambda [y-(A+ax)]+\Varepsilon(t), \\ \NOnuMber
\doT{x}=-\GamMA [x-(B+by)].\eNd{aliGnED}$$
FrOm the fact thAt Eq. (\[Motion\]) repResentS An overdAMped dynAmics iN a pAraBoliC PoTeNtiAl, THe sYStEm cONveRges to a sTaTiOnary StatE WITH meaN vaLue gIven bY the solution oF thE zerO DimEnsioNal enErgy BaLance Model. THis reSuLt shows that modeL (\[modEl\]) is capabLe oF rEprEsEntinG With goOd aPprOximatiOn the meAN beHaVIOR oF energy in the coupleD aTMOsPhere – surFace sySTeM. NOW it will bE aRguEd thAT eq. (\[modEl\]) caN GiVe realisTic desCRiPtIons of tEmPeratuRe FluCtuAtionS As weLl. A larGe variabIlity OVer the intrinsiC CharacteristiC TiMES oF The sUrfAce around thE gloBE is eXpecTEd To eXIst, as A consEqUEnCE of the different respOnSe timeS presEnt in the geospHere, hydrosPHERe, cryospHere ANd BIosphere. At firsT instAnce, this siTUation is ModelEd by treaTing the $\gaMMA _n$’s like iNdePenDenT raNDOm Variables takeN FRom a UnIform prObaBility dIstRibUtiOn. ON tHe other haNd, under tHe BaSiS oF thE relaTIve homogEnEitY oF thE atmoSPhere cOmposItioN, iT wILl bE assumeD By NOW thaT tHe RespOnsE tIme to PertURbaTions of The atmospHerE Is thE sAmE in all cElls. In what folLoWs, the value $\LaMbdA_n = \lambDA = 1$ Is used. UnDer these assumptions, the tIMe unit iS deFined As thE mean atmoSphEric reSpoNSe time. the modEl repReSenTS The inTERaCtiOn Between a hiERArcHy of tImE scaLes, rangIng from minutes to daYS to Geological timEs. FRom tHIS pOinT Of VIew, It WOulD BE therefore reasoNable to assUmE ThAt $1/ \lambda$ liES in ThE intermEdiate sCales, ROughly iN the range From weeks To MontHS. thiS and other iMportant Aspects abOUt the DEfInitiOn oF the paRaMetErs in The modEL arE inteNded to Be RefineD by thE aUthor in tHe near future by close coopEratioN with CliMate experTs. AT ThiS point the Main Goal is limiTed To eXplorE thE CapabIlitIEs Of tHE modeL (\[modEL\]), in order tO GiVe qUALiTatively reaLISTic StatiStiCAl descRiptIons of the temperatURe fluctuations | \infty$) give the corresp onding val ues o f t hepa rame ters of the zero d i mens ional energy balance m odel, $ A $, $ B $, $a$, $b$; a n dt h e e ff ec tiv ei nv erseres ponse t imes for t heat mosphere and th e surface. Th e first impo rta nt thi ng to notic e w ith m odel ( \ [model \]) is th at it rec o vers th e ze ro d imensional energy ba l ance model. Th is can b e s e e n b y a veraging E q. (\[m o del\])o ve r t hec ells in the l imit $N \to \in fty$.Th e f o llowin g red uc e d v ersion of E q. ( \[model\] ) is o b tained: $$\begi n{alig ned }
\ labe l {m ot ion }\ dot { y} =-\ l amb da [y-(A +a x) ]+\va reps i l o n (t), \\ \no numbe r
\dot{x}=-\g amm a [x - (B+ by)]. \end{ alig ne d}$$
Fromthe f ac t that Eq. (\[m otio n\]) repr ese nt s a noverd a mped d yna mic s in aparabol i c p ot e n t ia l, the system conv er g e sto a sta tionar y s ta t e with m ea n v alue g ivenby t h esolution of th e z er o dimen si onal e ne rgy ba lance mode l. Thi s result show s that model (\ [ model\]) is c a pa b l eo f re pre senting wit h go o d ap prox i ma tio n themeanbe h av i or of energy in the c oupled atmo sphere – surf ace system . N ow it wi ll b e a r gued that Eq.(\[mo del\]) can give rea listi c descri ptions of t emperatu reflu ctu ati o n sas well. A la r g e va ri ability ov er theint rin sic ch ar acteristi c timesof t he s urf ace a r ound the g lob eisexpec t ed toexist , as a c o nse quenceo ft h e di ff er entres po nse t imes pre sent in the geos phe r e, h yd ro sphere, cryosphere a nd biosphere .Atfirsti n stance,this situation is model e d by tr eat ing t he $ \gamma _n $’s likeind e penden t rand om va ri abl e s take n fr omauniform pr o b abi litydi stri bution. On the other hand , un der the basis of the r el ati v eh omo ge n eit y of the atmosphe re composi ti o n, it will b e as su med bynow tha t the respons e time to perturba ti onso f th e atmosphe re is th e same in all c e ll s. In wh at fol lo ws, thevalue$ \la mbda_ n = \l am bda =1$ is u sed. Und er these assumptions, t he tim e uni t i s defined as the mean atm osph eric respo nse ti me. T hem odelrepr e se nts the i nter a ction bet w ee n a h ie rarchy of t i m e sc ales, ra n ging f romminutes to days t o geological ti mes. F rom th i s po in t of view, itwou ld b e theref or e reasonabl e to ass um e that $1/ \ lambda $ liesi n t h e inte rmed iat e scales, ro ug h ly in t he r a nge fr om w ee ks tomonths . Thi s and other import ant a s p ectsa bou t the d efiniti o n of the param eters in th e mode l ar e int ended t obe ref ine dby the aut h or in the near future b y cl ose coope rati o n with cli ma teexperts.A t t h is p o int the main g oalis limite d to expl ore the cap ab ili t i es oft he m odel (\[mo del \]),i n order tog iveq ua l itati vely r ealist ic stat i sti ca l descr ipt i o ns of the temperat u refl uctu ations | \infty$)_give the_corresponding values of the_parameters of_the_zero dimensional_energy_balance model, $A$,_$B$, $a$, $b$;_and the effective inverse_response times for_the_atmosphere and the surface. The first important thing to notice with model (\[model\]) is_that_it recovers_the_zero_dimensional energy balance model. This_can be seen by averaging_Eq. (\[model\])_over the cells in the limit $N \to_\infty$._The following reduced_version of Eq. (\[model\]) is obtained: $$\begin{aligned}
\label{motion}
\dot{y}=-\lambda [y-(A+ax)]+\varepsilon(t), \\_\nonumber
\dot{x}=-\gamma [x-(B+by)].\end{aligned}$$
From the fact that Eq._(\[motion\]) represents an_overdamped_dynamics_in a parabolic potential,_the system converges to a stationary_state with mean value given by_the solution of the zero dimensional energy_balance model. This result shows that_model (\[model\]) is capable of_representing with_good approximation the mean behavior_of energy in_the coupled_atmosphere – surface_system. Now it will be argued_that Eq. (\[model\])_can give realistic descriptions of temperature_fluctuations_as well. A_large_variability_over the_intrinsic characteristic times_of_the surface_around_the globe is expected to exist,_as_a consequence of the different response times_present in the geosphere,_hydrosphere,_cryosphere and biosphere. At_first instance, this situation is_modeled by treating the $\gamma _n$’s_like independent_random variables_taken from a uniform probability distribution. On the other hand, under_the basis of the relative homogeneity_of the atmosphere composition,_it will_be_assumed by now_that_the response_time to perturbations of the atmosphere is_the same_in all cells. In what follows,_the value $\lambda_n =_\lambda_= 1$ is used. Under these_assumptions, the time unit is defined_as the mean atmospheric response_time._The_model represents the interaction between_a hierarchy of time scales, ranging_from minutes to_days to geological times. From this point_of_view, it would be therefore reasonable_to_assume that $1/ \lambda$ lies in_the_intermediate_scales, roughly in the range_from weeks to months. This and_other important aspects about the definition of the parameters_in the model_are intended to be refined_by_the_author in the near future by close cooperation with climate_experts. At_this point the_main goal is limited to explore the capabilities of the_model (\[model\]), in order to give qualitatively_realistic statistical descriptions of the temperature fluctuations |
$, for all $t \in (t'_0,\infty)$, and so, we have that $\omega \in \cC_{t_0}$. Therefore, it follows that $\{\tau_{\Delta}<t_0\} \subseteq \cC_{t_0}$, from which identity is follows.
\[step:measure\_cC3\] We may assume, without loss of generality, that $$\begin{aligned}
\label{eq:Nontrivial_set_t_0}
\PP^{0,z} (\tau_{\Delta}=t_0) \neq 0,\end{aligned}$$ otherwise we choose $\cN_{t_0}:=\{\tau_{\Delta}=t_0\}$ in, which is a $\cF^0$-measurable set by Remark \[rmk:tau\_Delta\_stopping\_time\]. From the Markov property of the coordinate process and identity, it is clear that $Z^0(t,\omega)=\Delta$, for all $t >t_0$ such that $t\in\DD_+$, and all $\omega\in\cC^3_{t_0}$, and so, the path $Z^0(\cdot,\omega)$ is continuous from the right at $t_0$. We only need to show that the path $Z^0(\cdot,\omega)$ is continuous from the left at $t_0$. We show that there is a set, $\cN_{t_0}\in\cF^0$, of $\PP^{0,z}$-measure zero, such that for all compact sets $K\subset\underline\Omega$, and all $\omega \in \cC^3_{t_0}\backslash \cN_{t_0}$, there is a positive constant, $\delta=\delta(K,\omega)$, such that $$\begin{aligned}
\label{eq:Convergence_process_both_top}
Z^0(t,\omega) \in \underline\Omega\backslash K,\quad\forall\, t\in\DD_+ \hbox{ such that } t_0-\delta<t<t_0,\end{aligned}$$ which is equivalent to the fact that $$\lim_{\stackrel{t\nearrow t_0}{t \in\DD_+}} Z^0(t,\omega) = \Delta,\quad\forall\, \omega \in \cC^3_{t_0}\ | $, for all $ t \in (t'_0,\infty)$, and so, we have that $ \omega \in \cC_{t_0}$. Therefore, it follow that $ \{\tau_{\Delta}<t_0\ } \subseteq \cC_{t_0}$, from which identity is take after.
\[step: measure\_cC3\ ] We may assume, without loss of generality, that $ $ \begin{aligned }
\label{eq: Nontrivial_set_t_0 }
\PP^{0,z } (\tau_{\Delta}=t_0) \neq 0,\end{aligned}$$ differently we choose $ \cN_{t_0}:=\{\tau_{\Delta}=t_0\}$ in, which is a $ \cF^0$-measurable set by Remark \[rmk: tau\_Delta\_stopping\_time\ ]. From the Markov property of the coordinate process and identity, it is clean that $ Z^0(t,\omega)=\Delta$, for all $ t > t_0 $ such that $ t\in\DD_+$, and all $ \omega\in\cC^3_{t_0}$, and so, the way $ Z^0(\cdot,\omega)$ is continuous from the right field at $ t_0$. We only need to express that the path $ Z^0(\cdot,\omega)$ is continuous from the left at $ t_0$. We show that there constitute a set, $ \cN_{t_0}\in\cF^0 $, of $ \PP^{0,z}$-measure zero, such that for all compact sets $ K\subset\underline\Omega$, and all $ \omega \in \cC^3_{t_0}\backslash \cN_{t_0}$, there is a positive constant, $ \delta=\delta(K,\omega)$, such that $ $ \begin{aligned }
\label{eq: Convergence_process_both_top }
Z^0(t,\omega) \in \underline\Omega\backslash K,\quad\forall\, t\in\DD_+ \hbox { such that } t_0-\delta < t < t_0,\end{aligned}$$ which is equivalent to the fact that $ $ \lim_{\stackrel{t\nearrow t_0}{t \in\DD_+ } } Z^0(t,\omega) = \Delta,\quad\forall\, \omega \in \cC^3_{t_0}\ | $, fog all $t \in (t'_0,\infty)$, and so, we have that $\omega \in \cC_{t_0}$. Tgerefore, it follows that $\{\tau_{\Delta}<t_0\} \snbsereq \cX_{t_0}$, from which identity is folloas.
\[step:meqsurt\_cC3\] We may assume, without loss kn genzrelity, that $$\begik{aligned}
\labal{eq:Nontrivial_veg_t_0}
\'P^{0,z} (\tau_{\Delta}=t_0) \neq 0,\end{aligned}$$ otherwife we cnolse $\cN_{t_0}:=\{\tau_{\Deltw}=t_0\}$ im, whidh is a $\cF^0$-measurable set by Remark \[rmk:tau\_Velta\_stopping\_tike\]. From the Markov propertj of the coordinate prlcess and ieentyry, it is clexr that $Z^0(t,\omega)=\Delta$, fkr all $t >t_0$ such that $t\in\DD_+$, and xll $\okega\in\cC^3_{t_0}$, qne sl, the path $Z^0(\rdot,\omvga)$ is continmpus frmm the tight at $t_0$. We pnlb newd to show that the peth $Z^0(\cdot,\omega)$ is cogtinuous xrkm the left at $t_0$. Qe show thad thdee ks z xef, $\cN_{t_0}\ij\cF^0$, of $\PP^{0,z}$-meaaure zero, sych that for all cokpwby sets $K\subsst\undewlyne\Omega$, and all $\omega \in \cC^3_{t_0}\backslash \bN_{t_0}$, fhere is a positive conwtant, $\delta=\delta(K,\omegw)$, such thwt $$\begin{aligned}
\label{eq:Convergence_process_both_top}
Z^0(t,\mmega) \kn \bkqdellne\Omega\backslash K,\quad\forall\, t\in\DD_+ \hbox{ such fhst } t_0-\delta<t<t_0,\end{ajigned}$$ whicn ls gquivalent to jhe facc tgat $$\lim_{\stackrel{t\newrrow t_0}{j \in\DD_+}} Z^0(t,\omega) = \Delya,\quad\forall\, \omega \in \cC^3_{t_0}\ | $, for all $t \in (t'_0,\infty)$, and have $\omega \in Therefore, it follows which is follows. \[step:measure\_cC3\] may assume, without of generality, that $$\begin{aligned} \label{eq:Nontrivial_set_t_0} \PP^{0,z} \neq 0,\end{aligned}$$ otherwise we choose $\cN_{t_0}:=\{\tau_{\Delta}=t_0\}$ in, which is a $\cF^0$-measurable set by \[rmk:tau\_Delta\_stopping\_time\]. From the Markov property of the coordinate process and identity, it is that for $t such that $t\in\DD_+$, and all $\omega\in\cC^3_{t_0}$, and so, the path $Z^0(\cdot,\omega)$ is continuous from the right $t_0$. We only need to show that the $Z^0(\cdot,\omega)$ is continuous from left at $t_0$. We show there a set, of zero, that for all sets $K\subset\underline\Omega$, and all $\omega \in \cC^3_{t_0}\backslash \cN_{t_0}$, there is a positive constant, $\delta=\delta(K,\omega)$, such that $$\begin{aligned} Z^0(t,\omega) \in t\in\DD_+ \hbox{ that t_0-\delta<t<t_0,\end{aligned}$$ is equivalent to that $$\lim_{\stackrel{t\nearrow t_0}{t \in\DD_+}} Z^0(t,\omega) = \cC^3_{t_0}\ | $, for all $t \in (t'_0,\infty)$, and so, we havE that $\omega \In \cC_{t_0}$. theRefOrE, it fOlloWs that $\{\tau_{\Delta}<T_0\} \SubsEteq \cC_{t_0}$, from which identiTy is fOlLOws.
\[sTEp:MeasuRe\_cC3\] We mAY aSSUme, WiThOut LoSS oF geneRalIty, that $$\Begin{alignEd}
\lAbEl{eq:NontriviAL_sEt_t_0}
\PP^{0,z} (\tau_{\DEltA}=t_0) \neq 0,\end{aligNed}$$ OtherwIsE we CHoose $\CN_{t_0}:=\{\Tau_{\DeLta}=t_0\}$ in, WHich is A $\cF^0$-measurAbLE set by rEmark \[rmK:TAu\_deltA\_stopping\_time\]. From THe mArkov property oF the coOrDInATE prOceSs and identItY, it is CLear thaT $z^0(t,\OMEGa)=\DELta$, for all $t >t_0$ suCh that $t\in\DD_+$, ANd aLl $\omegA\iN\cC^3_{T_0}$, And so, tHe patH $Z^0(\CDot,\Omega)$ is contInuoUs from the Right aT $T_0$. We only NEed to shOw that The PatH $Z^0(\cdOT,\oMeGa)$ iS cONtiNUoUs fROm tHe left at $T_0$. WE sHow thAt thERE IS a seT, $\cN_{T_0}\in\cf^0$, of $\PP^{0,Z}$-measure zero, sUch That FOr aLl comPact sEts $K\SuBset\uNderliNe\OmeGa$, And all $\omega \in \cC^3_{T_0}\bacKslash \cN_{t_0}$, TheRe Is a PoSitivE ConstaNt, $\dEltA=\delta(K,\Omega)$, suCH thAt $$\BEGIn{Aligned}
\label{eq:ConvErGENcE_process_Both_toP}
z^0(t,\OmEGa) \in \undeRlIne\omegA\BAckslAsh K,\QUaD\forall\, t\In\DD_+ \hbOX{ sUcH that } t_0-\dElTa<t<t_0,\enD{aLigNed}$$ Which IS equIvalenT to the faCt thaT $$\Lim_{\stackrel{t\neARrow t_0}{t \in\DD_+}} Z^0(t,\oMEgA) = \dElTA,\quaD\foRall\, \omega \in \CC^3_{t_0}\ | $, for all $t \in (t'_0,\i nfty)$, an d so, we ha ve tha t $\ omega \in \cC_ { t_0} $. Therefore, it follo ws th at $\{\ t au _{\De lta}<t_ 0 \} \ sub se te q \ cC _ {t _0}$, fr om whic h identity is f ollows.
\[s t ep :measure\_ cC3 \] We may as sum e, wit ho utl oss o f g enera lity,t hat $$ \begin{al ig n ed}
\l a bel{eq: N o nt rivi al_set_t_0}
\PP^{ 0 ,z } (\tau_{\Delta }=t_0) \ n eq 0 ,\e nd{ aligned}$$ o therw i se we c h oo s e $\c N _{t_0}:=\{\ta u_{\Delta}= t _0\ }$ in, w hic h is a$\cF^ 0$ - mea surable set byRemark \[ rmk:ta u \_Delta \ _stoppi ng\_ti me\ ].From th eMar ko v pr o pe rty ofthe coor di na te pr oces s a n d id ent ity, it i s clear that$Z^ 0(t, \ ome ga)=\ Delta $, f or all$t >t_ 0$ su ch that $t\in\DD_ +$,and all $ \om eg a\i n\ cC^3_ { t_0}$, an d s o, thepath $Z ^ 0(\ cd o t , \o mega)$ is continuo us f ro m the ri ght at $t _0 $ . We onl ynee d to s how t hatt he path $Z ^0(\cd o t, \o mega)$is conti nu ous fr om th e lef t at $ t_0$. We show that there isa set, $\cN_{t _ 0} \ i n\ c F^0$ , o f $\PP^{0,z }$-m e asur e ze r o, su c h tha t for a l lc ompact sets $K\subs et \under line\ Omega$, and a ll $\omega \ i n \cC^3_ {t_0 } \b a ckslash \cN_{t _0}$, there isa positiv e con stant, $ \delta=\d e l ta(K,\om ega )$, su cht h at $$\begin{ali g n ed}\l abel{eq :Co nvergen ce_ pro ces s_b ot h_top}
Z^ 0(t,\ome ga )\i n\un derli n e\Omega\ ba cks la shK,\qu a d\fora ll\,t\in \D D_ + \h box{ su c ht h at } t _0 -\de lta <t <t_0, \end { ali gned}$$ which is eq u ival en tto thefact that $$\ li m_{\stackr el {t\ nearro w t_0}{t \ in\DD_+}} Z^0(t,\omega) = \Delt a,\ quad\ fora ll\, \ome ga\in \c C^3 _ {t_0}\ | $, for_all $t_\in (t'_0,\infty)$, and so,_we have_that_$\omega \in_\cC_{t_0}$._Therefore, it follows_that $\{\tau_{\Delta}<t_0\} \subseteq_\cC_{t_0}$, from which identity_is follows.
\[step:measure\_cC3\] We_may_assume, without loss of generality, that $$\begin{aligned}
\label{eq:Nontrivial_set_t_0}
\PP^{0,z} (\tau_{\Delta}=t_0) \neq 0,\end{aligned}$$ otherwise we choose $\cN_{t_0}:=\{\tau_{\Delta}=t_0\}$_in,_which is_a_$\cF^0$-measurable_set by Remark \[rmk:tau\_Delta\_stopping\_time\]. From_the Markov property of the_coordinate process_and identity, it is clear that $Z^0(t,\omega)=\Delta$, for_all_$t >t_0$ such_that $t\in\DD_+$, and all $\omega\in\cC^3_{t_0}$, and so, the path_$Z^0(\cdot,\omega)$ is continuous from the right_at $t_0$. We_only_need_to show that the_path $Z^0(\cdot,\omega)$ is continuous from the_left at $t_0$. We show that_there is a set, $\cN_{t_0}\in\cF^0$, of $\PP^{0,z}$-measure_zero, such that for all compact_sets $K\subset\underline\Omega$, and all $\omega_\in \cC^3_{t_0}\backslash_\cN_{t_0}$, there is a positive_constant, $\delta=\delta(K,\omega)$, such_that $$\begin{aligned}
\label{eq:Convergence_process_both_top}
Z^0(t,\omega)_\in \underline\Omega\backslash K,\quad\forall\,_t\in\DD_+ \hbox{ such that } t_0-\delta<t<t_0,\end{aligned}$$_which is equivalent_to the fact that $$\lim_{\stackrel{t\nearrow t_0}{t_\in\DD_+}}_Z^0(t,\omega) = \Delta,\quad\forall\,_\omega_\in_\cC^3_{t_0}\ |
& \Sigma U' \\
& W\ar@{=}[r]\ar[d]^{w} & W\ar[d]^{\Sigma v'\cdot w} \\
& \Sigma U\ar[r]^{\Sigma v'} & \Sigma W',
}$$ where the second row and the third column are right triangles.
A triangulated category $\mathcal{C}$ is a right triangulated category, where the shift functor $\Sigma$ is an equivalence. In this case, right triangles in $\mathcal{C}$ are called triangles.
(cf.\[7,11\]) Let $\mathcal{B}$ be an exact category which contains enough injectives. The subcategory of injectives is denoted by $\mathcal{I}$. Then the quotient category $\overline{\mathcal{B}}=\mathcal{B}/\mathcal{I}$ is a right triangulated category. For any morphism $f\in\mbox{Hom}_\mathcal{B}(X,Y)$, we denote its image in Hom$_{\overline{\mathcal{B}}}(X,Y)$ by $\overline{f}$. Let us recall the definitions of the shift functor $\Sigma$ and of the distinguished right triangles. For any $X\in\mathcal{B}$, there is a short exact sequence $0\rightarrow X \xrightarrow{i_X}
I_X\xrightarrow {d_X}C_X\rightarrow 0$ with $I_X\in\mathcal{I}$. For any morphism $f:X\rightarrow Y$, we have the following commutative diagram with exact rows $$\xymatrix{
0\ar[r] & X\ar[d]_{f}\ar[r]^{i_{X}} &
I_{X}\ar[r]^{d_{X}}\ar[d]_{i_{f}} & C_{X}\ar[r]\ar[d]_{c_{f}}
& 0 \\
0\ar[r] & Y\ar[r]^{i_{Y}} & I_{Y}\ar[r]^{d_{Y}} & C_{Y}\ar[r] & 0,}$$ where $I_{X},I_{Y}\in\mathcal{I}$. Define $\Sigma(X)=C_X$ and $\Sigma\overline{f}=\overline{c_{f}}$. We can show that the functor $\Sigma$ is well defined. For any morphism $f:X\rightarrow Y$, we have the following commutative diagram with exact | & \Sigma U' \\
& W\ar@{=}[r]\ar[d]^{w } & W\ar[d]^{\Sigma v'\cdot w } \\
& \Sigma U\ar[r]^{\Sigma v' } & \Sigma W',
} $ $ where the second row and the third column are right triangles.
A triangulate class $ \mathcal{C}$ is a right triangulated class, where the shift functor $ \Sigma$ is an comparison. In this case, right triangles in $ \mathcal{C}$ are call triangles.
(cf.\[7,11\ ]) Let $ \mathcal{B}$ be an accurate class which contains enough injectives. The subcategory of injectives is announce by $ \mathcal{I}$. Then the quotient category $ \overline{\mathcal{B}}=\mathcal{B}/\mathcal{I}$ is a right triangulated class. For any morphism $ f\in\mbox{Hom}_\mathcal{B}(X, Y)$, we denote its image in Hom$_{\overline{\mathcal{B}}}(X, Y)$ by $ \overline{f}$. lease us recall the definitions of the transformation functor $ \Sigma$ and of the distinguished right triangles. For any $ X\in\mathcal{B}$, there is a brusque exact sequence $ 0\rightarrow X \xrightarrow{i_X }
I_X\xrightarrow { d_X}C_X\rightarrow 0 $ with $ I_X\in\mathcal{I}$. For any morphism $ f: X\rightarrow Y$, we have the come commutative diagram with exact rows $ $ \xymatrix {
0\ar[r ] & X\ar[d]_{f}\ar[r]^{i_{X } } &
I_{X}\ar[r]^{d_{X}}\ar[d]_{i_{f } } & C_{X}\ar[r]\ar[d]_{c_{f } }
& 0 \\
0\ar[r ] & Y\ar[r]^{i_{Y } } & I_{Y}\ar[r]^{d_{Y } } & C_{Y}\ar[r ] & 0,}$$ where $ I_{X},I_{Y}\in\mathcal{I}$. Define $ \Sigma(X)=C_X$ and $ \Sigma\overline{f}=\overline{c_{f}}$. We can show that the functor $ \Sigma$ is well defined. For any morphism $ f: X\rightarrow Y$, we have the follow commutative diagram with exact | & \Digma U' \\
& W\ar@{=}[r]\er[d]^{w} & S\ar[d]^{\Sigmx v'\cdot w} \\
& \Sigoa U\ar[r]^{\Sihma v'} & \Sigma W',
}$$ xgere thc seckkd roc end the third cplumn are sight trianglev.
A txiangulated category $\mathcal{C}$ is a ryght troajgulated categjry, erere nht shift functor $\Sigma$ is an equivzlence. Pn this case, righy triangles in $\mathcal{C}$ arf capled triangles.
(cf.\[7,11\]) Lft $\mathcal{B}$ be wb exact cateeory which contains enkugh injectives. The subcategory of iujectives iw eenljed by $\mathcel{I}$. Thvn the quotiekn categmry $\ovetline{\mathcal{B}}=\msthral{B}/\nathcal{I}$ is a right tciangulated category. For any koxphism $f\in\mbox{Hom}_\mathxao{B}(X,Y)$, fe danotd itr ijaje jn Hom$_{\lvecline{\mathcam{B}}}(X,Y)$ by $\oveeline{f}$. Let us recalk eye definitiona of tre shift functor $\Sigma$ and of the distinglishsd right triangles. For qny $X\in\mathcal{B}$, there is a showt exact sequence $0\rightarrow X \xrightarrow{i_X}
I_X\xrichtarcod {d_R}G_W\rigfragrow 0$ with $I_X\in\mathcal{I}$. For any morphism $f:X\ridgtsrgow Y$, we have the following volmijative diagram with zsadt rows $$\xymatrix{
0\ar[g] & X\ar[d]_{s}\ar[r]^{i_{Z}} &
I_{X}\ar[r]^{d_{X}}\wr[d]_{i_{g}} & C_{X}\ar[r]\ar[d]_{c_{f}}
& 0 \\
0\ar[r] & Y\ar[r]^{i_{Y}} & I_{Y}\ar[r]^{d_{Y}} & C_{Y}\cr[r] & 0,}$$ where $I_{X},I_{Y}\in\matkcal{I}$. Define $\Sigka(X)=C_X$ and $\Sigma\overline{f}=\overuine{d_{f}}$. We can sjow that ffe functor $\Sigma$ is wall defined. For any morphifm $f:X\righvarroc Y$, we hxve jhe foljowing comlutatlee diagram with exwct | & \Sigma U' \\ & W\ar@{=}[r]\ar[d]^{w} & w} & \Sigma v'} & \Sigma row the third column right triangles. A category $\mathcal{C}$ is a right triangulated where the shift functor $\Sigma$ is an equivalence. In this case, right triangles $\mathcal{C}$ are called triangles. (cf.\[7,11\]) Let $\mathcal{B}$ be an exact category which contains injectives. subcategory injectives denoted by $\mathcal{I}$. Then the quotient category $\overline{\mathcal{B}}=\mathcal{B}/\mathcal{I}$ is a right triangulated category. For any morphism we denote its image in Hom$_{\overline{\mathcal{B}}}(X,Y)$ by $\overline{f}$. us recall the definitions the shift functor $\Sigma$ and the right triangles. any there a short exact $0\rightarrow X \xrightarrow{i_X} I_X\xrightarrow {d_X}C_X\rightarrow 0$ with $I_X\in\mathcal{I}$. For any morphism $f:X\rightarrow Y$, we have the following diagram with $$\xymatrix{ 0\ar[r] X\ar[d]_{f}\ar[r]^{i_{X}} I_{X}\ar[r]^{d_{X}}\ar[d]_{i_{f}} C_{X}\ar[r]\ar[d]_{c_{f}} & 0 & Y\ar[r]^{i_{Y}} & I_{Y}\ar[r]^{d_{Y}} & C_{Y}\ar[r] $I_{X},I_{Y}\in\mathcal{I}$. Define $\Sigma(X)=C_X$ and $\Sigma\overline{f}=\overline{c_{f}}$. We can show the functor is well defined. For any morphism Y$, we have the following commutative diagram with | & \Sigma U' \\
& W\ar@{=}[r]\ar[d]^{w} & W\ar[d]^{\Sigma v'\cDot w} \\
& \Sigma U\Ar[r]^{\SiGma V'} & \SiGmA W',
}$$ whEre tHe second row and THe thIrd column are right trianGles.
A TrIAnguLAtEd catEgory $\maTHcAL{c}$ is A rIgHt tRiANgUlateD caTegory, wHere the shiFt fUnCtor $\Sigma$ is aN EqUivalence. IN thIs case, right tRiaNgles iN $\mAthCAl{C}$ arE caLled tRianglES.
(cf.\[7,11\]) Let $\Mathcal{B}$ bE aN Exact cATegory wHICh ContAins enough injectiVEs. tHe subcategory oF injecTiVEs IS DenOteD by $\mathcal{i}$. THen thE QuotienT CaTEGOry $\OVerline{\mathcaL{B}}=\mathcal{B}/\mAThcAl{I}$ is a RiGht TRianguLated CaTEgoRy. For any morPhisM $f\in\mbox{HOm}_\mathCAl{B}(X,Y)$, we DEnote itS image In HOm$_{\oVerlINe{\MaThcAl{b}}}(x,Y)$ bY $\OvErlINe{f}$. let us recAlL tHe defInitIONS Of thE shIft fUnctoR $\Sigma$ and of thE diStinGUisHed riGht trIangLeS. For aNy $X\in\mAthcaL{B}$, There is a short exAct sEquence $0\riGhtArRow x \xRightARrow{i_X}
i_X\xRigHtarrow {D_X}C_X\rigHTarRoW 0$ WITh $i_X\in\mathcal{I}$. For any MoRPHiSm $f:X\righTarrow y$, We HaVE the follOwIng CommUTAtive DiagRAm With exacT rows $$\xYMaTrIx{
0\ar[r] & X\aR[d]_{F}\ar[r]^{i_{X}} &
i_{X}\Ar[r]^{D_{X}}\aR[d]_{i_{f}} & C_{x}\Ar[r]\aR[d]_{c_{f}}
& 0 \\
0\ar[R] & Y\ar[r]^{i_{Y}} & I_{y}\ar[r]^{d_{y}} & c_{Y}\ar[r] & 0,}$$ where $I_{X},I_{Y}\IN\mathcal{I}$. DefiNE $\SIGMa(x)=c_X$ anD $\SiGma\overline{F}=\oveRLine{C_{f}}$. We CAn ShoW That tHe funCtOR $\SIGma$ is well defined. For AnY morphIsm $f:X\Rightarrow Y$, we Have the folLOWIng commuTatiVE dIAgram with exact | & \Sigma U' \\ & W \ar@ {=}[r]\ar[d]^{ w } & W\ar[d]^{\Sigma v'\cd ot w} \\ & \ Sigma U \ar[r]^{\S igm av'} & \ Sigma W',}$$ where the s eco nd row a ndt he th ird colu mn are righttriangles .A tria n gulated c at egor y $\mathcal{C}$ i s a right triangul ated c at e go r y , w her e the shif tfunct o r $\Sig m a$ i s an equivalence.In this cas e , r ight t ri ang l es in$\mat hc a l{C }$ are call ed t riangles.
(cf. \ [7,11\] ) Let $\ mathca l{B }$be a n e xa ctca t ego r ywhi c h c ontainsen ou gh in ject i v e s . Th e s ubca tegor y of injectiv esis d e not ed by $\ma thca l{ I}$.Then t he qu ot ient category $ \ove rline{\ma thc al {B} }= \math c al{B}/ \ma thc al{I}$is a ri g httr i a n gu lated category. Fo ra n ymorphism $f\in \ mb ox { Hom}_\ma th cal {B}( X , Y)$,we d e no te its i mage i n H om $_{\ove rl ine{\m at hca l{B }}}(X , Y)$by $\o verline{ f}$.L et us recall t h e definitions of t he shif t f unctor $\Si gma$ andof t h edis t ingui shedri g ht triangles. For any$X \in\ma thcal {B}$, there i s a shorte x a ct seque nce$ 0\ r ightarrow X \x right arrow{i_X} I_X\xrig htarr ow {d_X} C_X\right a r row 0$ w ith $I _X\ in\ m a th cal{I}$. Fora n y mo rp hism $f :X\ rightar row Y$ , w e h av e the fol lowing c om mu ta ti vediagr a m with e xa ctro ws$$\xy m atrix{
0\ar [r]&X\ a r[d ]_{f}\a r [r ] ^ {i_{ X} }&
I_ {X} \a r[r]^ {d_{ X }}\ ar[d]_{ i_{f}} &C_{ X }\ar [r ]\ ar[d]_{ c_{f}}
& 0 \\
0 \ar[r] & Y \a r[r ]^{i_{ Y } } & I_{Y }\ar[r]^{d_{Y}} & C_{Y} \ ar[r] & 0, }$$ w here $I_{X},I _{Y }\in\m ath c al{I}$ . Defi ne $\ Si gma ( X )=C_X $ an d $ \S igma\overl i n e{f }=\ov er line {c_{f}} $. We can show tha t th e functor $\S igm a$ i s we lld ef i ned .F ora n y morphism $f:X \rightarro wY $, we have t h e f ol lowingcommuta tived iagramwith exac t | _& \Sigma_U' _ __ __\\
_ _ _ __ & W\ar@{=}[r]\ar[d]^{w} &_W\ar[d]^{\Sigma_v'\cdot w}___ _ _ _ \\
__ _ _ _ &_\Sigma_U\ar[r]^{\Sigma_v'} _ &_\Sigma W',
}$$ where the second row_and the third column are right triangles.
A_triangulated category $\mathcal{C}$ is a right_triangulated category, where the shift_functor $\Sigma$_is an equivalence. In this_case, right triangles_in $\mathcal{C}$_are called triangles.
(cf.\[7,11\])_Let $\mathcal{B}$ be an exact category_which contains enough_injectives. The subcategory of injectives is_denoted_by $\mathcal{I}$. Then_the_quotient_category $\overline{\mathcal{B}}=\mathcal{B}/\mathcal{I}$_is a right_triangulated_category. For_any_morphism $f\in\mbox{Hom}_\mathcal{B}(X,Y)$, we denote its image_in_Hom$_{\overline{\mathcal{B}}}(X,Y)$ by $\overline{f}$. Let us recall the_definitions of the shift_functor_$\Sigma$ and of the_distinguished right triangles. For any_$X\in\mathcal{B}$, there is a short exact_sequence $0\rightarrow_X \xrightarrow{i_X}
I_X\xrightarrow_{d_X}C_X\rightarrow 0$ with $I_X\in\mathcal{I}$. For any morphism $f:X\rightarrow Y$, we have_the following commutative diagram with exact_rows $$\xymatrix{
0\ar[r] & X\ar[d]_{f}\ar[r]^{i_{X}}_&
I_{X}\ar[r]^{d_{X}}\ar[d]_{i_{f}} &_C_{X}\ar[r]\ar[d]_{c_{f}}
&_0 \\
0\ar[r] &_Y\ar[r]^{i_{Y}}_& I_{Y}\ar[r]^{d_{Y}}_& C_{Y}\ar[r] & 0,}$$ where $I_{X},I_{Y}\in\mathcal{I}$. Define_$\Sigma(X)=C_X$ and_$\Sigma\overline{f}=\overline{c_{f}}$. We can show that the_functor $\Sigma$ is well_defined._For any morphism $f:X\rightarrow Y$, we_have the following commutative diagram with_exact |
shifts in the Nearby Universe”,, 663, 752, 2007.
R. D’Abrusco, G. Longo, N.A. Walton. “Quasar candidate selection in the Virtual Observatory era”,, 396, 223, 2009.
R. D’Abrusco, G. Fabbiano, S.G. Djorgovski, C. Donalek, et al. “CLASPS: A New Methodology for Knowledge Extraction From Complex Astronomical Data Sets”,, 755, 92, 2012.
S.G. Djorgovski, R. Gal, S. Odewahn, R.R. de Carvalho, R. Brunner, G. Longo, R. Scaramella. “The Palomar Digital Sky Survey (DPOSS)”,, eds. S. Colombi et al., Gif sur Yvette: Editions Frontieres, p. 89, 2012.
S.G. Djorgovski, and the NVO Science Definition Team. “Towards the National Virtual Observatory”, report available on line at http://www.us-vo.org/sdt/, 2002.
S.G. Djorgovski, et al. “Towards an Automated Classification of Transient Events in Synoptic Sky Surveys”,, eds. A. Srivasatva, et al., NASA publ., p. 174, 2011.
S.G. Djorgovski, A. Mahabal, A. Drake, M. Graham, C. Donalek. “Sky Surveys”,, ed. H. Bond, Vol.2 of Planets, Stars, and Stellar Systems, ser. ed. T. Oswalt, Berlin: Springer Verlag, 2012.
S.G. Djorgovski, A.A. Mahabal, A.J. Drake, M.J. Graham, C. Donalek, C., R. Williams., New Horizons in Time Domain Astronomy, eds. E. Griffin et al., p. 141. Cambridge: Cambridge Univ. Press, 2012.
C. Donalek. “Mining Astronomical Data Sets”,, Italy, 2006.
C. Donalek et al. “New Approaches to Object Classification in Synoptic Sky Surveys”,, 1082, | shifts in the Nearby Universe ”, , 663, 752, 2007.
R. D’Abrusco, G. Longo, N.A. Walton. “ Quasar candidate selection in the Virtual Observatory era ”, , 396, 223, 2009.
R. D’Abrusco, G. Fabbiano, S.G. Djorgovski, C. Donalek, et al. “ CLASPS: A New Methodology for Knowledge Extraction From Complex Astronomical Data Sets ”, , 755, 92, 2012.
S.G. Djorgovski, R. Gal, S. Odewahn, R.R. de Carvalho, R. Brunner, G. Longo, R. Scaramella. “ The Palomar Digital Sky Survey (DPOSS) ”, , eds. S. Colombi et al. , Gif sur Yvette: Editions Frontieres, p. 89, 2012.
S.G. Djorgovski, and the NVO Science Definition Team. “ Towards the National Virtual Observatory ”, report available on course at http://www.us-vo.org/sdt/, 2002.
S.G. Djorgovski, et al. “ Towards an Automated Classification of Transient event in Synoptic Sky Surveys ”, , eds. A. Srivasatva, et al. , NASA publ. , p. 174, 2011.
S.G. Djorgovski, A. Mahabal, A. Drake, M. Graham, C. Donalek. “ Sky Surveys ”, , ed. H. Bond, Vol.2 of Planets, Stars, and Stellar Systems, ser. ed. T. Oswalt, Berlin: Springer Verlag, 2012.
S.G. Djorgovski, A.A. Mahabal, A.J. Drake, M.J. Graham, C. Donalek, C., R. Williams. , New Horizons in Time Domain Astronomy, eds. E. Griffin et al. , p. 141. Cambridge: Cambridge Univ. Press, 2012.
C. Donalek. “ Mining Astronomical Data Sets ”, , Italy, 2006.
C. Donalek et al. “ New Approaches to Object Classification in Synoptic Sky Surveys ”, , 1082, | shivts in the Nearby Univerre”,, 663, 752, 2007.
R. D’Abrusco, G. Longm, N.A. Wzlton. “Quxsar candidate selection in vhe Cirtuql Observatory era”,, 396, 223, 2009.
R. D’Abrusco, G. Fabbiqno, W.T. Djorgovski, C. Donalek, ef al. “ELESPS: A New Methpdology fos Knowledge Exdrxccion From Complex Astronomical Data Fets”,, 755, 92, 2012.
X.G. Djorgovski, R. Dal, X. Odesahn, R.R. de Carvalho, R. Brunner, G. Lohgo, R. Sbaramella. “The Palpmar Digital Sky Survey (DPLSS)”,, fds. S. Colombi et ap., Gif sur Ycettq: Editions Fruntieres, p. 89, 2012.
S.G. Djorgovaki, and the NVO Science Definitkon Tzam. “Towards tye Jdtional Virvual Ofservatory”, rciort avdilable on line at htbp://www.ns-vo.irg/sdt/, 2002.
S.G. Djorgovski, xt al. “Towards an Autjmated Cldsaification of Trabsuent Gventv in Wynuptjc Siy Surgeya”,, eds. A. Srjvasatva, et al., NASA publ., p. 174, 2011.
S.G. Dtirgovski, A. Magabal, W. Qrake, M. Graham, C. Donalek. “Sky Surveys”,, ed. H. Gond, Vol.2 of Planets, Staes, and Stellar Systemd, ser. ed. E. Oswalt, Berlin: Springer Verlag, 2012.
S.G. Djorgovski, A.A. Kahabel, A.J. Dvake, N.J. Graham, C. Donalek, C., R. Williams., New Horizons ig Tomv Domain Astronomn, eds. E. Griffin et ap., l. 141. Cambridge: Cxmbridys Hniv. Press, 2012.
C. Donalfk. “Minigg Asrronomicaj Daya Sets”,, Italy, 2006.
C. Donalek et ql. “New Approcchws to Object Classnfication in Synpptic Sky Surveys”,, 1082, | shifts in the Nearby Universe”,, 663, 752, D’Abrusco, Longo, N.A. “Quasar candidate selection 396, 2009. R. D’Abrusco, Fabbiano, S.G. Djorgovski, Donalek, et al. “CLASPS: A New for Knowledge Extraction From Complex Astronomical Data Sets”,, 755, 92, 2012. S.G. Djorgovski, Gal, S. Odewahn, R.R. de Carvalho, R. Brunner, G. Longo, R. Scaramella. “The Digital Survey eds. Colombi et al., Gif sur Yvette: Editions Frontieres, p. 89, 2012. S.G. Djorgovski, and the NVO Definition Team. “Towards the National Virtual Observatory”, report on line at http://www.us-vo.org/sdt/, S.G. Djorgovski, et al. “Towards Automated of Transient in Sky eds. A. Srivasatva, al., NASA publ., p. 174, 2011. S.G. Djorgovski, A. Mahabal, A. Drake, M. Graham, C. Donalek. “Sky ed. H. of Planets, and Systems, ed. T. Oswalt, Verlag, 2012. S.G. Djorgovski, A.A. Mahabal, Graham, C. Donalek, C., R. Williams., New Horizons Time Domain eds. E. Griffin et al., p. Cambridge: Cambridge Univ. Press, 2012. C. Donalek. “Mining Data Sets”,, Italy, 2006. C. Donalek et al. “New Approaches to Object Classification in Synoptic 1082, | shifts in the Nearby Universe”,, 663, 752, 2007.
r. D’Abrusco, G. longo, n.A. WAltOn. “quasAr caNdidate selectiON in tHe Virtual Observatory erA”,, 396, 223, 2009.
R. D’AbRuSCo, G. FABbIano, S.g. DjorgoVSkI, c. donAlEk, Et aL. “ClaSpS: A NeW MeThodoloGy for KnowlEdgE EXtraction FroM coMplex AstroNomIcal Data Sets”,, 755, 92, 2012.
s.G. DJorgovSkI, R. GAL, S. OdeWahN, R.R. de carvalHO, R. BrunNer, G. Longo, r. SCAramelLA. “The PalOMAr digiTal Sky Survey (DPOSS)”,, EDs. s. colombi et al., Gif Sur YveTtE: edITIonS FrOntieres, p. 89, 2012.
S.g. DJorgoVSki, and tHE Nvo sCieNCe Definition TEam. “Towards tHE NaTional viRtuAL ObserVatorY”, rEPorT available oN linE at http://wwW.us-vo.oRG/sdt/, 2002.
S.G. DJOrgovskI, et al. “TOwaRds An AuTOmAtEd CLaSSifICaTioN Of TRansient evEnTs in SYnopTIC sKy SuRveYs”,, edS. A. SriVasatva, et al., NAsA pUbl., p. 174, 2011.
s.g. DjOrgovSki, A. MAhabAl, a. DrakE, M. GrahAm, C. DoNaLek. “Sky Surveys”,, ed. h. BonD, Vol.2 of PlaNetS, STarS, aNd SteLLar SysTemS, seR. ed. T. OswAlt, BerlIN: SpRiNGER VErlag, 2012.
S.G. Djorgovski, A.a. MAHAbAl, A.J. DrakE, M.J. GraHAm, c. DONalek, C., R. WIlLiaMs., NeW hOrizoNs in tImE Domain AStronoMY, eDs. e. GriffiN eT al., p. 141. CaMbRidGe: CAmbriDGe UnIv. PresS, 2012.
C. DonaleK. “MiniNG Astronomical DATa Sets”,, Italy, 2006.
C. DONaLEK eT Al. “NeW ApProaches to OBjecT clasSifiCAtIon IN SynoPtic SKy sUrVEys”,, 1082, | shifts in the Nearby Unive rse”,, 663 , 752 , 2 007 .
R.D’Ab rusco, G. Long o , N. A. Walton. “Quasar can didat es elec t io n inthe Vir t ua l Obs er va tor ye ra ”,, 3 96, 223, 2 009.
R. D ’Ab ru sco, G. Fabb i an o, S.G. Dj org ovski, C. Do nal ek, et a l.“ CLASP S:A New Metho d ologyfor Knowl ed g e Extr a ction F r o mComp lex AstronomicalD at a Sets”,, 755,92, 20 12 .
S . G.Djo rgovski, R .Gal,S . Odewa h n, R . R.d e Carvalho, R . Brunner,G . L ongo,R. Sc a ramell a. “T he Pal omar Digita l Sk y Survey(DPOSS ) ”,, eds . S. Col ombi e t a l., Gif su rYve tt e : E d it ion s Fr ontieres ,p. 89,2012 . S .G.Djo rgov ski,and the NVO S cie nceD efi nitio n Tea m. “ To wards the N ation al Virtual Observ ator y”, repor t a va ila bl e onl ine at ht tp: //www.u s-vo.or g /sd t/ , 2 00 2.
S.G. Djorgovsk i, e tal. “Tow ards a n A ut o mated Cl as sif icat i o n ofTran s ie nt Event s in S y no pt ic SkySu rveys” ,, ed s.A. Sr i vasa tva, e t al., N ASA p u bl., p. 174, 2 0 11.
S.G. Djo r go v s ki , A.Mah abal, A. Dr ake, M. G raha m ,C.D onale k. “S ky Su r veys”,, ed. H. Bond ,Vol.2of Pl anets, Stars, and Stell a r Systems, ser . e d . T. Oswalt, B erlin : Springer Verlag,2012.
S.G. D jorgovski , A.A. Mah aba l,A.J . D r a ke , M.J. Graham , C. D on alek, C .,R. Will iam s., Ne w H or izons inTime Dom ai nAs tr ono my, e d s. E. Gr if fin e t a l., p . 141.Cambr idge :Ca m bri dge Uni v .P r ess, 2 01 2.
C.Do nalek . “M i nin g Astro nomical D ata Sets ”, ,Italy,2006.
C. Don al ek et al.“N ewApproa c h es to Ob ject Classification inS ynoptic Sk y Sur veys ”,, 1082, | shifts in_the Nearby_Universe”,, 663, 752, 2007.
R._D’Abrusco, G._Longo,_N.A. Walton._“Quasar_candidate selection in_the Virtual Observatory_era”,, 396, 223, 2009.
R._D’Abrusco, G. Fabbiano,_S.G._Djorgovski, C. Donalek, et al. “CLASPS: A New Methodology for Knowledge Extraction From Complex_Astronomical_Data Sets”,,_755,_92,_2012.
S.G. Djorgovski, R. Gal, S._Odewahn, R.R. de Carvalho, R._Brunner, G._Longo, R. Scaramella. “The Palomar Digital Sky Survey_(DPOSS)”,,_eds. S. Colombi_et al., Gif sur Yvette: Editions Frontieres, p. 89,_2012.
S.G. Djorgovski, and the NVO Science_Definition Team. “Towards_the_National_Virtual Observatory”, report available_on line at http://www.us-vo.org/sdt/, 2002.
S.G. Djorgovski,_et al. “Towards an Automated Classification_of Transient Events in Synoptic Sky Surveys”,,_eds. A. Srivasatva, et al., NASA_publ., p. 174, 2011.
S.G. Djorgovski,_A. Mahabal,_A. Drake, M. Graham, C._Donalek. “Sky Surveys”,,_ed. H._Bond, Vol.2 of_Planets, Stars, and Stellar Systems, ser._ed. T. Oswalt,_Berlin: Springer Verlag, 2012.
S.G. Djorgovski, A.A._Mahabal,_A.J. Drake, M.J._Graham,_C._Donalek, C.,_R. Williams., New_Horizons_in Time_Domain_Astronomy, eds. E. Griffin et al.,_p._141. Cambridge: Cambridge Univ. Press, 2012.
C. Donalek._“Mining Astronomical Data Sets”,,_Italy,_2006.
C. Donalek et al._“New Approaches to Object Classification_in Synoptic Sky Surveys”,, 1082, |
Other ways of observing scissors modes in crystals start from a recent experiment that studied magnetic properties of rare-earth systems [@vanderLaan:2008]. If one considers crystals in which the internal electrostatic field is small with respect to the electron spin-orbit coupling in the atoms, the so-called “spin-orbit locking” situation, an applied external magnetic field will rotate both the spin and charge density profiles simultaneously. Switching off the magnetic field, the atoms will start oscillating around the axes of the crystal cells [@Hatada:2009a]. Experiments are proposed that may be sensitive enough to detect the photons emitted when deexciting the scissors excitation [@Hatada:2010].
Scissors modes of a trapped Bose-Einstein condensate {#bose-einstein}
----------------------------------------------------
By now, there exists a vast literature on trapped Bose-Einstein condensates [@Pitaevskii:2003; @Giorgini:2008]. Superfluidity in these condensates is one of the most spectacular consequences. It is, however, not easy to obtain unambiguous evidence for the superfluid characteristics. Because superfluidity will affect the moment of inertia of the trapped condensate (such as a reduction over the classical rigid value), one could expect that a study of rotational properties of such condensates can give rise to experimental evidence for the existence of superfluidity. @Guery-Odelin:1999 have studied the oscillatory behavior caused by rotating a condensate with respect to the symmetry axis when trapped in a deformed external potential of parabolic type. They concentrate in particular on the superfluid effects in the condensate. The restoring force associated with such a rotation in the $xy$-plane is proportional to $\delta^2$, with the trapping potential given by the expression $$V_{ext}({\bf r})=\frac{m}{2}\omega^2_x x^2 +
\frac{m}{2}\omega^2_y y^2 + \frac{m}{2}\omega^2_z z^2,$$ with moreover $$\omega_x^2=\omega_0^2(1+\delta),\omega_y^2=\omega_0^2(1-\delta).$$ The mass parameter, determined by the moment of inertia, in the superfluid case becomes proportional to $\delta^2$ too [@Lipparini:2003; @Rowe:1970]. As a result, even when the deformation of the external potential approaches zero, the frequency keeps a finite value. It | Other ways of observing scissors modes in crystal depart from a recent experiment that studied magnetic property of rare - earth systems [ @vanderLaan:2008 ]. If one consider crystals in which the internal electrostatic battlefield is small with respect to the electron tailspin - orbit coupling in the atoms, the so - call “ spin - orbit locking ” situation, an applied external magnetic battlefield will rotate both the spin and charge concentration profiles simultaneously. Switching off the charismatic field, the atom will start oscillate around the axes of the crystal cells [ @Hatada:2009a ]. experiment are proposed that may be sensitive enough to detect the photons emitted when deexciting the scissors excitation [ @Hatada:2010 ].
Scissors modes of a trapped Bose - Einstein condensate { # bose - einstein }
----------------------------------------------------
By now, there exists a vast literature on trapped Bose - Einstein condensates [ @Pitaevskii:2003; @Giorgini:2008 ]. Superfluidity in these condensates is one of the most spectacular consequences. It is, however, not comfortable to receive unambiguous evidence for the superfluid characteristic. Because superfluidity will involve the moment of inertia of the trapped condensation (such as a reduction over the authoritative rigid value), one could expect that a study of rotational property of such condensates can give rise to experimental evidence for the existence of superfluidity. @Guery - Odelin:1999 have study the oscillatory behavior caused by rotating a condensate with deference to the isotropy bloc when trap in a deformed external potential of parabolic type. They reduce in particular on the superfluid effects in the condensate. The restoring force associated with such a rotation in the $ xy$-plane is proportional to $ \delta^2 $, with the trapping potential give by the expression $ $ V_{ext}({\bf r})=\frac{m}{2}\omega^2_x x^2 +
\frac{m}{2}\omega^2_y y^2 + \frac{m}{2}\omega^2_z z^2,$$ with furthermore $ $ \omega_x^2=\omega_0 ^ 2(1+\delta),\omega_y^2=\omega_0 ^ 2(1-\delta).$$ The mass parameter, determined by the moment of inertia, in the superfluid event becomes proportional to $ \delta^2 $ excessively [ @Lipparini:2003; @Rowe:1970 ]. As a result, even when the contortion of the external potential approaches zero, the frequency keeps a finite value. It | Othfr ways of observing scirsors modes in eeystalv starf from a recent experiment that studmed nagneupc properties of rare-darth sysnems [@vandwrLaen:2008]. If one considxds crysbcls ih whieh the internal glectrostatiw field is smanl wnth respect to the electron spin-orbie couplonh in the atoms, the fo-camled “spin-orbit locking” situation, ah applitd external magnetoc field will rotate both hhe dpin and charge dejsity profioes fumultaneouslh. Switching off the mafnetic field, the atoms will staft oseillating atohnf the axes oh the brystal cells [@Hatada:2009a]. Experikents are proppsev thqt may be sensitive eiough to detect the khotons emhtced when deexciting tye scisvors exckratkon [@Hetasa:2010].
Scisdora modes of a trapped Vose-Einstein condenxaev {#bose-einsteih}
----------------------------------------------------
By nor, ehere exists a vast literature on trapptd Boae-Einstein condensates [@Pitaevskii:2003; @Giorgini:2008]. Soperfluidiey in these condensates is one of the most spectawular zonwezjwnfes. It is, however, not easy to obtain unambigujhs enidence for the smperfluid charactetidtoss. Because sukerfluibjtg will affect the loment jf inwrtia of uhe ttapped condensate (such as a reduction oner rhe classical rigib value), one eould gxpect that a study of rotatiunal properties of such dundensates can gkve rhse to experimental evidense for thx exixtence uf soperfluydity. @Guerj-Odellt:1999 have studied the oscipldtory behagior caused by rotating a condeixate with rexpact to the fymmebry axis when twapped in a deyormed erternau potentiam of pacabolic type. They concentswte in partirular on ehe wupeefluid dwfects in the vondensatv. Uhe restorung force associatcd wijh such a rotatiou un the $xy$-plane os oro[ogtipnwn to $\delta^2$, whth ghe yrappkng pottuticl givrn by the expression $$V_{exf}({\bf r})=\frac{m}{2}\omega^2_x x^2 +
\fvac{m}{2}\omega^2_i y^2 + \frac{i}{2}\omega^2_z z^2,$$ witn moreover $$\omega_x^2=\olega_0^2(1+\dxlta),\omxga_y^2=\omrga_0^2(1-\qelta).$$ The mass parameter, deterjined by hhe moment of igertla, ig the supexfluid case becomes proportional to $\delte^2$ too [@Lipparini:2003; @Rowe:1970]. As a result, even when uhe deformatipn of the extesnal potential approqches zero, the frtquency keeps a finite balue. Ht | Other ways of observing scissors modes in from recent experiment studied magnetic properties one crystals in which internal electrostatic field small with respect to the electron coupling in the atoms, the so-called “spin-orbit locking” situation, an applied external magnetic will rotate both the spin and charge density profiles simultaneously. Switching off the field, atoms start around the axes of the crystal cells [@Hatada:2009a]. Experiments are proposed that may be sensitive enough detect the photons emitted when deexciting the scissors [@Hatada:2010]. Scissors modes of trapped Bose-Einstein condensate {#bose-einstein} ---------------------------------------------------- now, exists a literature trapped condensates [@Pitaevskii:2003; @Giorgini:2008]. in these condensates is one of the most spectacular consequences. It is, however, not easy to obtain evidence for characteristics. Because will the of inertia of condensate (such as a reduction over value), one could expect that a study of properties of condensates can give rise to experimental for the existence of superfluidity. @Guery-Odelin:1999 have studied oscillatory behavior caused by rotating a condensate with respect to the symmetry axis when trapped deformed external potential of type. They concentrate particular the effects the condensate. restoring force associated with such a rotation in the $xy$-plane is to $\delta^2$, with the trapping potential given by the expression x^2 \frac{m}{2}\omega^2_y y^2 + z^2,$$ with moreover $$\omega_x^2=\omega_0^2(1+\delta),\omega_y^2=\omega_0^2(1-\delta).$$ mass determined by the moment in superfluid to too @Rowe:1970]. As a result, when the deformation of the potential approaches zero, the It | Other ways of observing scissOrs modes in CrystAls StaRt From A recEnt experiment tHAt stUdied magnetic propertieS of raRe-EArth SYsTems [@vAnderLaAN:2008]. IF ONe cOnSiDerS cRYsTals iN whIch the iNternal eleCtrOsTatic field is SMaLl with respEct To the electroN spIn-orbiT cOupLIng in The Atoms, The so-cALled “spIn-orbit loCkINg” situATion, an aPPLiEd exTernal magnetic fieLD wILl rotate both thE spin aNd CHaRGE deNsiTy profiles SiMultaNEously. SWItCHINg oFF the magnetic fIeld, the atomS WilL start OsCilLAting aRound ThE AxeS of the crystAl ceLls [@Hatada:2009A]. ExperIMents arE ProposeD that mAy bE seNsitIVe EnOugH tO DetECt The PHotOns emittEd WhEn deeXcitING THe scIssOrs eXcitaTion [@Hatada:2010].
SciSsoRs moDEs oF a traPped BOse-EInStein CondenSate {#bOsE-einstein}
----------------------------------------------------
By now, tHere Exists a vaSt lItEraTuRe on tRApped BOse-einStein coNdensatES [@PiTaEVSKiI:2003; @Giorgini:2008]. SuperfluiDiTY In These conDensatES iS oNE of the moSt SpeCtacULAr conSequENcEs. It is, hoWever, nOT eAsY to obtaIn UnambiGuOus EviDence FOr thE superFluid chaRacteRIstics. Because sUPerfluidity wiLL aFFEcT The mOmeNt of inertia Of thE TrapPed cONdEnsATe (sucH as a rEdUCtIOn over the classical rIgId valuE), one cOuld expect thaT a study of rOTATional prOperTIeS Of such condensaTes caN give rise tO ExperimeNtal eVidence fOr the exisTENce of supErfLuiDitY. @GuERY-ODelin:1999 have studIED the OsCillatoRy bEhavior CauSed By rOtaTiNg a condenSate with ReSpEcT tO thE symmETry axis wHeN trApPed In a deFOrmed eXternAl poTeNtIAl oF paraboLIc TYPe. ThEy CoNcenTraTe In parTicuLAr oN the supErfluid efFecTS in tHe CoNdensatE. The restoring FoRce associaTeD wiTh such A ROtation iN the $xy$-plane is proportionAL to $\deltA^2$, wiTh the TrapPing potenTiaL given By tHE expreSsion $$V_{Ext}({\bf R})=\fRac{M}{2}\OMega^2_x X^2 +
\FRaC{m}{2}\oMeGa^2_y y^2 + \frac{m}{2}\oMEGa^2_z Z^2,$$ with MoReovEr $$\omega_X^2=\omega_0^2(1+\delta),\omega_y^2=\oMEga_0^2(1-\Delta).$$ The mass pAraMeteR, DEtErmINeD By tHe MOmeNT Of inertia, in the sUperfluid cAsE BeComes propoRTioNaL to $\deltA^2$ too [@LipParinI:2003; @rowe:1970]. As a Result, eveN when the dEfOrmaTIOn oF the externAl potentIal approaCHes zeRO, tHe freQueNcy keePs A fiNite vAlue. It | Other ways of observing sc issors mod es in cr yst al s st artfrom a recente xper iment that studied mag netic p r oper t ie s ofrare-ea r th s yst em s[@v an d er Laan: 200 8]. Ifone consid ers c rystals in w h ic h the inte rna l electrosta tic field i s s m all w ith resp ect to the el ectron sp in - orbitc oupling i ntheatoms, the so-cal l ed “spin-orbit lo cking” s i tu a t ion , a n appliedex terna l magnet i cf i e ldw ill rotate bo th the spin and charg eden s ity pr ofile ss imu ltaneously. Swi tching of f them agnetic field,the at oms wi ll s t ar tosc il l ati n garo u ndthe axes o fthe c ryst a l c ells [@ Hata da:20 09a]. Experim ent s ar e pr opose d tha t ma ybe se nsitiv e eno ug h to detect the pho tons emit ted w hen d eexci t ing th e s cis sors ex citatio n [@ Ha t a d a: 2010].
Scissors m od e s o f a trap ped Bo s e- Ei n stein co nd ens ate{ # bose- eins t ei n}
----- ------ - -- -- ------- -- ------ -- --- --- ----- - ---- ---
B y now, t heree xists a vast l i terature on t r ap p e dB ose- Ein stein conde nsat e s [@ Pita e vs kii : 2003; @Gio rg i ni : 2008]. Superfluidit yin the se co ndensates isone of the m o st spect acul a rc onsequences. I t is, however,n ot easyto ob tain una mbiguouse v idence f orthe su per f l ui d characteris t i cs.Be cause s upe rfluidi tywil l a ffe ct the mome nt of in er ti aof th e tra p ped cond en sat e(su ch as a redu ction ove rth e cl assical ri g i d va lu e) , on e c ou ld ex pect tha t a stu dy of rot ati o nalpr op ertiesof such conde ns ates can g iv e r ise to e xperimen tal evidence for the ex i stenceofsuper flui dity. @Gu ery -Odeli n:1 9 99 hav e stud ied t he os c i llato r y b eha vi or causedb y ro tatin ga co ndensat e with respect tot hesymmetry axis wh en t r a pp edi na de fo r med e xternal potenti al of para bo l ic type. The y co nc entrate in par ticul a r on th e superfl uid effec ts int h e c ondensate. The res toring fo r ce as s oc iated wi th suc ha r otati on int he$xy$- planeis propo rtion al to $\de lta^2$, with the trappi ng pot entia l g iven by t hee xpr ession $$ V_{e xt}({\bf r })= \fr ac{m} {2} \ omega ^2_x x^ 2 + \frac {m}{ 2 }\omega^2 _ yy^2 + \ frac{m}{2}\ o m e ga^ 2_z z ^2, $ $ with mor eover $$\omega_x^ 2 =\omega_0^2(1+ \del t a ),\ ome g a_y^ 2= \omega_0^2(1-\ del ta ) . $$ The m as s parameter , determ in e d bythe mo ment o f inert i a ,i n thesupe rfl uid casebec om e s propo rt io n al to$\de lt a^2$ t oo [@L i ppar i n i:2003; @Rowe:19 70].A s a re s ult , eve nwhen th e def ormation o f the exter nal po tent ial a pproach es zero, th efrequencyk eeps a fi nitevalue.It | Other ways_of observing_scissors modes in crystals_start from_a_recent experiment_that_studied magnetic properties_of rare-earth systems_[@vanderLaan:2008]. If one considers_crystals in which_the_internal electrostatic field is small with respect to the electron spin-orbit coupling in the_atoms,_the so-called_“spin-orbit_locking”_situation, an applied external magnetic_field will rotate both the_spin and_charge density profiles simultaneously. Switching off the magnetic_field,_the atoms will_start oscillating around the axes of the crystal cells_[@Hatada:2009a]. Experiments are proposed that may_be sensitive enough_to_detect_the photons emitted when_deexciting the scissors excitation [@Hatada:2010].
Scissors modes_of a trapped Bose-Einstein condensate {#bose-einstein}
----------------------------------------------------
By_now, there exists a vast literature on_trapped Bose-Einstein condensates [@Pitaevskii:2003; @Giorgini:2008]. Superfluidity_in these condensates is one_of the_most spectacular consequences. It is,_however, not easy_to obtain_unambiguous evidence for_the superfluid characteristics. Because superfluidity will_affect the moment_of inertia of the trapped condensate_(such_as a reduction_over_the_classical rigid_value), one could_expect_that a_study_of rotational properties of such condensates_can_give rise to experimental evidence for the_existence of superfluidity. @Guery-Odelin:1999_have_studied the oscillatory behavior_caused by rotating a condensate_with respect to the symmetry axis_when trapped_in a_deformed external potential of parabolic type. They concentrate in particular on_the superfluid effects in the condensate._The restoring force associated_with such_a_rotation in the_$xy$-plane_is proportional_to $\delta^2$, with the trapping potential given_by the_expression $$V_{ext}({\bf r})=\frac{m}{2}\omega^2_x x^2 +
\frac{m}{2}\omega^2_y y^2_+ \frac{m}{2}\omega^2_z z^2,$$ with_moreover_$$\omega_x^2=\omega_0^2(1+\delta),\omega_y^2=\omega_0^2(1-\delta).$$ The mass parameter, determined by_the moment of inertia, in the_superfluid case becomes proportional to_$\delta^2$_too_[@Lipparini:2003; @Rowe:1970]. As a result,_even when the deformation of the_external potential approaches_zero, the frequency keeps a finite value._It |
s is limited by the background pressure in the UHV chamber. The lifetime suggests a pressure of around 10$^{-10}$ Torr and is limited by the gas load from the Rb dispensers and pumping speed from the 40 l/s Varian diode ion pump, not the leak rate of the viewports.
![Measurement of the number of Rb atoms in a CO$_2$ laser optical dipole trap for increasing hold time. A 1/[e]{} lifetime of 9.5 $\pm$ 0.5 s is deduced from an exponential fit to the data.](fig2.pdf){width="8.0cm"}
The specialized requirement in the design presented here is in the manufacturing process. Non-standard vacuum components are used including, in the present design, the mounting flange welded onto a standard, albeit shortened, conflat half nipple. The design is also bulkier than many viewports which presents a disadvantage for optical experiments requiring high numerical aperture (NA). To address these issues we have also investigated two simplified designs. First of all we note that the key to the robustness of the design presented so far in comparison with our previous design [@cox2003] is the use of a length of tube between the conflat flange, that attaches to the rest of the vacuum system, and the optic to metal seal. As this seal is typically only about 2 mm wide we implemented the design shown in Figure 3, where we have discarded the mounting flange and sealed directly on the end of the tube. For a standard 70CF half nipple the tube has a wall thickness of 1.5 mm. Prior to sealing, the end of the pipe was lapped to ensure it was completely flat. The sealing process follows the same principles as outlined above. In this case the clamping flange was tightened onto the conflat flange with 12 M4 (0.7 mm pitch) bolts with spring washers. A torque of about 1.2 Nm was sufficient to ensure a good seal. Figure 3 shows the design of this window including the specially designed clamping flange. This viewport design has been leak tested to $<10^{-10}$ atm cm$^3$/s and we have successfully operated a vacuum system with eight windows of this design at a pressure below $10^{-8}$ Torr for an extended period of time.
In order to increase the robustness of this design we have investigated the use of an ultra-high vacuum compatible epoxy to improve the glass-to-metal seal. | s is limited by the background pressure in the UHV chamber. The life suggest a pressure of around 10$^{-10}$ Torr and is limited by the accelerator burden from the Rb dispensers and pumping amphetamine from the 40 fifty / s Varian diode ion pump, not the leak rate of the viewports.
! [ Measurement of the number of Rb atom in a CO$_2 $ laser optical dipole trap for increase detention time. A 1/[e ] { } life of 9.5 $ \pm$ 0.5 s is deduce from an exponential paroxysm to the data.](fig2.pdf){width="8.0 cm " }
The specialized requirement in the design presented here is in the fabrication process. Non - standard vacuum component are used include, in the present design, the mounting flange welded onto a standard, albeit shortened, conflat one-half nipple. The design is also bulkier than many viewports which presents a disadvantage for optical experiments command high numerical aperture (NA). To address these issues we have also investigated two simplified designs. First of all we note that the key to the robustness of the design introduce so far in comparison with our previous invention [ @cox2003 ] is the use of a length of tube between the conflat flange, that attaches to the remainder of the vacuum system, and the optic to metal cachet. As this seal is typically only about 2 mm wide we implemented the purpose shown in Figure 3, where we have discarded the mounting flange and sealed immediately on the end of the metro. For a standard 70CF one-half nipple the pipe has a wall thickness of 1.5 mm. Prior to sealing, the goal of the pipe was lapped to ensure it was wholly flat. The waterproofing process follows the same principles as outlined above. In this sheath the clamping flange was tightened onto the conflat flange with 12 M4 (0.7 mm sales talk) bolts with give washers. A torsion of about 1.2 Nm was sufficient to ensure a beneficial seal. Figure 3 show the design of this windowpane including the specially designed clamp flange. This viewport design has been leak tested to $ < 10^{-10}$ atm cm$^3$/s and we have successfully operated a void system with eight windows of this blueprint at a pressure below $ 10^{-8}$ Torr for an extended period of time.
In order to increase the robustness of this design we have investigated the manipulation of an ultra - high vacuum compatible epoxy to improve the glass - to - metal sealing wax. | s id limited by the backgromnd pressure in jhw UHV rhamber. The lifdtime suggests a pressure of aeound 10$^{-10}$ Torr and is limited bh the gas load frim tie Rb dispensers and puminng slced fxon the 40 l/s Varisn diode imn pump, not tha ueck rate of the viewports.
![Measurement jf the mulber of Rb atois im a CK$_2$ laser optical dipole trap for indreasinj hold time. A 1/[e]{} lifetime of 9.5 $\pm$ 0.5 s is dedufed vrom an exponentiap fit to thg daeq.](fig2.pdf){width="8.0zm"}
The specpclized requjrement in the design presented here is in the mqnuvdcturing pricess. Non-standard vacuum womponemts are used ikcludmng, un the present design, the mounting flande welded outo a standard, albeit syorteted, wonfuqt falr iiplle. Thf dxsign is alao bulkier rhan many viewports wrpvh presents z disaqvwntage for optical experiments requirinc hjgh numerical aperture (BA). To address these idsues we rave also investigated two simplified designs. Firvt of xll wc nogw hhat the key to the robustness of the design [dexekted so far in cjmparison wotj por previous derign [@cox2003] js the use of a lejgth of tube between uhe cpnflat flange, that attaches to the rest if the vacuum systzm, and the o'tic tp metsl seal. As this seal is typjcally only about 2 mj wide we implemevtec dhe design shown in Figure 3, where we havz discaraed jhe mougting flanhe and sealed directly oj the etd of the hube. For a standard 70CF half nip'ke the tube nav a wall thncknesx of 1.5 mm. Priow to sealing, tke end oy the oipe was lzpped tm ensure it was completeni flat. The seeling prosess foloows thd same principlrs as outlined abovw. In this case the clxjping flange waw tughtened onto tne zonslwt fjdnge with 12 M4 (0.7 mm oitzn) bolgs with sprlng wasners. A torque of abogt 1.2 Nj was sufficient tp cnsure a tood seaj. Figure 3 shoes the design of tjis wmndow mncludong the specially designed clampihg flange. Thls viewport dqsigk haf been leal tested to $<10^{-10}$ atm cm$^3$/s and we have succeswfully operated a cacuum system with gigmt windows oh this design ad a pressure below $10^{-8}$ Rorr for an extenced period of time.
In krder do infrease the robustness of this design we have investigated the use of an uotra-hijh vacuum comlatinle eioxv tp imprjve vhx glass-to-metal sesl. | s is limited by the background pressure UHV The lifetime a pressure of limited the gas load the Rb dispensers pumping speed from the 40 l/s diode ion pump, not the leak rate of the viewports. {width="8.0cm"} The specialized requirement in the presented here is in the manufacturing process. Non-standard components are used including, the present design, the mounting welded a standard, shortened, half The design is bulkier than many viewports which presents a disadvantage for optical experiments requiring high numerical aperture (NA). To these issues also investigated simplified First all we note key to the robustness of the far in comparison with our previous design [@cox2003] the use a length of tube between the flange, that attaches to the rest of the system, and the optic to metal seal. As this seal is typically only about 2 we implemented the design in Figure 3, we discarded mounting and sealed on the end of the tube. For a standard 70CF half the tube has a wall thickness of 1.5 mm. Prior the of the pipe lapped to ensure it completely The sealing process follows principles outlined case clamping was tightened onto the flange with 12 M4 (0.7 pitch) bolts with spring 1.2 Nm was sufficient to ensure a good Figure 3 shows the design of this including the specially designed clamping flange. This viewport design has been leak to $<10^{-10}$ and we have successfully operated a vacuum system eight windows of this at a pressure below $10^{-8}$ Torr for an extended of In order increase the robustness this design we investigated the use ultra-high vacuum epoxy improve | s is limited by the background Pressure in The UHv chAmbEr. the lIfetIme suggests a prESsurE of around 10$^{-10}$ Torr and is limiTed by ThE Gas lOAd From tHe Rb disPEnSERs aNd PuMpiNg SPeEd froM thE 40 l/s VariAn diode ion PumP, nOt the leak ratE Of The viewporTs.
![MEasurement of The Number Of rb aTOms in A CO$_2$ Laser OpticaL Dipole Trap for inCrEAsing hOLd time. A 1/[E]{} LIfEtimE of 9.5 $\pm$ 0.5 s is deduced frOM aN Exponential fit To the dAtA.](FiG2.PDf){wIdtH="8.0cm"}
The specIaLized REquiremENt IN THe dESign presented Here is in the MAnuFacturInG prOCess. NoN-stanDaRD vaCuum componeNts aRe used incLuding, IN the preSEnt desiGn, the mOunTinG flaNGe WeLdeD oNTo a STaNdaRD, alBeit shorTeNeD, confLat hALF NIpplE. ThE desIgn is Also bulkier thAn mAny vIEwpOrts wHich pReseNtS a disAdvantAge foR oPtical experimenTs reQuiring hiGh nUmEriCaL aperTUre (NA). TO adDreSs these Issues wE HavE aLSO InVestigated two simplIfIED dEsigns. FiRst of aLL wE nOTe that thE kEy tO the ROBustnEss oF ThE design pResentED sO fAr in comPaRison wItH ouR prEviouS DesiGn [@cox2003] iS the use oF a lenGTh of tube betweeN The conflat flaNGe, THAt ATtacHes To the rest of The vACuum SystEM, aNd tHE optiC to meTaL SeAL. As this seal is typicaLlY only aBout 2 mM wide we implemEnted the deSIGN shown in figuRE 3, wHEre we have discaRded tHe mounting FLange and SealeD directlY on the end OF The tube. FOr a StaNdaRd 70Cf HAlF nipple the tubE HAs a wAlL thicknEss Of 1.5 mm. PriOr tO seAliNg, tHe End of the pIpe was laPpEd To EnSurE it waS CompleteLy FlaT. THe sEalinG ProcesS follOws tHe SaME prIncipleS As OUTlinEd AbOve. IN thIs Case tHe clAMpiNg flangE was tightEneD Onto ThE cOnflat fLange with 12 M4 (0.7 mm pItCh) bolts witH sPriNg washERS. A torque Of about 1.2 Nm was sufficient tO Ensure a GooD seal. figuRe 3 shows thE deSign of ThiS Window IncludIng thE sPecIALly deSIGnEd cLaMping flangE. tHis ViewpOrT desIgn has bEen leak tested to $<10^{-10}$ atm CM$^3$/s aNd we have succeSsfUlly OPErAteD A vACuuM sYSteM WIth eight windows Of this desiGn AT a Pressure beLOw $10^{-8}$ TOrR for an eXtended PerioD Of time.
IN order to iNcrease thE rObusTNEss Of this desiGn we have InvestigaTEd the USe Of an uLtrA-high vAcUum CompaTible ePOxy To impRove thE gLass-to-Metal SeAl. | s is limited by the backgr ound press ure i n t heUH V ch ambe r. The lifetim e sug gests a pressure of ar ound10 $ ^{-1 0 }$ Torr and is li m i ted b ythe g a sloadfro m the R b dispense rsan d pumping sp e ed from the40l/s Varian d iod e ionpu mp, not t heleakrate o f the v iewports.
! [Measu r ement o f th e nu mber of Rb atomsi na CO$_2$ laseroptica ld ip o l e t rap for incre as ing h o ld time . A 1 / [e] { } lifetime of 9.5 $\pm$0 .5s is d ed uce d froman ex po n ent ial fit tothedata.](fi g2.pdf ) {width= " 8.0cm"}
Thespe cia lize d r eq uir em e nti nthe des ign pres en te d her e is i n theman ufac turin g process. No n-s tand a rdvacuu m com pone nt s are usedinclu di ng, in the pres entdesign, t hemo unt in g fla n ge wel ded on to a st andard, alb ei t s ho rtened, conflat ha lf n ip ple. The desig n i sa lso bulk ie r t hanm a ny vi ewpo r ts which p resent s a d isadvan ta ge for o pti cal expe r imen ts req uiring h igh n u merical apertu r e (NA). To ad d re s s t h eseiss ues we have als o inv esti g at edt wo si mplif ie d d e signs. First of all w e note that the key to t he robustn e s s of thedesi g np resented so fa r incomparison with our prev ious des ign [@cox 2 0 03] is t heuse of al e ng th of tube be t w eenth e confl atflange, th atatt ach es to the r est of t he v ac uu m s ystem , and the o pti ctometal seal.As th is s ea li s t ypicall y o n l y ab ou t2 mm wi de we i mple m ent ed thedesign sh own in F ig ur e 3, wh ere we have d is carded the m oun ting f l a nge andsealed directly on thee nd of t hetube. For a standa rd70CF h alf nipple the t ube h as aw a ll th i c kn ess o f 1.5 mm.P r ior to s ea ling , the e nd of the pipe was lap ped to ensure it was c om ple t el y fl at . Th e sealing process follows t he sa me princip l esas outlin ed abov e. In this ca se the cl amping fl an ge w a s ti ghtened on to the c onflat fl a nge w i th 12 M 4 ( 0.7 mm p itc h) bo lts wi t h s pring washe rs . A to rqueof about 1 .2 Nm was sufficient to ensur e a g ood seal. Fi gur e 3shows the des ign of thi s w ind ow in clu d ing t he s p ec ial l y des igne d clamping fl ang e . T his viewpor t d esi gn ha s b e en lea k te sted to $<10^{-10 } $ atm cm$^3$/s and w e h ave succ es sfully operate d a v a c uum syst em with eight windows o f this desig n at a pressu r e b e low $1 0^{- 8}$ Torr for an e x tendedpe ri o d of t ime.
In ord er toi ncre a s e the robustness of t h i s des i gnwe ha ve invest i gate d the useof an ultra -highvacu um co mpatibl eepoxytoim prove theg lass-to-m etalseal. | s is_limited by_the background pressure in_the UHV_chamber._The lifetime_suggests_a pressure of_around 10$^{-10}$ Torr and_is limited by the_gas load from_the_Rb dispensers and pumping speed from the 40 l/s Varian diode ion pump, not_the_leak rate_of_the_viewports.
![Measurement of the number of_Rb atoms in a CO$_2$_laser optical_dipole trap for increasing hold time. A 1/[e]{}_lifetime_of 9.5 $\pm$_0.5 s is deduced from an exponential fit to the_data.](fig2.pdf){width="8.0cm"}
The specialized requirement in the design_presented here is_in_the_manufacturing process. Non-standard vacuum_components are used including, in the_present design, the mounting flange welded_onto a standard, albeit shortened, conflat half_nipple. The design is also bulkier_than many viewports which presents_a disadvantage_for optical experiments requiring high_numerical aperture (NA)._To address_these issues we_have also investigated two simplified designs._First of all_we note that the key to_the_robustness of the_design_presented_so far_in comparison with_our_previous design_[@cox2003]_is the use of a length_of_tube between the conflat flange, that attaches_to the rest of_the_vacuum system, and the_optic to metal seal. As_this seal is typically only about_2 mm_wide we_implemented the design shown in Figure 3, where we have discarded the_mounting flange and sealed directly on_the end of the_tube. For_a_standard 70CF half_nipple_the tube_has a wall thickness of 1.5 mm._Prior to_sealing, the end of the pipe_was lapped to ensure_it_was completely flat. The sealing process_follows the same principles as outlined_above. In this case the_clamping_flange_was tightened onto the conflat_flange with 12 M4 (0.7 mm pitch)_bolts with spring_washers. A torque of about 1.2 Nm was_sufficient_to ensure a good seal. Figure_3_shows the design of this window_including_the_specially designed clamping flange. This_viewport design has been leak tested_to $<10^{-10}$ atm cm$^3$/s and we have successfully operated_a vacuum system_with eight windows of this_design_at_a pressure below $10^{-8}$ Torr for an extended period of_time.
In order_to increase the_robustness of this design we have investigated the use of_an ultra-high vacuum compatible epoxy to improve_the glass-to-metal seal. |
} } \sum_{q}
\frac{\partial m_N }{\partial m_q^{\mbox{\scriptsize renorm} } }
\frac{\partial
m_q^{\mbox{\scriptsize renorm} } }{\partial m_s^{\mbox{\scriptsize bare} } }
\label{scalmix} \\ & \approx &
\left. m_s \langle N |\bar{s} s| N\rangle \right|_{\mbox{\scriptsize renorm} }
+ m_s^{\mbox{\scriptsize renorm} }
\frac{\partial m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} } }
\langle N |\bar{u} u + \bar{d} d | N \rangle^{\mbox{\scriptsize renorm} }
\nonumber\end{aligned}$$ where $m_{res,l} $ denotes the light quark flavor residual mass, and where only the dominant correction has been kept in the second line by using $m_{s}^{\mbox{\scriptsize bare} } \gg m_{res,s} $ and $\langle N |\bar{u} u + \bar{d} d | N\rangle \gg
\langle N |\bar{s} s | N\rangle $. No direct data for the factor $\partial m_{res,l} /\partial m_s^{\mbox{\scriptsize bare} } $ are available, but a rough order of magnitude estimate can be constructed from related data on the residual mass obtained within the LHPC program and in the present work. This estimate is discussed in the Appendix, with the result $$\left.
\frac{\partial m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} } }
\right|_{am_s =0.081, am_l =0.0081} \approx -0.00035$$ cf. (\[simpest\]), at the lowest light quark mass considered in the numerical calculations in this work. In turn, an upper bound for the light quark scalar matrix element $\langle N |\bar{u} u + \bar{d} d | N \rangle^{\mbox{\scriptsize renorm} } $ is given by its magnitude at the physical point; combining a typical phenomenological value for the nucleon sigma term, $m_l \langle N |\bar{u} u + \bar{d} d | N \rangle \approx 60\, \mbox{MeV} $, cf | } } \sum_{q }
\frac{\partial m_N } { \partial m_q^{\mbox{\scriptsize renorm } } }
\frac{\partial
m_q^{\mbox{\scriptsize renorm } } } { \partial m_s^{\mbox{\scriptsize bare } } }
\label{scalmix } \\ & \approx &
\left. m_s \langle N |\bar{s } s| N\rangle \right|_{\mbox{\scriptsize renorm } }
+ m_s^{\mbox{\scriptsize renorm } }
\frac{\partial m_{res, l } } { \partial m_s^{\mbox{\scriptsize bare } } }
\langle N |\bar{u } u + \bar{d } d | N \rangle^{\mbox{\scriptsize renorm } }
\nonumber\end{aligned}$$ where $ m_{res, l } $ announce the faint quark flavor residual mass, and where only the prevailing correction has been kept in the second line by practice $ m_{s}^{\mbox{\scriptsize bare } } \gg m_{res, s } $ and $ \langle N |\bar{u } u + \bar{d } vitamin d | N\rangle \gg
\langle N |\bar{s } s | N\rangle $. No direct data for the gene $ \partial m_{res, l } /\partial m_s^{\mbox{\scriptsize bare } } $ are available, but a boisterous order of magnitude estimate can be constructed from related data on the residual mass obtain within the LHPC program and in the present work. This estimate is hash out in the Appendix, with the result $ $ \left.
\frac{\partial m_{res, l } } { \partial m_s^{\mbox{\scriptsize bare } } }
\right|_{am_s = 0.081, am_l = 0.0081 } \approx -0.00035$$ cf. (\[simpest\ ]), at the lowest unaccented quark mass considered in the numerical calculations in this work. In turning, an upper bound for the light quark scalar matrix element $ \langle N |\bar{u } u + \bar{d } d | N \rangle^{\mbox{\scriptsize renorm } } $ is move over by its magnitude at the physical point; combining a typical phenomenological value for the nucleon sigma term, $ m_l \langle N |\bar{u } u + \bar{d } d | N \rangle \approx 60\, \mbox{MeV } $, cf | } } \skm_{q}
\frac{\partial m_N }{\partiau m_q^{\mbox{\scriptsnze renmrm} } }
\fdac{\partixl
m_q^{\mbox{\scriptsize renorm} } }{\pertiql m_s^{\nbox{\scriptsize bare} } }
\lxbel{scalmpx} \\ & \apprix &
\ltft. m_s \langle N |\bed{s} s| N\rangle \dlght|_{\myoe{\scriptsize renprm} }
+ m_s^{\mbof{\scriptsize retofm} }
\frac{\partial m_{res,l} }{\partial m_s^{\mbox{\scwiptsizr hare} } }
\langle N |\bar{l} t + \bzg{d} d | N \rangle^{\mbox{\scriptsize rehorm} }
\noiumber\end{alignec}$$ where $m_{res,l} $ denotes the ligjt quark flavor redidual mass, and qhere only tfe dominanu eorrection gas been kept in the second lind by bsing $m_{s}^{\mboz{\sxriojsize bare} } \jg m_{ref,s} $ and $\langle N |\bar{g} u + \bat{d} d | N\rangle \ng
\lanjle B |\bar{s} s | N\rangle $. No direct data for tre factor $\pcrtial m_{res,l} /\partial n_s^{\nbox{\swripdsizd bafe} } $ ere availwblx, but a roufh order of magnitude estimate cwb be construcfed frjm related data on the residual mass obtapned within the LHPC progran and in the present aork. This estimate is discussed in the Appendix, with the rasult $$\uefu.
\fvwz{\pagtial m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} } }
\ridgt|_{sm_x =0.081, am_l =0.0081} \approx -0.00035$$ cf. (\[simpest\]), ah yre lowest ligft quaxi jass considered in the nuiericql calculwtioms in this work. In turn, an ypper bound yor the light quark sealar matrix elekent $\kangle N |\bar{u} u + \bar{d} d | N \dangle^{\mbox{\sfriptsize fenorm} } $ is givev bj itv magnitude at the physicaj point; cimbiuing a thpicsl phegomenologifal value for the nucleoj sigla term, $m_l \lwngle N |\bar{u} u + \bar{d} d | N \ranglx \approx 60\, \mboc{MaV} $, cf | } } \sum_{q} \frac{\partial m_N }{\partial m_q^{\mbox{\scriptsize } m_q^{\mbox{\scriptsize renorm} }{\partial m_s^{\mbox{\scriptsize bare} \approx \left. m_s \langle |\bar{s} s| N\rangle renorm} } + m_s^{\mbox{\scriptsize renorm} } m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} } } \langle N |\bar{u} u + \bar{d} d N \rangle^{\mbox{\scriptsize renorm} } \nonumber\end{aligned}$$ where $m_{res,l} $ denotes the light quark flavor mass, where the correction has been kept in the second line by using $m_{s}^{\mbox{\scriptsize bare} } \gg m_{res,s} $ $\langle N |\bar{u} u + \bar{d} d | \gg \langle N |\bar{s} | N\rangle $. No direct for factor $\partial /\partial bare} $ are available, a rough order of magnitude estimate can be constructed from related data on the residual mass obtained the LHPC in the work. estimate discussed in the the result $$\left. \frac{\partial m_{res,l} }{\partial } \right|_{am_s =0.081, am_l =0.0081} \approx -0.00035$$ cf. at the light quark mass considered in the calculations in this work. In turn, an upper for the light quark scalar matrix element $\langle N |\bar{u} u + \bar{d} d | renorm} } $ is by its magnitude the point; a phenomenological value the nucleon sigma term, $m_l \langle N |\bar{u} u + \bar{d} | N \rangle \approx 60\, \mbox{MeV} $, cf | } } \sum_{q}
\frac{\partial m_N }{\partial m_Q^{\mbox{\scripTsize RenOrm} } }
\FrAc{\paRtiaL
m_q^{\mbox{\scriptsIZe reNorm} } }{\partial m_s^{\mbox{\scripTsize BaRE} } }
\labEL{sCalmiX} \\ & \approx &
\LEfT. M_S \laNgLe n |\baR{s} S| n\rAngle \RigHt|_{\mbox{\sCriptsize rEnoRm} }
+ M_s^{\mbox{\scriptSIzE renorm} }
\fraC{\paRtial m_{res,l} }{\paRtiAl m_s^{\mbOx{\ScrIPtsizE baRe} } }
\lanGle N |\baR{U} u + \bar{d} D | N \rangle^{\mBoX{\ScriptSIze renoRM} }
\NoNumbEr\end{aligned}$$ where $M_{ReS,L} $ denotes the ligHt quarK fLAvOR ResIduAl mass, and wHeRe onlY The domiNAnT CORreCTion has been kePt in the secoND liNe by usInG $m_{s}^{\MBox{\scrIptsiZe BAre} } \Gg m_{res,s} $ and $\lAnglE N |\bar{u} u + \baR{d} d | N\raNGle \gg
\laNGle N |\bar{S} s | N\ranGle $. no dIrecT DaTa For ThE FacTOr $\ParTIal M_{res,l} /\parTiAl M_s^{\mboX{\scrIPTSIze bAre} } $ Are aVailaBle, but a rough oRdeR of mAGniTude eStimaTe caN bE consTructeD from ReLated data on the rEsidUal mass obTaiNeD wiThIn the lhPC proGraM anD in the pResent wORk. THiS ESTiMate is discussed in tHe aPPeNdix, with The resULt $$\LeFT.
\frac{\parTiAl m_{Res,l} }{\PARtial M_s^{\mbOX{\sCriptsizE bare} } }
\rIGhT|_{aM_s =0.081, am_l =0.0081} \apPrOx -0.00035$$ cf. (\[siMpEst\]), At tHe lowESt liGht quaRk mass coNsideREd in the numericAL calculations IN tHIS wORk. In TurN, an upper bouNd foR The lIght QUaRk sCAlar mAtrix ElEMeNT $\langle N |\bar{u} u + \bar{d} d | N \RaNgle^{\mbOx{\scrIptsize renorm} } $ Is given by iTS MAgnitude At thE PhYSical point; combIning A typical phENomenoloGical Value for The nucleoN SIgma term, $M_l \lAngLe N |\Bar{U} U + \BaR{d} d | N \rangle \appROX 60\, \mboX{MEV} $, cf | } } \sum_{q}
\frac{\partia l m_N }{\p artia l m _q^ {\ mbox {\sc riptsize renor m } }}
\frac{\partial
m_q^{ \mbox {\ s crip t si ze re norm} } }{ \ p art ia lm_s ^{ \ mb ox{\s cri ptsizebare} } }\la be l{scalmix} \ \ & \approx &
\l eft. m_s \la ngl e N |\ ba r{s } s| N \ra ngle\right | _{\mbo x{\script si z e reno r m} }
+m _ s^ {\mb ox{\scriptsize re n or m } }
\frac{\par tial m _{ r es , l } } {\p artial m_s ^{ \mbox { \script s iz e b are } } }
\langleN |\bar{u}u +\bar{d }d | N \ran gle^{ \m b ox{ \scriptsize ren orm} }
\n onumbe r \end{al i gned}$$ where $m _{r es,l } $ d eno te s th e l igh t qu ark flav or r esidu al m a s s , and wh ereonlythe dominantcor rect i onhas b een k eptin thesecond line b y using $m_{s}^ {\mb ox{\scrip tsi ze ba re } } \ g g m_{r es, s}$ and $ \langle N | \b a r { u} u + \bar{d} d | N \r a n gl e \gg
\l angleN | \b a r{s} s | N \ra ngle $ . Nodire c tdata for the f a ct or $\part ia l m_{r es ,l} /\ parti a l m_ s^{\mb ox{\scri ptsiz e bare} } $ are available, bu t a r ou g h or der of magnitu de e s tima te c a nbec onstr ucted f r om related data on the r esidua l mas s obtained wi thin the L H P C program and in the present wo rk. T his estima t e is dis cusse d in the Appendix , with the re sul t $ $\l e f t.
\frac{\parti a l m_{ re s,l} }{ \pa rtial m _s^ {\m box {\s cr iptsize b are} } }
\ ri gh t| _{a m_s = 0 .081, am _l =0 .0 081 } \ap p rox -0 .0003 5$$cf .( \[s impest\ ] ), a t th elo west li gh t qua rk m a ssconside red in th e n u meri ca lcalcula tions in this w ork. In tu rn , a n uppe r bound fo r the light quark scala r matrix el ement $\l angle N | \ba r{u} u +\ bar{d} d | N \ran gl e^{ \ m box{\ s c ri pts iz e renorm}} $ i s giv en byits mag nitude at the phys i cal point; combi nin g at y pi cal ph e nom en o log i c al value for th e nucleonsi g ma term, $m_ l \l an gle N | \bar{u} u +\ bar{d}d | N \ra ngle \app ro x 60 \ , \m box{MeV} $ , cf | } }_\sum_{q}
\frac{\partial m_N_}{\partial m_q^{\mbox{\scriptsize renorm} }_}
\frac{\partial
m_q^{\mbox{\scriptsize renorm}_}_}{\partial m_s^{\mbox{\scriptsize_bare}_} }
\label{scalmix} \\_& \approx &
\left._m_s \langle N |\bar{s}_s| N\rangle \right|_{\mbox{\scriptsize_renorm}_}
+ m_s^{\mbox{\scriptsize renorm} }
\frac{\partial m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} } }
\langle N |\bar{u} u +_\bar{d}_d |_N_\rangle^{\mbox{\scriptsize_renorm} }
\nonumber\end{aligned}$$ where $m_{res,l} $_denotes the light quark flavor_residual mass,_and where only the dominant correction has been_kept_in the second_line by using $m_{s}^{\mbox{\scriptsize bare} } \gg m_{res,s} $_and $\langle N |\bar{u} u +_\bar{d} d |_N\rangle_\gg
\langle_N |\bar{s} s |_N\rangle $. No direct data for_the factor $\partial m_{res,l} /\partial m_s^{\mbox{\scriptsize_bare} } $ are available, but a_rough order of magnitude estimate can_be constructed from related data_on the_residual mass obtained within the_LHPC program and_in the_present work. This_estimate is discussed in the Appendix,_with the result_$$\left.
\frac{\partial m_{res,l} }{\partial m_s^{\mbox{\scriptsize bare} }_}
\right|_{am_s_=0.081, am_l =0.0081}_\approx_-0.00035$$_cf. (\[simpest\]), at_the lowest light_quark_mass considered_in_the numerical calculations in this work._In_turn, an upper bound for the light_quark scalar matrix element_$\langle_N |\bar{u} u +_\bar{d} d | N \rangle^{\mbox{\scriptsize_renorm} } $ is given by_its magnitude_at the_physical point; combining a typical phenomenological value for the nucleon sigma_term, $m_l \langle N |\bar{u} u_+ \bar{d} d |_N \rangle_\approx_60\, \mbox{MeV} $,_cf |
\vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-\frac{a_n}{\left\Vert a\right\Vert_{N,\ell^2}}\right\}\right\Vert_{N,\ell^2} \leq \frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1-\kappa_N^2)^{\frac{1}{2}}} \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right).$$ Define $H_N=(1+\delta_N)^{-1} \left\Vert a\right\Vert_{N,\ell^2} h_N$, and then we have $H_N\in K_{\Theta}$ and $\left\Vert H_N\right\Vert_{2}\leq \left\Vert a\right\Vert_{N,\ell^2}$. Using the last estimate and adding and subtracting the quantity $\frac{a_n}{(1+\delta_N)}$ yields that: $$\begin{aligned}
\left\Vert \left\{H_N(\lambda_n)\left(\frac{1-\left\vert \lambda_n\right\vert^2}{1-\left\vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-a_n\right\}\right\Vert_{N,\ell^2} \leq & &
\\ \left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1+\delta_N)(1-\kappa_N^2)^{\frac{1}{2}}} \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right)+\delta_N\right)\left\Vert a\right\Vert_{N,\ell^2}.\end{aligned}$$ Note that the quantity: $$\left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\ | \vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-\frac{a_n}{\left\Vert a\right\Vert_{N,\ell^2}}\right\}\right\Vert_{N,\ell^2 } \leq \frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1-\kappa_N^2)^{\frac{1}{2 } } } \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right).$$ Define $ H_N=(1+\delta_N)^{-1 } \left\Vert a\right\Vert_{N,\ell^2 } h_N$, and then we have $ H_N\in K_{\Theta}$ and $ \left\Vert H_N\right\Vert_{2}\leq \left\Vert a\right\Vert_{N,\ell^2}$. Using the last appraisal and lend and subtracting the quantity $ \frac{a_n}{(1+\delta_N)}$ render that: $ $ \begin{aligned }
\left\Vert \left\{H_N(\lambda_n)\left(\frac{1-\left\vert \lambda_n\right\vert^2}{1-\left\vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-a_n\right\}\right\Vert_{N,\ell^2 } \leq & &
\\ \left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1+\delta_N)(1-\kappa_N^2)^{\frac{1}{2 } } } \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right)+\delta_N\right)\left\Vert a\right\Vert_{N,\ell^2}.\end{aligned}$$ Note that the measure: $ $ \left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\ | \verh \Theta(\lambda_n)\right\vert^2}\rlght)^{\frac{1}{2}}-\frac{a_n}{\leyr\Vert e\right\Vsrt_{N,\ell^2}}\rkght\}\right\Vert_{N,\ell^2} \leq \frac{\su'_{m\gew N}\ledt\vert\Theta(\lambda_m)\righg\vert}{(1-\kappw_N^2)^{\frac{1}{2}}} \lwft(1+\ltft\Vert \left\{g_N(\lamusa_n)\left(1-\left\vedb \lamyde_n\right\vert^2\righj)^{\frac{1}{2}}\right\}\richt\Vert_{N,\ell^2}\rigvt).$$ Dzfine $H_N=(1+\delta_N)^{-1} \left\Vert a\right\Vert_{N,\ejl^2} h_N$, amd then we have $R_N\in H_{\Thefa}$ and $\left\Vert H_N\right\Vert_{2}\leq \leff\Vert a\gight\Vert_{N,\ell^2}$. Usimg the last estimate and afdinh and subtracting hhe quantiti $\frwx{a_n}{(1+\delta_N)}$ yidlds that: $$\begin{aligned}
\meft\Vert \left\{H_N(\lambda_n)\left(\frac{1-\ldft\vext \lambda_n\rugyt\vftt^2}{1-\left\vert \Tieta(\laibda_n)\right\vevn^2}\right)^{\fsac{1}{2}}-a_n\ribht\}\right\Vert_{N,\ekl^2} \lwq & &
\\ \left(\frac{\sup_{m\gxq N}\left\vert\Theta(\lamfda_m)\right\eext}{(1+\delta_N)(1-\kappa_N^2)^{\frac{1}{2}}} \ledt(1+\oeft\Vgrt \laft\{g_V(\oamcda_h)\lxft(1-\meft\vegt \mambda_n\riggt\vert^2\right)^{\drac{1}{2}}\right\}\right\Vert_{N,\tll^2}\wpbht)+\delta_N\riggt)\left\Dewt a\right\Vert_{N,\ell^2}.\end{aligned}$$ Note that tve suantity: $$\left(\frac{\sup_{m\gew N}\left\vert\Theta(\lambdw_m)\right\ | \vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-\frac{a_n}{\left\Vert a\right\Vert_{N,\ell^2}}\right\}\right\Vert_{N,\ell^2} \leq \frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1-\kappa_N^2)^{\frac{1}{2}}} \left(1+\left\Vert Define \left\Vert a\right\Vert_{N,\ell^2} and then we H_N\right\Vert_{2}\leq a\right\Vert_{N,\ell^2}$. Using the estimate and adding subtracting the quantity $\frac{a_n}{(1+\delta_N)}$ yields that: \left\Vert \left\{H_N(\lambda_n)\left(\frac{1-\left\vert \lambda_n\right\vert^2}{1-\left\vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-a_n\right\}\right\Vert_{N,\ell^2} \leq & & \\ \left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1+\delta_N)(1-\kappa_N^2)^{\frac{1}{2}}} \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right)+\delta_N\right)\left\Vert Note that the quantity: $$\left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\ | \vert \Theta(\lambda_n)\right\vert^2}\Right)^{\frac{1}{2}}-\fRac{a_n}{\LefT\VeRt A\rigHt\VeRt_{N,\ell^2}}\right\}\rigHT\VerT_{N,\ell^2} \leq \frac{\sup_{m\geq N}\leFt\verT\THEta(\lAMbDa_m)\riGht\vert}{(1-\KApPA_n^2)^{\frAc{1}{2}}} \LeFt(1+\lEfT\veRt \lefT\{g_N(\Lambda_n)\Left(1-\left\veRt \lAmBda_n\right\verT^2\RiGht)^{\frac{1}{2}}\rigHt\}\rIght\Vert_{N,\ell^2}\RigHt).$$ DefiNe $h_N=(1+\dELta_N)^{-1} \lEft\vert a\Right\VERt_{N,\ell^2} H_N$, and then We HAve $H_N\iN k_{\Theta}$ aND $\LeFt\VeRt H_N\right\Vert_{2}\leq \lEFt\vErt a\right\Vert_{N,\Ell^2}$. UsiNg THe LASt eStiMate and addInG and sUBtractiNG tHE QUanTIty $\frac{a_n}{(1+\deltA_N)}$ yields thaT: $$\BegIn{aligNeD}
\leFT\Vert \lEft\{H_N(\LaMBda_N)\left(\frac{1-\leFt\veRt \lambda_n\Right\vERt^2}{1-\left\vERt \Theta(\Lambda_N)\riGht\Vert^2}\RIgHt)^{\FraC{1}{2}}-a_N\RigHT\}\rIghT\verT_{N,\ell^2} \leq & &
\\ \LeFt(\Frac{\sUp_{m\gEQ n}\LEft\vErt\thetA(\lambDa_m)\right\vert}{(1+\dEltA_N)(1-\kaPPa_N^2)^{\Frac{1}{2}}} \lEft(1+\leFt\VeRt \Left\{g_n(\lambdA_n)\lefT(1-\lEft\vert \lambda_n\rIght\Vert^2\right)^{\FraC{1}{2}}\rIghT\}\rIght\VERt_{N,\ell^2}\RigHt)+\dElta_N\riGht)\left\vErt A\rIGHT\VErt_{N,\ell^2}.\end{aligned}$$ NOtE THaT the quanTity: $$\leFT(\fRaC{\Sup_{m\geq N}\LeFt\vErt\THETa(\lamBda_m)\RIgHt\ | \vert \Theta(\lambda_n)\ri ght\vert^2 }\rig ht) ^{\ fr ac{1 }{2} }-\frac{a_n}{\ l eft\ Vert a\right\Vert_{N,\ ell^2 }} \ righ t \} \righ t\Vert_ { N, \ e ll^ 2} \ leq \ f ra c{\su p_{ m\geq N }\left\ver t\T he ta(\lambda_m ) \r ight\vert} {(1 -\kappa_N^2) ^{\ frac{1 }{ 2}} } \lef t(1 +\lef t\Vert \left\ {g_N(\lam bd a _n)\le f t(1-\le f t \v ert\lambda_n\right\v e rt ^ 2\right)^{\fra c{1}{2 }} \ ri g h t\} \ri ght\Vert_{ N, \ell^ 2 }\right ) .$ $ D efi n e $H_N=(1+\de lta_N)^{-1} \le ft\Ver ta\r i ght\Ve rt_{N ,\ e ll^ 2} h_N$, an d th en we hav e $H_N \ in K_{\ T heta}$and $\ lef t\V ertH _N \r igh t\ V ert _ {2 }\l e q \ left\Ver ta\ right \Ver t _ { N ,\el l^2 }$.Using the last est ima te a n d a dding andsubt ra cting the q uanti ty $\frac{a_n}{(1 +\de lta_N)}$yie ld s t ha t: $$ \ begin{ ali gne d}
\lef t\Vert\ lef t\ { H _ N( \lambda_n)\left(\f ra c { 1- \left\ve rt \la m bd a_ n \right\v er t^2 }{1- \ l eft\v ert\ Th eta(\lam bda_n) \ ri gh t\vert^ 2} \right )^ {\f rac {1}{2 } }-a_ n\righ t\}\righ t\Ver t _{N,\ell^2} \ leq & &
\\\l e f t( \ frac {\s up_{m\geq N }\le f t\ve rt\T h et a(\ l ambda _m)\r ig h t\ v ert}{(1+\delta_N)(1 -\ kappa_ N^2)^ {\frac{1}{2}} } \left(1+ \ l e ft\Vert\lef t \{ g _N(\lambda_n)\ left( 1-\left\ve r t \lambd a_n\r ight\ver t^2\right ) ^ {\frac{1 }{2 }}\ rig ht\ } \ ri ght\Vert_{N,\ e l l^2} \r ight)+\ del ta_N\ri ght )\l eft \Ve rt a\right\ Vert_{N, \e ll ^2 }. \en d{ali g ned}$$ N ot e t ha t t he qu a ntity: $$\l eft( \f ra c {\s up_{m\g e qN } \lef t\ ve rt\T het a( \lamb da_m ) \ri ght\ | \vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-\frac{a_n}{\left\Vert_a\right\Vert_{N,\ell^2}}\right\}\right\Vert_{N,\ell^2} \leq_\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1-\kappa_N^2)^{\frac{1}{2}}} \left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert_\lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right).$$ Define_$H_N=(1+\delta_N)^{-1}_\left\Vert a\right\Vert_{N,\ell^2}_h_N$,_and then we_have $H_N\in K_{\Theta}$_and $\left\Vert H_N\right\Vert_{2}\leq \left\Vert_a\right\Vert_{N,\ell^2}$. Using the_last_estimate and adding and subtracting the quantity $\frac{a_n}{(1+\delta_N)}$ yields that: $$\begin{aligned}
\left\Vert \left\{H_N(\lambda_n)\left(\frac{1-\left\vert \lambda_n\right\vert^2}{1-\left\vert \Theta(\lambda_n)\right\vert^2}\right)^{\frac{1}{2}}-a_n\right\}\right\Vert_{N,\ell^2}__ \leq_&_&_
\\ \left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\vert}{(1+\delta_N)(1-\kappa_N^2)^{\frac{1}{2}}}_\left(1+\left\Vert \left\{g_N(\lambda_n)\left(1-\left\vert \lambda_n\right\vert^2\right)^{\frac{1}{2}}\right\}\right\Vert_{N,\ell^2}\right)+\delta_N\right)\left\Vert a\right\Vert_{N,\ell^2}.\end{aligned}$$ Note_that the_quantity: $$\left(\frac{\sup_{m\geq N}\left\vert\Theta(\lambda_m)\right\ |
is at $ k \sim 1.5 ~h {\rm Mpc}^{-1}$. Initial velocities for WDM are drawn from a Fermi-Dirac distribution and added to the proper velocity assigned by linear theory.
In our second prescription, WDM is implemented starting directly from a properly designed CAMB power spectrum, by setting $\Delta N_{\rm eff}$ to express departures from the canonical $N_{\rm eff}$ value according to the relic mass – where $N_{\rm eff} = 3.046 + \Delta N_{\rm eff}$. These runs are simply indicated as WDM in Table \[table\_supporting\_sims\_wdm\], and we adopt this implementation for our [*Grid Suite*]{}, since it is more physically motivated and it is also straightforward (although delicate) to include massive neutrinos within this framework. The inclusion is simply achieved by accounting for the neutrino mass degeneracy, where: $$\Delta N_{\rm eff} = \Big ( {93.14 \cdot \Omega_{\rm WDM} h^2 \over m_{\rm WDM} } \Big )^{4/3}.$$ We eventually use the one-to-one mapping between thermal relics and keV sterile neutrinos to simulate massive sterile neutrino cosmologies in the DW mechanism – as previously explained.
Within the second prescription, we also simulate for the first time cosmologies with both massive active and sterile neutrinos, in a consistent fashion. This is achieved by introducing two neutrino eigenstates (one for the active ones, and one for the sterile counterpart) in a three-component simulation, and then by considering a neutrino mass splitting – so that $\Omega_{\nu} = \Omega_{\rm m} - \Omega_{\rm b}$ now contains two distinct contributions. In essence, three active massive-with-degenerate-mass neutrinos contribute as $N_{\rm eff} =3.046$, while a non-thermalized massive sterile neutrino contributes with an additional $\Delta N_{\rm eff}$ to the number of effective neutrino species, directly proportional to its mass. Such runs represent a unique novelty in the literature, and allow us to study the combined effects of massive and sterile neutrinos, as well as to put constraints on pure WDM scenarios directly from Ly$\alpha$ forest data. The latter simulations are indicated as $\sum m_{\nu}^{+} m_{\rm WDM}^{+}$ in Table \[table\_grid\_sims\_cross\_terms\] ([*Grid*]{} runs), and as [*NU\_WDM*]{} in Table \[table\_ | is at $ k \sim 1.5 ~h { \rm Mpc}^{-1}$. Initial velocities for WDM are drawn from a Fermi - Dirac distribution and added to the proper velocity impute by analogue theory.
In our second prescription, WDM is follow through begin directly from a properly design CAMB might spectrum, by setting $ \Delta N_{\rm eff}$ to carry departures from the canonical $ N_{\rm eff}$ value accord to the relic mass – where $ N_{\rm eff } = 3.046 + \Delta N_{\rm eff}$. These runs are just argue as WDM in Table \[table\_supporting\_sims\_wdm\ ], and we adopt this execution for our [ * Grid Suite * ] { }, since it is more physically motivated and it is also square (although finespun) to include massive neutrinos within this framework. The inclusion body is simply achieved by accounting for the neutrino mass degeneracy, where: $ $ \Delta N_{\rm eff } = \Big ({ 93.14 \cdot \Omega_{\rm WDM } h^2 \over m_{\rm WDM } } \Big) ^{4/3}.$$ We eventually use the one - to - one function between thermal relics and keV aseptic neutrinos to simulate massive sterile neutrino cosmologies in the DW mechanism – as previously explain.
Within the second prescription, we also simulate for the first time cosmology with both massive active and sterile neutrinos, in a consistent fashion. This is achieved by introducing two neutrino eigenstates (one for the active ones, and one for the sterile counterpart) in a three - component simulation, and then by consider a neutrino mass splitting – so that $ \Omega_{\nu } = \Omega_{\rm m } - \Omega_{\rm b}$ now control two clear-cut contributions. In essence, three active massive - with - debauched - mass neutrino contribute as $ N_{\rm eff } = 3.046 $, while a non - thermalized massive sterile neutrino contributes with an extra $ \Delta N_{\rm eff}$ to the number of effective neutrino species, directly proportional to its mass. Such runs represent a singular novelty in the literature, and allow us to study the combined impression of massive and aseptic neutrino, equally well as to put constraints on pure WDM scenarios directly from Ly$\alpha$ forest data. The latter simulations are indicated as $ \sum m_{\nu}^{+ } m_{\rm WDM}^{+}$ in Table \[table\_grid\_sims\_cross\_terms\ ] ([ * Grid * ] { } rivulet), and as [ * NU\_WDM * ] { } in Table \[table\ _ | is at $ k \sim 1.5 ~h {\rm Mpc}^{-1}$. Iniuial velocities for WDM ere drasn from x Fermi-Dirac distribution anv adeed ti the proper velocity xssigned hy lineae thtory.
In our second prescriicion, SFM iv implemented sjarting direwtly from a prmpdrpy designed CAMB power spectrum, by fetting $\Dflta N_{\rm eff}$ tj exkrefs dsiavtures from the canonical $N_{\rm efr}$ value according to the relic mass – where $N_{\rm eff} = 3.046 + \Delta N_{\rm eff}$. Hhese runs qre fumply indicaged as WDM in Table \[tagle\_supporting\_sims\_wdm\], and we adoot thns implemenjcrioj for our [*Grmd Suine*]{}, since it is more pvysicalky motivated akd it is also straightforward (although delicate) to incluge massive neutrinow qithit thhs ffqmedori. Vhe includioi is simply achieved bt accounting for tht nqlyrino mass dsgenerwcr, where: $$\Delta N_{\rm eff} = \Big ( {93.14 \cdot \Omega_{\gm WSM} h^2 \over m_{\rm WDM} } \Big )^{4/3}.$$ We eventually use thg one-to-one mapping between thermal relics and keV sterile nautriior ti fknupate massive sterile neutrino cosmologies in ege DE mechanism – af previouslu fxljained.
Within jhe second prescription, we apso simolate dor the fyrst time cosmologies with both massive actpve qnd sterile neutriuos, in a conristgnt faxhion. This is achieved yy intdoducing twl neutrink eigenstates (one fog tha active ones, and one for ehe sterioe cpunterpxrt) on a trree-componfnt slkulation, and then hy cousidesing a neuhrino mass splitting – so that $\Omega_{\nu} = \Omega_{\rk k} - \Mmega_{\rm y}$ now gontains two diftinct contribotions. In essevce, three zctive kassive-with-qegenerate-masv neutrinos cmntributq as $N_{\rm eff} =3.046$, wfkle a non-thermslized massive sterule neutrino contrlbuter with an additiinao $\Delta N_{\rm eff}$ to thq jukbqs of effectiee ndutfono soecies, diregtlh prpportional to its mavs. Shch runs represent a unique bovelty yn the literayure, and allow us uo stuvy the comboneq effects of massive and sterime neutrijos, as well as eo pmt cjnstraints on pure WDM scenarios directly from Ly$\elpha$ forest data. The lqtter simulations ate lndicated as $\sum i_{\nu}^{+} m_{\rm WGM}^{+}$ in Table \[table\_grie\_sims\_cross\_terms\] ([*Gvid*]{} runs), and as [*NU\_WDM*]{} jn Tabne \[tahle\_ | is at $ k \sim 1.5 ~h Initial for WDM drawn from a the velocity assigned by theory. In our prescription, WDM is implemented starting directly a properly designed CAMB power spectrum, by setting $\Delta N_{\rm eff}$ to express from the canonical $N_{\rm eff}$ value according to the relic mass – where eff} 3.046 \Delta eff}$. These runs are simply indicated as WDM in Table \[table\_supporting\_sims\_wdm\], and we adopt this implementation our [*Grid Suite*]{}, since it is more physically and it is also (although delicate) to include massive within framework. The is achieved accounting for the mass degeneracy, where: $$\Delta N_{\rm eff} = \Big ( {93.14 \cdot \Omega_{\rm WDM} h^2 \over m_{\rm WDM} \Big )^{4/3}.$$ use the mapping thermal and keV sterile simulate massive sterile neutrino cosmologies in – as previously explained. Within the second prescription, also simulate the first time cosmologies with both active and sterile neutrinos, in a consistent fashion. is achieved by introducing two neutrino eigenstates (one for the active ones, and one for counterpart) in a three-component and then by a mass – that $\Omega_{\nu} \Omega_{\rm m} - \Omega_{\rm b}$ now contains two distinct contributions. In three active massive-with-degenerate-mass neutrinos contribute as $N_{\rm eff} =3.046$, while massive neutrino contributes with additional $\Delta N_{\rm eff}$ the of effective neutrino species, to mass. a novelty the literature, and allow to study the combined effects massive and sterile neutrinos, constraints on pure WDM scenarios directly from Ly$\alpha$ data. The latter simulations are indicated as m_{\nu}^{+} m_{\rm WDM}^{+}$ in Table \[table\_grid\_sims\_cross\_terms\] ([*Grid*]{} runs), and as [*NU\_WDM*]{} in \[table\_ | is at $ k \sim 1.5 ~h {\rm Mpc}^{-1}$. Initial veloCities for WdM are DraWn fRoM a FeRmi-DIrac distributiON and Added to the proper velociTy assIgNEd by LInEar thEory.
In oUR sECOnd PrEsCriPtIOn, wDM is ImpLementeD starting dIreCtLy from a propeRLy Designed CAmB pOwer spectrum, By sEtting $\deLta n_{\Rm eff}$ To eXpresS deparTUres frOm the canoNiCAl $N_{\rm eFF}$ value aCCOrDing To the relic mass – wheRE $N_{\RM eff} = 3.046 + \Delta N_{\rm efF}$. These RuNS aRE SimPly Indicated aS WdM in TABle \[tablE\_SuPPORtiNG\_sims\_wdm\], and we Adopt this imPLemEntatiOn For OUr [*Grid suite*]{}, SiNCe iT is more physIcalLy motivatEd and iT Is also sTRaightfOrward (AltHouGh deLIcAtE) to InCLudE MaSsiVE neUtrinos wItHiN this FramEWORK. The IncLusiOn is sImply achieved By aCcouNTinG for tHe neuTrinO mAss deGeneraCy, wheRe: $$\delta N_{\rm eff} = \Big ( {93.14 \cDot \OMega_{\rm WDM} H^2 \ovEr M_{\rm wDm} } \Big )^{4/3}.$$ WE EventuAllY usE the one-To-one maPPinG bETWEeN thermal relics and kEV STErIle neutrInos to SImUlATe massivE sTerIle nEUTrino CosmOLoGies in thE DW mecHAnIsM – as prevIoUsly exPlAinEd.
WIthin THe seCond prEscriptiOn, we aLSo simulate for tHE first time cosMOlOGIeS With BotH massive actIve aND steRile NEuTriNOs, in a ConsiStENt FAshion. This is achieveD bY introDucinG two neutrino eIgenstates (ONE For the acTive ONeS, And one for the stErile CounterparT) In a three-CompoNent simuLation, and THEn by consIdeRinG a nEutRINo Mass splitting – SO That $\omEga_{\nu} = \OmEga_{\Rm m} - \OmegA_{\rm B}$ noW coNtaInS two distiNct contrIbUtIoNs. in eSsencE, Three actIvE maSsIve-With-dEGeneraTe-masS neuTrInOS coNtributE As $n_{\RM eff} =3.046$, WhIlE a noN-thErMalizEd maSSivE sterilE neutrino ConTRibuTeS wIth an adDitional $\Delta n_{\rM eff}$ to the nUmBer Of effeCTIve neutrIno species, directly propoRTional tO itS mass. such Runs repreSenT a uniqUe nOVelty iN the liTeratUrE, anD ALlow uS TO sTudY tHe combined EFFecTs of mAsSive And sterIle neutrinos, as well AS to Put constraintS on Pure wdm sCenARiOS diReCTly FROm Ly$\alpha$ forest Data. The latTeR SiMulations aRE inDiCated as $\Sum m_{\nu}^{+} m_{\Rm WDM}^{+}$ IN Table \[tAble\_grid\_sIms\_cross\_tErMs\] ([*GrID*]{} RunS), and as [*NU\_WDm*]{} in Table \[Table\_ | is at $ k \sim 1.5 ~h {\r m Mpc}^{-1 }$. I nit ial v eloc itie s for WDM ared rawn from a Fermi-Dirac di strib ut i on a n dadded to the pr o p erve lo cit ya ss igned by linear theory.
Inou r second pre s cr iption, WD M i s implemente d s tartin gdir e ctlyfro m a p roperl y desig ned CAMBpo w er spe c trum, b y se ttin g $\Delta N_{\rme ff } $ to express d epartu re s f r o m t hecanonical$N _{\rm eff}$ v a lu e a cco r ding to the r elic mass – whe re $N_ {\ rme ff} =3.046 + \De lta N_{\rmeff} $. Theseruns a r e simpl y indica ted as WD M i n Ta b le \ [ta bl e \_s u pp ort i ng\ _sims\_w dm \] , and wea d o p t th isimpl ement ation for our [* Grid Sui te*]{ }, si nceit is m ore ph ysica ll y motivated and itis also s tra ig htf or ward( althou ghdel icate)to incl u dema s s i ve neutrinos withinth i s f ramework . Thei nc lu s ion is s im ply ach i e ved b y ac c ou nting fo r then eu tr ino mas sdegene ra cy, wh ere:$ $\De lta N_ {\rm eff } = \ B ig ( {93.14 \c d ot \Omega_{\r m W D M }h ^2 \ ove r m_{\rm WD M} } \Big )^{ 4 /3 }.$ $ We e ventu al l yu se the one-to-one m ap ping b etwee n thermal rel ics and ke V s terile n eutr i no s to simulate m assiv e sterilen eutrinocosmo logies i n the DWm e chanism– a s p rev iou s l yexplained.
W i t hinth e secon d p rescrip tio n,weals osimulatefor thefi rs tti mecosmo l ogies wi th bo th ma ssive active andster il en eut rinos,i na cons is te nt f ash io n. Th is i s ac hievedby introd uci n g tw one utrinoeigenstates ( on e for theac tiv e ones , and onefor the sterile counter p art) in athree -com ponent si mul ation, an d thenby con sider in g a n eutri n o m ass s plitting – s o t hat $ \O mega _{\nu}= \Omega_{\rm m} - \Om ega_{\rm b}$now con t a in s t w od ist in c t c o n tributions. Inessence, t hr e eactive mas s ive -w ith-deg enerate -mass neutrin os contri bute as $ N_ {\rm e ff} =3.046$,while anon-therm a lized ma ssive st erilene utr ino c ontrib u tes with an ad di tional $\De lt a N_{\rm eff}$ to the number of effec tiveneu trino spe cie s , d irectly p ropo rtional to it s m ass.Suc h runs rep r es ent a uni quen ovelty in th e l i t er ature, anda l l owus to st u dy the com bined effects ofm assive and ste rile n eut rin o s, a swell as to put co ns t r aints on p ure WDM sce narios d ir e ctlyfrom L y$\alp ha$ for e s td ata. T he l att er simula tio ns are ind ic at e d as $ \sum m _{\nu} ^{+} m _ {\rm W DM}^{+}$ in Tabl e \[t a b le\_g r id\ _sims \_ cross\_ t erms \] ([*Grid *]{} runs), and a s [* NU\_W DM*]{}in Table \[ ta ble\_ | is_at $_k \sim 1.5 ~h_{\rm Mpc}^{-1}$._Initial_velocities for_WDM_are drawn from_a Fermi-Dirac distribution_and added to the_proper velocity assigned_by_linear theory.
In our second prescription, WDM is implemented starting directly from a properly designed_CAMB_power spectrum,_by_setting_$\Delta N_{\rm eff}$ to express_departures from the canonical $N_{\rm_eff}$ value_according to the relic mass – where $N_{\rm_eff}_= 3.046 +_\Delta N_{\rm eff}$. These runs are simply indicated as_WDM in Table \[table\_supporting\_sims\_wdm\], and we_adopt this implementation_for_our_[*Grid Suite*]{}, since it_is more physically motivated and it_is also straightforward (although delicate) to_include massive neutrinos within this framework. The_inclusion is simply achieved by accounting_for the neutrino mass degeneracy,_where: $$\Delta_N_{\rm eff} = \Big (_{93.14 \cdot \Omega_{\rm_WDM} h^2_\over m_{\rm WDM}_} \Big )^{4/3}.$$ We eventually use_the one-to-one mapping_between thermal relics and keV sterile_neutrinos_to simulate massive_sterile_neutrino_cosmologies in_the DW mechanism_–_as previously_explained.
Within_the second prescription, we also simulate_for_the first time cosmologies with both massive_active and sterile neutrinos,_in_a consistent fashion. This_is achieved by introducing two_neutrino eigenstates (one for the active_ones, and_one for_the sterile counterpart) in a three-component simulation, and then by considering_a neutrino mass splitting – so_that $\Omega_{\nu} = \Omega_{\rm_m} -_\Omega_{\rm_b}$ now contains_two_distinct contributions._In essence, three active massive-with-degenerate-mass neutrinos contribute_as $N_{\rm_eff} =3.046$, while a non-thermalized massive_sterile neutrino contributes with_an_additional $\Delta N_{\rm eff}$ to the_number of effective neutrino species, directly_proportional to its mass. Such_runs_represent_a unique novelty in the_literature, and allow us to study_the combined effects_of massive and sterile neutrinos, as well_as_to put constraints on pure WDM_scenarios_directly from Ly$\alpha$ forest data. The_latter_simulations_are indicated as $\sum m_{\nu}^{+}_m_{\rm WDM}^{+}$ in Table \[table\_grid\_sims\_cross\_terms\] ([*Grid*]{}_runs), and as [*NU\_WDM*]{} in Table \[table\_ |
[
\left( -1+\sqrt{1+16a^2\mu^2}\right) /(4a\mu)\right].$
2. If $\mu<0$, then $S$ has also only a normal magnetic parallel corresponding to $t_1=2\pi -t_0$.
[Finally, let us consider the cone $M$ generated by the line $\alpha (t)=(at,0,bt), \; a^2+b^2=1, \; t>0$ around the $z$-axis. Then, it can be seen that for any $\mu>0$, there exists a unique normal magnetic parallel given by $t=1/ \mu$.]{}
Relativistic particles with rigidity of order one
=================================================
The search for Lagrangians describing spinning particles (both massive and massless) has a long history. An interesting and unconventional approach is to provide the necessary extra degrees of freedom by actions whose densities depend on higher order geometrical invariants. In particular, this means that those extra bosonic variables must be encoded in the geometry of the world trajectories. The simplest models are those involving density Lagrangians that depend on the curvature, $\kappa$, of the worldlines ([@Arroyo]-[@Nesterenko1], etc.). In particular, actions that depend linearly from $\kappa$ ([@Plyushchay2],[@Arroyo-Barros-Garay]-[@Plyushchay3], etc.) will be considered in this section. These models describing a massive relativistic boson [@Plyushchay].
Suppose that $\Lambda$ is a suitable space of curves (closed curves or clamped curves, for instance) in a Riemannian surface $(M,g)$. Define a one-parameter family of functionals $\mathcal{F}_{m}:\Lambda\rightarrow\mathbb{R}, \; m\in \R,$ by $$\label{funcional}
\mathcal{F}_{m}(\gamma)=\int_{\gamma}(\kappa+m)ds,$$where $s$ stands for the arclength parameter of curves $\gamma\in\Lambda$. In order to obtain the first variation of these actions, we use the following standard machinery (see for instance [@Barros]). For a curve $\gamma:[0,L]\rightarrow M$, we take variations $\Theta=\Theta(t,r):[0,L]\times(-\varepsilon,\varepsilon)\rightarrow
M$ with $\Theta( | [
\left (-1+\sqrt{1 + 16a^2\mu^2}\right) /(4a\mu)\right].$
2. If $ \mu<0 $, then $ S$ has also only a normal magnetic parallel correspond to $ t_1=2\pi -t_0$.
[ ultimately, let us consider the cone $ M$ generated by the cable $ \alpha (t)=(at,0,bt), \; a^2+b^2=1, \; t>0 $ around the $ z$-axis. Then, it can be seen that for any $ \mu>0 $, there exists a singular normal charismatic parallel given by $ t=1/ \mu$. ] { }
Relativistic particles with rigidity of order one
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
The search for Lagrangians identify spinning particles (both massive and massless) has a farseeing history. An interesting and improper approach is to provide the necessary excess degrees of exemption by natural process whose densities depend on higher orderliness geometrical invariants. In especial, this means that those excess bosonic variables must be encode in the geometry of the world trajectories. The simplest models are those involve density Lagrangians that depend on the curvature, $ \kappa$, of the worldlines ([ @Arroyo]-[@Nesterenko1 ], etc .). In particular, actions that depend linearly from $ \kappa$ ([ @Plyushchay2],[@Arroyo - Barros - Garay]-[@Plyushchay3 ], etc .) will be regard in this section. These models describing a massive relativistic boson [ @Plyushchay ].
Suppose that $ \Lambda$ is a suitable space of curves (closed curves or clamped curve, for instance) in a Riemannian surface $ (M, g)$. specify a one - argument syndicate of functionals $ \mathcal{F}_{m}:\Lambda\rightarrow\mathbb{R }, \; m\in \R,$ by $ $ \label{funcional }
\mathcal{F}_{m}(\gamma)=\int_{\gamma}(\kappa+m)ds,$$where $ s$ stands for the arclength parameter of curves $ \gamma\in\Lambda$. In order to obtain the first variation of these action, we use the following standard machinery (see for instance [ @Barros ]). For a curve $ \gamma:[0,L]\rightarrow M$, we take variation $ \Theta=\Theta(t, r):[0,L]\times(-\varepsilon,\varepsilon)\rightarrow
M$ with $ \Theta ( | [
\left( -1+\sqrt{1+16a^2\mu^2}\right) /(4a\mu)\rinht].$
2. If $\mu<0$, then $S$ has anso onmy a noroal magnetic parallel corres'ondung ti $t_1=2\pi -t_0$.
[Finally, let us zonsider nhe cone $N$ geierated by the lmhe $\alpha (t)=(at,0,bf), \; a^2+b^2=1, \; t>0$ around the $e$-axis. Then, id can be seen dhxt for any $\mu>0$, there exists a unique njrmal msgjetic parallel givtn fy $t=1/ \mu$.]{}
Relativistic particles with rigjdity oh order one
=================================================
The xearch for Lagrangians desfriblng spinning partifles (both mqssidw and masslers) has a long history. Zn interesting and unconventionxl ap'roach is ti proghde the necxssary extra degrecx of fseedom ny actions whoxe vensuties depend on highec order geometrical ynvariantv. Nn particular, this meqnw thaj thove ebrra boaoiic variahlea must be sncoded in rhe geometry of the wjgkd trajectorjes. Thq fimplest models are those involving denvitg Lagrangians that depebd on the curvature, $\kwppa$, of tre worldlines ([@Arroyo]-[@Nesterenko1], etc.). In particular, dctiois thcb awpfnd linearly from $\kappa$ ([@Plyushchay2],[@Arroyo-Barrof-Fataj]-[@Plyushchay3], etc.) wlll be considered on tnys section. Thgse modzms describing a masslve relwtiviwtic bosog [@Pluushchay].
Suppose that $\Lambda$ is a suitabjw space of curves (elosed curver or clamled curves, for instance) in z Riemanniaj surface $(O,g)$. Define a one-pxrakeder family of functionals $\iathcal{F}_{m}:\Oambba\rightafrow\kathbb{W}, \; m\in \R,$ bj $$\labcn{funcional}
\mathcal{F}_{l}(\gammc)=\int_{\gdmma}(\kappa+m)fs,$$where $s$ stands for the arclenjvh parameter pf cugves $\gammc\in\Lamnda$. In order to obtain the fitst variacion ow these acnions, we nse the folljwing standarg machinery (sxe for inftanxe [@Bqrros]). Fuf a curve $\gamms:[0,L]\rightargoc M$, we taje variations $\Thete=\Theja(f,r):[0,L]\times(-\varepsioon,\carepsilon)\rightsrruw
M$ wpth $\Trata( | [ \left( -1+\sqrt{1+16a^2\mu^2}\right) /(4a\mu)\right].$ 2. If $\mu<0$, has only a magnetic parallel corresponding us the cone $M$ by the line (t)=(at,0,bt), \; a^2+b^2=1, \; t>0$ around $z$-axis. Then, it can be seen that for any $\mu>0$, there exists a normal magnetic parallel given by $t=1/ \mu$.]{} Relativistic particles with rigidity of order ================================================= search Lagrangians spinning particles (both massive and massless) has a long history. An interesting and unconventional approach is provide the necessary extra degrees of freedom by whose densities depend on order geometrical invariants. In particular, means those extra variables be in the geometry the world trajectories. The simplest models are those involving density Lagrangians that depend on the curvature, $\kappa$, the worldlines In particular, that linearly $\kappa$ ([@Plyushchay2],[@Arroyo-Barros-Garay]-[@Plyushchay3], etc.) considered in this section. These models relativistic boson [@Plyushchay]. Suppose that $\Lambda$ is a space of (closed curves or clamped curves, for in a Riemannian surface $(M,g)$. Define a one-parameter of functionals $\mathcal{F}_{m}:\Lambda\rightarrow\mathbb{R}, \; m\in \R,$ by $$\label{funcional} \mathcal{F}_{m}(\gamma)=\int_{\gamma}(\kappa+m)ds,$$where $s$ stands for the arclength parameter $\gamma\in\Lambda$. In order to the first variation these we the standard machinery for instance [@Barros]). For a curve $\gamma:[0,L]\rightarrow M$, we take variations M$ with $\Theta( | [
\left( -1+\sqrt{1+16a^2\mu^2}\right) /(4a\mu)\right].$
2. IF $\mu<0$, then $S$ haS also OnlY a nOrMal mAgneTic parallel corREspoNding to $t_1=2\pi -t_0$.
[Finally, let uS consIdER the COnE $M$ genErated bY ThE LIne $\AlPhA (t)=(aT,0,bT), \; A^2+b^2=1, \; T>0$ arouNd tHe $z$-axis. then, it can bE seEn That for any $\mu>0$, THeRe exists a uNiqUe normal magnEtiC paralLeL giVEn by $t=1/ \Mu$.]{}
RElatiVistic PArticlEs with rigIdITy of orDEr one
=================================================
ThE SEaRch fOr Lagrangians descRIbINg spinning partIcles (bOtH MaSSIve And Massless) haS a Long hIStory. An INtEREStiNG and unconventIonal approaCH is To provIdE thE NecessAry exTrA DegRees of freedOm by Actions whOse denSIties dePEnd on hiGher orDer GeoMetrICaL iNvaRiANts. iN pArtICulAr, this meAnS tHat thOse eXTRA BosoNic VariAbles Must be encoded In tHe geOMetRy of tHe worLd trAjEctorIes. The SimplEsT models are those InvoLving densIty laGraNgIans tHAt depeNd oN thE curvatUre, $\kappA$, Of tHe WORLdLines ([@Arroyo]-[@NestereNkO1], ETc.). in particUlar, acTIoNs THat depenD lIneArly FROm $\kapPa$ ([@PlYUsHchay2],[@ArrOyo-BarROs-gaRay]-[@PlyuShChay3], etC.) wIll Be cOnsidERed iN this sEction. ThEse moDEls describing a MAssive relativIStIC BoSOn [@PlYusHchay].
SupposE thaT $\lambDa$ is A SuItaBLe spaCe of cUrVEs (CLosed curves or clampeD cUrves, fOr insTance) in a RiemaNnian surfaCE $(m,G)$. Define a One-pARaMEter family of fuNctioNals $\mathcaL{f}_{m}:\Lambda\RightArrow\matHbb{R}, \; m\in \R,$ bY $$\LAbel{funcIonAl}
\mAthCal{f}_{M}(\GaMma)=\int_{\gamma}(\kaPPA+m)ds,$$WhEre $s$ staNds For the aRclEngTh pAraMeTer of curvEs $\gamma\iN\LAmBdA$. IN orDer to OBtain the FiRst VaRiaTion oF These aCtionS, we uSe ThE FolLowing sTAnDARd maChInEry (sEe fOr InstaNce [@BARroS]). For a cuRve $\gamma:[0,L]\RigHTarrOw m$, wE take vaRiations $\Theta=\thEta(t,r):[0,L]\timeS(-\vArePsilon,\VARepsilon)\Rightarrow
M$ with $\Theta( | [
\left( -1+\sqrt{1+16 a^2\mu^2}\ right ) / (4a \m u)\r ight ].$
2. If $\ m u<0$ , then $S$ has also on ly ano r malm ag netic parall e lc o rre sp on din gt o$t_1= 2\p i -t_0$ .
[Finall y,le t us conside r t he cone $M $ g enerated bythe line$\ alp h a (t) =(a t,0,b t), \; a^2+b^ 2=1, \; t >0 $ aroun d the $z $ - ax is.Then, it can be s e en that for any $ \mu>0$ ,t he r e ex ist s a unique n ormal magneti c p a r a lle l given by $t= 1/ \mu$.]{}
Re lativi st icp articl es wi th rig idity of or derone
===== ====== = ======= = ======= ====== === === ==== = == == =
Th e se a rc h f o r L agrangia ns d escri bing s p i nnin g p arti cles(both massive an d ma s sle ss) h as along h istor y. Aninter es ting and unconv enti onal appr oac histo prov i de the ne ces sary ex tra deg r ees o f f re edom by actions wh os e de nsitiesdepend on h i gher ord er ge omet r i cal i nvar i an ts. In p articu l ar ,this me an s that t hos e e xtrab oson ic var iables m ust b e encoded in th e geometry oft he w or l d tr aje ctories. Th e si m ples t mo d el s a r e tho se in vo l vi n g density Lagrangia ns thatdepen d on the curv ature, $\k a p p a$, of t he w o rl d lines ([@Arroy o]-[@ Nesterenko 1 ], etc.) . Inparticul ar, actio n s that de pen d l ine arl y fr om $\kappa$ ( [ @ Plyu sh chay2], [@A rroyo-B arr os- Gar ay] -[ @Plyushch ay3], et c. )wi ll be cons i dered in t his s ect ion.T hese m odels des cr ib i nga massi v er e lati vi st ic b oso n[@Ply ushc h ay] .
Supp ose that$\L a mbda $is a suit able space of c urves (clo se d c urveso r clamped curves, for instance)i n a Rie man niansurf ace $(M,g )$. Defin e a one-pa ramete r fam il y o f funct i o na ls$\ mathcal{F} _ { m}: \Lamb da \rig htarrow \mathbb{R}, \; m\i n \R ,$ by $$\labe l{f unci o n al }
\ m at h cal {F } _{m } ( \gamma)=\int_{\ gamma}(\ka pp a +m )ds,$$wher e $s $standsfor the arcl e ngth pa rameter o f curves$\ gamm a \ in\ Lambda$. I n orderto obtain the f i rs t var iat ion of t hes e act ions,w e u se th e foll ow ing st andar dmachiner y (see for instance [@B arros] ). Fo r a curve $\ gam m a:[ 0,L]\righ tarr ow M$, wetak e v ariat ion s $\Th eta= \ Th eta ( t,r): [0,L ] \times(-\ v ar eps i l on ,\varepsilo n ) \ rig htarr owM $ with $\T heta( | [
_ _\left( -1+\sqrt{1+16a^2\mu^2}\right) /(4a\mu)\right].$
2. _If $\mu<0$,_then_$S$ has_also_only a normal_magnetic parallel corresponding_to $t_1=2\pi -t_0$.
[Finally, let_us consider the_cone_$M$ generated by the line $\alpha (t)=(at,0,bt), \; a^2+b^2=1, \; t>0$ around the $z$-axis._Then,_it can_be_seen_that for any $\mu>0$, there_exists a unique normal magnetic_parallel given_by $t=1/ \mu$.]{}
Relativistic particles with rigidity of order_one
=================================================
The_search for Lagrangians_describing spinning particles (both massive and massless) has a_long history. An interesting and unconventional_approach is to_provide_the_necessary extra degrees of_freedom by actions whose densities depend_on higher order geometrical invariants. In_particular, this means that those extra bosonic_variables must be encoded in the_geometry of the world trajectories._The simplest_models are those involving density_Lagrangians that depend_on the_curvature, $\kappa$, of_the worldlines ([@Arroyo]-[@Nesterenko1], etc.). In particular,_actions that depend_linearly from $\kappa$ ([@Plyushchay2],[@Arroyo-Barros-Garay]-[@Plyushchay3], etc.) will_be_considered in this_section._These_models describing_a massive relativistic_boson_[@Plyushchay].
Suppose that_$\Lambda$_is a suitable space of curves_(closed_curves or clamped curves, for instance) in_a Riemannian surface $(M,g)$._Define_a one-parameter family of_functionals $\mathcal{F}_{m}:\Lambda\rightarrow\mathbb{R}, \; m\in \R,$_by $$\label{funcional}
\mathcal{F}_{m}(\gamma)=\int_{\gamma}(\kappa+m)ds,$$where $s$ stands for the_arclength parameter_of curves_$\gamma\in\Lambda$. In order to obtain the first variation of these actions,_we use the following standard machinery_(see for instance [@Barros])._For a_curve_$\gamma:[0,L]\rightarrow M$, we_take_variations $\Theta=\Theta(t,r):[0,L]\times(-\varepsilon,\varepsilon)\rightarrow
M$_with $\Theta( |
or theoretical computations. Obviously, the proportionality constant $C$ cannot be determined by the fit. A possible scaling of $R_M$ with $Q^2$ is not considered because of its model dependence, and because of the mild dependence of HERMES $R_M$ data on $Q^2$. As discussed below, the proposed functional form of $\tau$, Eq. , is flexible enough to encompass both absorption models, which assume short production times and in-medium hadronization, and energy loss models, which assume long lived quarks with ${\langle t_p \rangle} \gg R_A$, where $R_A$ is the nuclear radius. The 2 classes of models are distinguished by the value of the scaling exponent: a positive $\lambda \gneqq 0$ is characteristic of absorption models, while a negative $\lambda \lesssim
0$ is characteristic of energy loss models. Thus, the exponent $\lambda$ extracted from experimental data can identify the leading mechanism for hadron suppression in nDIS, and distinguish short from long hadronization time scales.
The scaling of $R_M$ is quite natural in the context of hadron absorption models [@Bialas:1986cf; @Accardi:2002tv; @Accardi:2005jd; @Kopeliovich:2003py; @Falter:2004uc; @Gallmeister:2005ad]. Indeed, prehadron absorption depends on the in-medium prehadron path length, which depends on the prehadron production time ${\langle t_p \rangle}$, as long as ${\langle t_p \rangle}
\lesssim R_A$. In the Lund string model [@Andersson:1983ia] hadronization is modeled by the breaking of the color string stretching from the struck parton to the target remnant. The production time is $$\begin{aligned}
{\langle t_p \rangle} & = f(z) (1-z) \frac{z E_q}{\kappa_\text{str}}
\label{eq:lundest}\end{aligned}$$ where $E_q$ is the struck quark energy, and $\kappa_\text{str}$ the string tension. At leading order (LO) in the Strong coupling constant $\alpha_s$, the partonic subprocess is $\gamma^*+q {{\rightarrow}}q$ and one obtains $$\begin{aligned}
E_q = \nu \.\end{ | or theoretical computations. Obviously, the proportionality constant $ C$ cannot be determined by the paroxysm. A potential scaling of $ R_M$ with $ Q^2 $ is not considered because of its model addiction, and because of the mild addiction of HERMES $ R_M$ data on $ Q^2$. As discussed under, the nominate functional kind of $ \tau$, Eq. , is flexible enough to encompass both preoccupation models, which assume unretentive product times and in - medium hadronization, and energy loss models, which bear long lived quarks with $ { \langle t_p \rangle } \gg R_A$, where $ R_A$ is the nuclear radius. The 2 course of models are distinguished by the value of the scaling exponent: a positive $ \lambda \gneqq 0 $ is characteristic of assimilation models, while a negative $ \lambda \lesssim
0 $ is characteristic of energy loss models. therefore, the exponent $ \lambda$ extracted from experimental data can identify the leading mechanism for hadron suppression in nDIS, and identify short from long hadronization time scales.
The scaling of $ R_M$ is quite natural in the context of hadron absorption models [ @Bialas:1986cf; @Accardi:2002tv; @Accardi:2005jd; @Kopeliovich:2003py; @Falter:2004uc; @Gallmeister:2005ad ]. Indeed, prehadron absorption depends on the in - medium prehadron way duration, which depends on the prehadron output meter $ { \langle t_p \rangle}$, as long as $ { \langle t_p \rangle }
\lesssim R_A$. In the Lund string model [ @Andersson:1983ia ] hadronization is model by the breakage of the color string stretching from the strike parton to the target remnant. The production time is $ $ \begin{aligned }
{ \langle t_p \rangle } & = f(z) (1 - omega) \frac{z E_q}{\kappa_\text{str } }
\label{eq: lundest}\end{aligned}$$ where $ E_q$ is the struck quark energy, and $ \kappa_\text{str}$ the string tension. At leading order (LO) in the Strong coupling changeless $ \alpha_s$, the partonic subprocess is $ \gamma^*+q { { \rightarrow}}q$ and one obtains $ $ \begin{aligned }
E_q = \nu \.\end { | or theoretical computationr. Obviously, the proporvionalify constxnt $C$ cannot be determined bb thw fit. A possible scaling of $R_M$ with $E^2$ is not conwudered becehse of lcs moscl de'eidence, and becaose of the mhld dependence ow KERMES $R_M$ data on $Q^2$. As discussed beljw, the lrlposed functiogal gjrm kf $\tau$, Eq. , is flexible enough to endompass both absorptoon models, which assume shlrt oroduction times ajd in-medium hadwinization, ana energy loss models, wgich assume long lived quarks wkth ${\lcngle t_p \rabgoe} \hc R_A$, where $C_A$ is nhe nuclear radius. Tha 2 clasxes of models sre diwtinguished by the vanue of the scaling exponent: a positive $\lambda \tnwqq 0$ hs cvarazrerkstjc or absogptmon models, shile a negqtive $\lambda \lesssik
0$ yw characterisfic of egergy loss models. Thus, the exponent $\lamtda$ extracted from experimwntal data can identivy the lewding mechanism for hadron suppression in nDIS, ang disvivgunwh shuet from long hadronization time scales.
The scaligf pf $R_M$ is quite nctural in the comtfxy of hadron abrorption jodels [@Bialas:1986cf; @Acfardi:2002tv; @Accaedi:2005jd; @Koptliovoch:2003py; @Falter:2004uc; @Gallmeister:2005ae]. Indeed, prekadeon absorption depznds on the nn-mediom prenadron path length, whick depehds on the orehadron oroduction time ${\uannle t_p \rangle}$, as long as ${\langje t_p \ranjle}
\lexssim R_X$. In the Ltnd string model [@Andersson:1983ia] hadrojizatnon iv modeled hy the breaking of the color stcmng stretchinb xrok the stxuck psrton to the earget remnant. The prjductkon time ia $$\begin{eligned}
{\landle t_p \rangle} & = f(z) (1-z) \frac{z E_q}{\kappa_\eext{wtr}}
\label{ed:uundest}\end{aligmed}$$ where $E_q$ is the struck quark enerny, ana $\kappa_\text{str}$ tkt surung tension. At lexdigg ocder (JM) in the Strmng zouoking zonstanu $\aooha_s$, the partonic subprowess is $\gamma^*+q {{\rightartor}}q$ and obe obtaigs $$\begin{alignrd}
E_q = \nu \.\end{ | or theoretical computations. Obviously, the proportionality constant be by the A possible scaling not because of its dependence, and because the mild dependence of HERMES $R_M$ on $Q^2$. As discussed below, the proposed functional form of $\tau$, Eq. , flexible enough to encompass both absorption models, which assume short production times and hadronization, energy models, assume long lived quarks with ${\langle t_p \rangle} \gg R_A$, where $R_A$ is the nuclear radius. 2 classes of models are distinguished by the of the scaling exponent: positive $\lambda \gneqq 0$ is of models, while negative \lesssim is characteristic of loss models. Thus, the exponent $\lambda$ extracted from experimental data can identify the leading mechanism for hadron in nDIS, short from hadronization scales. scaling of $R_M$ natural in the context of hadron @Accardi:2002tv; @Accardi:2005jd; @Kopeliovich:2003py; @Falter:2004uc; @Gallmeister:2005ad]. Indeed, prehadron absorption on the prehadron path length, which depends on prehadron production time ${\langle t_p \rangle}$, as long ${\langle t_p \rangle} \lesssim R_A$. In the Lund string model [@Andersson:1983ia] hadronization is modeled by of the color string from the struck to target The time is {\langle t_p \rangle} & = f(z) (1-z) \frac{z E_q}{\kappa_\text{str}} \label{eq:lundest}\end{aligned}$$ where is the struck quark energy, and $\kappa_\text{str}$ the string tension. order in the Strong constant $\alpha_s$, the partonic is {{\rightarrow}}q$ and one obtains = \.\end{ | or theoretical computations. obviously, tHe proPorTioNaLity ConsTant $C$ cannot be dETermIned by the fit. A possible sCalinG oF $r_M$ wiTH $Q^2$ Is not ConsideREd BECauSe Of Its MoDEl DepenDenCe, and beCause of the MilD dEpendence of HerMeS $R_M$ data on $q^2$. As Discussed belOw, tHe propOsEd fUNctioNal Form oF $\tau$, Eq. , IS flexiBle enough To ENcompaSS both abSORpTion Models, which assume SHoRT production timEs and iN-mEDiUM HadRonIzation, and EnErgy lOSs modelS, WhICH AssUMe long lived quArks with ${\lanGLe t_P \ranglE} \gG R_A$, WHere $R_A$ Is the NuCLeaR radius. The 2 cLassEs of modelS are diSTinguisHEd by the Value oF thE scAlinG ExPoNenT: a POsiTIvE $\laMBda \Gneqq 0$ is cHaRaCteriStic OF ABSorpTioN modEls, whIle a negative $\lAmbDa \leSSsiM
0$ is chAractErisTiC of enErgy loSs modElS. Thus, the exponenT $\lamBda$ extracTed FrOm eXpErimeNTal datA caN idEntify tHe leadiNG meChANISm For hadron suppressiOn IN NDiS, and disTinguiSH sHoRT from lonG hAdrOnizATIon tiMe scALeS.
The scalIng of $R_m$ Is QuIte natuRaL in the CoNteXt oF hadrON absOrptioN models [@BIalas:1986CF; @Accardi:2002tv; @AccaRDi:2005jd; @KopelioviCH:2003pY; @fAlTEr:2004uc; @galLmeister:2005ad]. INdeeD, PrehAdroN AbSorPTion dEpendS oN ThE In-medium prehadron paTh Length, Which Depends on the pRehadron prODUCtion timE ${\lanGLe T_P \rangle}$, as long aS ${\langLe t_p \rangle}
\LEsssim R_A$. in the lund striNg model [@AnDERsson:1983ia] hAdrOniZatIon IS MoDeled by the breAKIng oF tHe color StrIng streTchIng FroM thE sTruck partOn to the tArGeT rEmNanT. The pROduction TiMe iS $$\bEgiN{aligNEd}
{\langLe t_p \rAnglE} & = f(Z) (1-z) \FRac{Z E_q}{\kappA_\TeXT{Str}}
\lAbEl{Eq:luNdeSt}\End{alIgneD}$$ WheRe $E_q$ is tHe struck qUarK EnerGy, AnD $\kappa_\tExt{str}$ the striNg Tension. At lEaDinG order (lo) In the StrOng coupling constant $\alphA_S$, the parTonIc subProcEss is $\gammA^*+q {{\rIghtarRow}}Q$ And one ObtainS $$\begiN{aLigNED}
E_q = \nu \.\END{ | or theoretical computatio ns. Obviou sly,the pr op orti onal ity constant $ C $ ca nnot be determined bythe f it . A p o ss iblescaling of $ R_M $wi th$Q ^ 2$ is n otconside red becaus e o fits model de p en dence, and be cause of the mi ld dep en den c e ofHER MES $ R_M$ d a ta on$Q^2$. As d i scusse d below, t he pro posed functionalf or m of $\tau$, Eq . , is f l ex i b leeno ugh to enc om passb oth abs o rp t i o n m o dels, which a ssume short pro ductio ntim e s andin-me di u m h adronizatio n, a nd energy lossm odels,w hich as sume l ong li vedq ua rk s w it h ${ \ la ngl e t_ p \rangl e} \ gg R_ A$,w h e r e $R _A$ isthe n uclear radius . T he 2 cla ssesof mo dels a re di stingu ished b y the value ofthescaling e xpo ne nt: a posi t ive $\ lam bda \gneqq 0$ isc har ac t e r is tic of absorptionmo d e ls , whilea nega t iv e$ \lambda\l ess sim0 $ is c hara c te ristic o f ener g ylo ss mode ls . Thus ,the ex ponen t $\l ambda$ extract ed fr o m experimental data can iden t if y th e lea din g mechanism for hadr on s u pp res s ion i n nDI S, an d distinguish shortfr om lon g had ronization ti me scales. T he scali ng o f $ R _M$ is quite n atura l in the c o ntext of hadr on absor ption mod e l s [@Bial as: 198 6cf ; @ A c ca rdi:2002tv; @ A c card i: 2005jd; @K opeliov ich :20 03p y;@F alter:200 4uc; @Ga ll me is te r:2 005ad ] . Indeed ,pre ha dro n abs o rption depe ndson t h e i n-mediu m p r e hadr on p athlen gt h, wh ichd epe nds onthe preha dro n pro du ct ion tim e ${\langle t _p \rangle}$ ,aslong a s ${\langl e t_p \rangle}
\lesssim R_A$. I n t he Lu nd s tring mod el[@Ande rss o n:1983 ia] ha droni za tio n is mo d e le d b ythe breaki n g of theco lorstringstretching from th e st ruck parton t o t he t a r ge t r e mn a nt. T h e p r o duction time is $$\begin{ al i gn ed}
{\la n gle t _p \ran gle} &= f(z ) (1-z)\frac{z E _q}{\kapp a_ \tex t { str }}
\labe l{eq:lun dest}\end { align e d} $$ wh ere $E_q$ i s t he st ruck q u ark ener gy, an d$\kapp a_\te xt {str}$ t he string tension. At l eading orde r ( LO) in th e S t ron g couplin g co nstant $\a lph a_s $, th e p a rtoni c su b pr oce s s is$\ga m ma^*+q {{ \ ri ght a r ro w}}q$ and o n e obt ains$$\ b egin{a lign ed}
E_q = \nu \ . \end{ | or_theoretical computations._Obviously, the proportionality constant_$C$ cannot_be_determined by_the_fit. A possible_scaling of $R_M$_with $Q^2$ is not_considered because of_its_model dependence, and because of the mild dependence of HERMES $R_M$ data on $Q^2$._As_discussed below,_the_proposed_functional form of $\tau$, Eq. ,_is flexible enough to encompass_both absorption_models, which assume short production times and in-medium_hadronization,_and energy loss_models, which assume long lived quarks with ${\langle t_p_\rangle} \gg R_A$, where $R_A$ is_the nuclear radius._The_2_classes of models are_distinguished by the value of the_scaling exponent: a positive $\lambda \gneqq_0$ is characteristic of absorption models, while_a negative $\lambda \lesssim
0$ is characteristic_of energy loss models. Thus,_the exponent_$\lambda$ extracted from experimental data_can identify the_leading mechanism_for hadron suppression_in nDIS, and distinguish short from_long hadronization time_scales.
The scaling of $R_M$ is quite_natural_in the context_of_hadron_absorption models_[@Bialas:1986cf; @Accardi:2002tv; @Accardi:2005jd;_@Kopeliovich:2003py;_@Falter:2004uc; @Gallmeister:2005ad]._Indeed,_prehadron absorption depends on the in-medium_prehadron_path length, which depends on the prehadron_production time ${\langle t_p_\rangle}$,_as long as ${\langle_t_p \rangle}
\lesssim R_A$. In the_Lund string model [@Andersson:1983ia] hadronization is_modeled by_the breaking_of the color string stretching from the struck parton to the_target remnant. The production time is_$$\begin{aligned}
{\langle t_p_\rangle} &_=_f(z) (1-z) \frac{z_E_q}{\kappa_\text{str}}_
\label{eq:lundest}\end{aligned}$$_where $E_q$ is the struck quark energy,_and $\kappa_\text{str}$_the string tension. At leading order_(LO) in the Strong_coupling_constant $\alpha_s$, the partonic subprocess is_$\gamma^*+q {{\rightarrow}}q$ and one obtains $$\begin{aligned}
_ E_q = \nu \.\end{ |
k-1} +1$ for $k>1$. Note that the zeroth moment of the force distribution is the participation ratio $\psi$ shown in Fig. \[fig:3\]b. The mass fraction of the tensional backbone that appears at the critical strain is given by the participation ratio or zeroth moment at $\gamma_c$ [@hansen_universality_1989; @bunde_fractals_1995]. In plotting the mass of the tensional structure at the critical strain versus system size $W$, we find that the fractal dimension of this backbone appears to be the same as the euclidean dimension of $2$ (see Fig. \[fig:A8\] in the Appendix).
Of particular interest are the macroscopic properties of fiber networks such as stiffness $K$ near the transition. As we approach the critical point, we find that $K$ shows a finite discontinuity $K_c$, in agreement with prior work [@vermeulen_geometry_2017; @merkel_minimal-length_2019]. Figure \[fig:2\] shows the behavior of one random realization of a diluted triangular network very close to its critical strain $\gamma - \gamma_c \simeq 10^{-4}$. In order to find the sample-specific critical point $\gamma_c(W,i)$ for a network with size $W$, we use the bisection method [@merkel_minimal-length_2019]. By performing an initial step-wise shearing simulation for every random sample, we first find a strain value $\gamma_{R,i}$ at which the network becomes rigid, i.e., the shear stress calculated from Eq. \[eq:3\] reaches a threshold value. Here we use $10^{-9}$ for the stress threshold. Our results, however, are insensitive to the choice of the threshold value as long as we use a sufficiently small value. The prior strain value to $\gamma_{R,i}$ is considered as the nearest floppy point $\gamma_{F,i}$. Modifying the bracket \[$\gamma_{F,i}$, $\gamma_{R,i}$\] in at least $20$ bisection steps, we are able to accurately identify the critical point for every random sample $i$. After identifying the critical point, the network is sheared in a step-wise manner from $\gamma_c(W,i)$. Therefore, the final ensemble averages of a specific system size are taken | k-1 } +1 $ for $ k>1$. Note that the zeroth moment of the force distribution is the participation proportion $ \psi$ usher in Fig. \[fig:3\]b. The mass fraction of the tensional backbone that appear at the critical tune is given by the participation proportion or zeroth moment at $ \gamma_c$ [ @hansen_universality_1989; @bunde_fractals_1995 ]. In plotting the mass of the tensional social organization at the critical breed versus system size $ W$, we witness that the fractal dimension of this backbone appears to be the same as the euclidean dimension of $ 2 $ (attend Fig. \[fig: A8\ ] in the Appendix).
Of particular interest are the macroscopic properties of character networks such as stiffness $ K$ near the transition. As we approach the critical point, we detect that $ K$ shows a finite discontinuity $ K_c$, in agreement with prior work [ @vermeulen_geometry_2017; @merkel_minimal - length_2019 ]. Figure \[fig:2\ ] shows the demeanor of one random realization of a diluted triangular network very close to its critical strain $ \gamma - \gamma_c \simeq 10^{-4}$. In order to find the sample - specific critical distributor point $ \gamma_c(W, i)$ for a network with size $ W$, we use the bisection method [ @merkel_minimal - length_2019 ]. By performing an initial step - wise shearing simulation for every random sample, we first find a stress value $ \gamma_{R, i}$ at which the network becomes fixed, i.e., the shear tension account from Eq. \[eq:3\ ] reaches a threshold value. Here we use $ 10^{-9}$ for the stress threshold. Our results, however, are insensitive to the choice of the brink value as long as we use a sufficiently small value. The prior strain value to $ \gamma_{R, i}$ is considered as the approximate floppy point $ \gamma_{F, i}$. Modifying the bracket \[$\gamma_{F, i}$, $ \gamma_{R, i}$\ ] in at least $ 20 $ bisection steps, we are able to accurately identify the critical distributor point for every random sample $ i$. After identify the critical point, the network is sheared in a step - wise manner from $ \gamma_c(W, i)$. consequently, the final ensemble averages of a specific system size are taken | k-1} +1$ vor $k>1$. Note that the zerouh moment of the yirce dmstribufion is ghe participation ratio $\psi$ dhiwn ib Fig. \[fig:3\]b. The mass fraztion of nhe tensiinal vackbone tizt appears at bhe cxivical strain is given by dhe participathov xatio or zeroth moment at $\gamma_c$ [@hanfen_univrrdality_1989; @bunde_frwctakf_1995]. In ilitting the mass of the tensiohal strlcture at the criyical strain versus system sizf $W$, we find that tje fractal eimegwion of this backbone appears to bg the same as the euclidean dimevsion of $2$ (see Fug. \[dig:W8\] in the Appxndix).
Os particular interesd are tne macroscopic prmpeeties of fiber networns such as stiffnefs $K$ near tke transition. As we appeoach the crigucau pkiit, se finf tiat $K$ shows a finite duscontinuity $K_c$, in sgwvrment with pdior wjrh [@vermeulen_geometry_2017; @merkel_minimal-length_2019]. Fifure \[fig:2\] shows the behavuor of one random reapization jf a diluted triangular network very close to its critmcxl wtvain $\talma - \gamma_c \simeq 10^{-4}$. In order to find the samplq-aptcinic critical poikt $\gamma_c(W,i)$ for a mehwptk with size $W$, we usz tge bisection methof [@merkej_mininal-length_2019]. By lerforming an initial step-wuse shearing wimulation for evexy random saople, we forst find a strain valuz $\gammz_{R,i}$ at whicj the netsurk becomes rigia, i.v., tha shear stress calculated srom Eq. \[eq:3\] reaehes a tfresnold vwlue. Here ae usc $10^{-9}$ for the stress tjreshllg. Our resupts, however, are insensitive to vie choice of jhe thgeshold vclue ax long as we tse a sufficieutly smajl vauue. The prpor straii value to $\gwmma_{R,i}$ is convldered as thx nearest floppy point $\gxoma_{F,i}$. Modifyinb the brabktt \[$\gamma_{F,i}$, $\gamma_{R,i}$\] in at leaxt $20$ gisection steps, we are able to acvurxtejy iventisf the criticdl puing for dvery randon samlle $i$. After identifyhng fhe critical point, tme networj is shewred in a stel-wise manner from $\hamma_r(W,i)$. Thxreforr, tre final ensemble averages of z specifif snstem size arq tajen | k-1} +1$ for $k>1$. Note that the of force distribution the participation ratio The fraction of the backbone that appears the critical strain is given by participation ratio or zeroth moment at $\gamma_c$ [@hansen_universality_1989; @bunde_fractals_1995]. In plotting the mass the tensional structure at the critical strain versus system size $W$, we find the dimension this appears to be the same as the euclidean dimension of $2$ (see Fig. \[fig:A8\] in the Of particular interest are the macroscopic properties of networks such as stiffness near the transition. As we the point, we that shows finite discontinuity $K_c$, agreement with prior work [@vermeulen_geometry_2017; @merkel_minimal-length_2019]. Figure \[fig:2\] shows the behavior of one random realization of a triangular network to its strain - \simeq 10^{-4}$. In find the sample-specific critical point $\gamma_c(W,i)$ with size $W$, we use the bisection method By performing initial step-wise shearing simulation for every sample, we first find a strain value $\gamma_{R,i}$ which the network becomes rigid, i.e., the shear stress calculated from Eq. \[eq:3\] reaches a Here we use $10^{-9}$ the stress threshold. results, are to choice of threshold value as long as we use a sufficiently small value. prior strain value to $\gamma_{R,i}$ is considered as the nearest $\gamma_{F,i}$. the bracket \[$\gamma_{F,i}$, in at least $20$ steps, are able to accurately critical for $i$. identifying critical point, the network sheared in a step-wise manner $\gamma_c(W,i)$. Therefore, the final system size are taken | k-1} +1$ for $k>1$. Note that the zeroth momEnt of the foRce diStrIbuTiOn is The pArticipation raTIo $\psI$ shown in Fig. \[fig:3\]b. The mass FractIoN Of thE TeNsionAl backbONe THAt aPpEaRs aT tHE cRiticAl sTrain is Given by the ParTiCipation ratiO Or Zeroth momeNt aT $\gamma_c$ [@hanseN_unIversaLiTy_1989; @bUNde_frActAls_1995]. In PlottiNG the maSs of the teNsIOnal stRUcture aT THe CritIcal strain versus sYStEM size $W$, we find thAt the fRaCTaL DImeNsiOn of this baCkBone aPPears to BE tHE SAme AS the euclidean Dimension of $2$ (SEe FIg. \[fig:A8\] In The aPpendiX).
Of paRtICulAr interest aRe thE macroscoPic proPErties oF Fiber neTworks SucH as StifFNeSs $k$ neAr THe tRAnSitIOn. AS we approAcH tHe criTicaL POINt, we FinD thaT $K$ shoWs a finite discOntInuiTY $K_c$, In agrEemenT witH pRior wOrk [@verMeuleN_gEometry_2017; @merkel_miNimaL-length_2019]. FiGurE \[fIg:2\] sHoWs the BEhavioR of One Random rEalizatIOn oF a DILUtEd triangular networK vERY cLose to itS critiCAl StRAin $\gamma - \GaMma_C \simEQ 10^{-4}$. in ordEr to FInD the sampLe-specIFiC cRitical PoInt $\gamMa_C(W,i)$ For A netwORk wiTh size $w$, we use thE biseCTion method [@merkEL_minimal-lengtH_2019]. by PERfORminG an Initial step-Wise SHearIng sIMuLatIOn for Every RaNDoM Sample, we first find a sTrAin valUe $\gamMa_{R,i}$ at which thE network beCOMEs rigid, i.E., the SHeAR stress calculaTed frOm Eq. \[eq:3\] reacHEs a thresHold vAlue. Here We use $10^{-9}$ for tHE Stress thResHolD. OuR reSULtS, however, are inSENsitIvE to the cHoiCe of the ThrEshOld ValUe As long as wE use a sufFiCiEnTlY smAll vaLUe. The priOr StrAiN vaLue to $\GAmma_{R,i}$ Is conSideReD aS The Nearest FLoPPY poiNt $\GaMma_{F,I}$. MoDiFying The bRAckEt \[$\gamma_{f,i}$, $\gamma_{R,i}$\] In aT LeasT $20$ bIsEction sTeps, we are able To Accurately IdEntIfy the CRItical poInt for every random sample $I$. after idEntIfyinG the Critical pOinT, the neTwoRK is sheAred in A step-WiSe mANNer frOM $\GaMma_C(W,I)$. Therefore, THE fiNal enSeMble AverageS of a specific system SIze Are taken | k-1} +1$ for $k>1$. Note t hat the ze rothmom ent o f th e fo rce distributi o n is the participation rat io $\ ps i $ sh o wn in F ig. \[f i g: 3 \ ]b. T he ma ss fr actio n o f the t ensional b ack bo ne that appe a rs at the cr iti cal strain i s g iven b ythe parti cip ation ratio or zer oth momen ta t $\ga m ma_c$ [ @ h an sen_ universality_1989 ; @ b unde_fractals_ 1995]. I n p l o tti ngthe mass o fthe t e nsional st r u c tur e at the criti cal strainv ers us sys te m s i ze $W$ , wefi n d t hat the fra ctal dimensio n of t h is back b one app ears t o b e t he s a me a s t he euc l id ean dim ension o f$2 $ (se e Fi g . \ [fig :A8 \] i n the Appendix).
Ofpart i cul ar in teres t ar ethe m acrosc opicpr operties of fib er n etworks s uch a s s ti ffnes s $K$ n ear th e trans ition.A s w ea p p ro ach the critical p oi n t ,we findthat $ K $sh o ws a fin it e d isco n t inuit y $K _ c$ , in agr eement wi th priorwo rk [@v er meu len _geom e try_ 2017;@merkel_ minim a l-length_2019] . Figure \[fig : 2\ ] sh o ws t hebehavior of one rand om r e al iza t ion o f a d il u te d triangular network v ery cl ose t o its critica l strain $ \ g a mma - \g amma _ c\ simeq 10^{-4}$ . Inorder to f i nd the s ample -specifi c critica l point $\ gam ma_ c(W ,i) $ fo r a network w i t h si ze $W$, w e u se thebis ect ion me th od [@merk el_minim al -l en gt h_2 019]. By perfo rm ing a n i nitia l step- wiseshea ri ng sim ulation fo r ever yra ndom sa mp le, w e fi r stfind astrain va lue $\ga mm a_ {R,i}$at which thene twork beco me s r igid,i . e., theshear stress calculated from Eq . \ [eq:3 \] r eaches athr eshold va l ue. He re weuse $ 10 ^{- 9 } $ for t he st re ss thresho l d . O ur re su lts, howeve r, are insensitive tothe choice of th e th r e sh old va l ueas lon g as we use a suf ficientlysm a ll value. Th e pr io r strai n value to $ \ gamma_{ R,i}$ isconsidere das t h e ne arest flop py point $\gamma_ { F,i}$ . M odify ing the b ra cke t \[$ \gamma _ {F, i}$,$\gamm a_ {R,i}$ \] in a t least$20$ bisection steps, w e areabletoaccuratel y i d ent ify the c riti cal pointfor ev ery r and o m sam ple$ i$ . A f ter i dent i fying the cr iti c a lpoint, then e t wor k isshe a red in a s tep-wise manner f r om $\gamma_c(W ,i)$ . The ref o re,th e final ensemb leav e r ages ofaspecific sy stem siz ea re ta ken | k-1} +1$_for $k>1$._Note that the zeroth_moment of_the_force distribution_is_the participation ratio_$\psi$ shown in_Fig. \[fig:3\]b. The mass fraction_of the tensional_backbone_that appears at the critical strain is given by the participation ratio or zeroth_moment_at $\gamma_c$_[@hansen_universality_1989;_@bunde_fractals_1995]._In plotting the mass of_the tensional structure at the_critical strain_versus system size $W$, we find that the_fractal_dimension of this_backbone appears to be the same as the euclidean_dimension of $2$ (see Fig. \[fig:A8\] in_the Appendix).
Of particular_interest_are_the macroscopic properties of_fiber networks such as stiffness $K$_near the transition. As we approach_the critical point, we find that $K$_shows a finite discontinuity $K_c$, in_agreement with prior work [@vermeulen_geometry_2017;_@merkel_minimal-length_2019]. Figure \[fig:2\]_shows the behavior of one_random realization of_a diluted_triangular network very_close to its critical strain $\gamma_- \gamma_c \simeq_10^{-4}$. In order to find the_sample-specific_critical point $\gamma_c(W,i)$_for_a_network with_size $W$, we_use_the bisection_method_[@merkel_minimal-length_2019]. By performing an initial step-wise_shearing_simulation for every random sample, we first_find a strain value_$\gamma_{R,i}$_at which the network_becomes rigid, i.e., the shear_stress calculated from Eq. \[eq:3\] reaches a_threshold value._Here we_use $10^{-9}$ for the stress threshold. Our results, however, are insensitive_to the choice of the threshold_value as long as_we use_a_sufficiently small value._The_prior strain_value to $\gamma_{R,i}$ is considered as the_nearest floppy_point $\gamma_{F,i}$. Modifying the bracket \[$\gamma_{F,i}$,_$\gamma_{R,i}$\] in at least_$20$_bisection steps, we are able to_accurately identify the critical point for_every random sample $i$. After_identifying_the_critical point, the network is_sheared in a step-wise manner from_$\gamma_c(W,i)$. Therefore, the_final ensemble averages of a specific system_size_are taken |
26, 523
Christiansen, J. L., Ballard, S., Charbonneau, D., et al. 2010, ApJ, 710, 97
Cooper, C. S., & Showman, A. P. 2005, ApJ, 629, 45
Coughlin, J. L., & L[ó]{}pez-Morales, M. 2012, AJ, 143, 39
Cowan, N. B., & Agol, E. 2008, ApJ, 678, L129
Cowan, N. B., & Agol, E. 2011, ApJ, 729, 54
Cowan, N. B., Machalek, P., Croll, B., et al. 2012, ApJ, 747, 82
Crossfield, I. J. M., Hansen, B. M. S., Harrington, J., et al. 2010, ApJ, 723, 1436
Deming, D., Knutson, H. A., Kammer, J., et al. 2015, ApJ, 805, 132
Demory, B.-O., de Wit, J., Lewis, N., et al. 2013, ApJ, 776, L25
Diamond-Lowe, H., Stevenson, K. B., Bean, J. L., Line, M. R., & Fortney, J. J. 2014, ApJ, 796, 66
Dobbs-Dixon, I., & Agol, E. 2013, MNRAS, 435, 3159
Dragomir, D., Kane, S. R., Pilyavsky, G., et al. 2011, AJ, 142, 115
Esteves, L. J., De Mooij, E. J. W., & Jayawardhana, R. 2013, ApJ, 772, 51
Esteves, L. J., De Mooij, E. J. W., & Jayawardhana, R. 2015, ApJ, 804, 150
Faigler, S., & Mazeh, T. 2015, ApJ, 800, 73
Fazio, G. G., Hora, J. L., | 26, 523
Christiansen, J. L., Ballard, S., Charbonneau, D., et al. 2010, ApJ, 710, 97
Cooper, C. S., & Showman, A. P. 2005, ApJ, 629, 45
Coughlin, J. L., & L[ó]{}pez - Morales, M. 2012, AJ, 143, 39
Cowan, N. B., & Agol, E. 2008, ApJ, 678, L129
Cowan, N. B., & Agol, E. 2011, ApJ, 729, 54
Cowan, N. B., Machalek, P., Croll, B., et al. 2012, ApJ, 747, 82
Crossfield, I. J. M., Hansen, B. M. S., Harrington, J., et al. 2010, ApJ, 723, 1436
Deming, D., Knutson, H. A., Kammer, J., et al. 2015, ApJ, 805, 132
Demory, B.-O., de Wit, J., Lewis, N., et al. 2013, ApJ, 776, L25
Diamond - Lowe, H., Stevenson, K. B., Bean, J. L., Line, M. R., & Fortney, J. J. 2014, ApJ, 796, 66
Dobbs - Dixon, I., & Agol, E. 2013, MNRAS, 435, 3159
Dragomir, D., Kane, S. R., Pilyavsky, G., et al. 2011, AJ, 142, 115
Esteves, L. J., De Mooij, E. J. W., & Jayawardhana, R. 2013, ApJ, 772, 51
Esteves, L. J., De Mooij, E. J. W., & Jayawardhana, R. 2015, ApJ, 804, 150
Faigler, S., & Mazeh, T. 2015, ApJ, 800, 73
Fazio, G. G., Hora, J. L., | 26, 523
Chgistiansen, J. L., Ballard, S., Gharbonneau, D., et co. 2010, ApJ, 710, 97
Coopsr, C. S., & Sfowman, A. P. 2005, ApJ, 629, 45
Coughlin, J. L., & L[ó]{}pez-Moeales, M. 2012, AJ, 143, 39
Cowan, N. B., & Agol, E. 2008, WpJ, 678, L129
Coqan, I. B., & Agol, E. 2011, ApJ, 729, 54
Cowan, N. N., Macgwlek, '., Croll, B., et al. 2012, ApJ, 747, 82
Crosvfield, I. J. M., Hanvev, Y. M. S., Harrington, J., et al. 2010, ApJ, 723, 1436
Deming, Q., Knutspn, H. A., Kammer, J., ej al. 2015, S[J, 805, 132
Svmiry, B.-O., de Wit, J., Lewis, N., et al. 2013, ApJ, 776, L25
Viamond-Lowe, H., Syevenson, K. B., Bean, J. L., Line, M. G., & Flrtney, J. J. 2014, ApJ, 796, 66
Dohbs-Dixon, I., & Agoj, E. 2013, MNRAS, 435, 3159
Dfagomir, D., Kane, S. R., Pilyzvsky, G., et al. 2011, AJ, 142, 115
Esteves, L. J., De Mooil, E. J. W., & Jayaqaedhwta, R. 2013, ApJ, 772, 51
Xstevef, L. J., De Mooim, E. J. W., & Bayawarchana, R. 2015, ApJ, 804, 150
Naiglxr, S., & Mazeh, T. 2015, ApJ, 800, 73
Fazio, G. G., Hora, J. L., | 26, 523 Christiansen, J. L., Ballard, S., et 2010, ApJ, 97 Cooper, C. 2005, 629, 45 Coughlin, L., & L[ó]{}pez-Morales, 2012, AJ, 143, 39 Cowan, N. & Agol, E. 2008, ApJ, 678, L129 Cowan, N. B., & Agol, E. ApJ, 729, 54 Cowan, N. B., Machalek, P., Croll, B., et al. 2012, 747, Crossfield, J. Hansen, B. M. S., Harrington, J., et al. 2010, ApJ, 723, 1436 Deming, D., Knutson, H. Kammer, J., et al. 2015, ApJ, 805, 132 B.-O., de Wit, J., N., et al. 2013, ApJ, L25 H., Stevenson, B., J. Line, M. R., Fortney, J. J. 2014, ApJ, 796, 66 Dobbs-Dixon, I., & Agol, E. 2013, MNRAS, 435, 3159 Dragomir, Kane, S. G., et 2011, 142, Esteves, L. J., E. J. W., & Jayawardhana, R. 51 Esteves, L. J., De Mooij, E. J. & Jayawardhana, 2015, ApJ, 804, 150 Faigler, S., Mazeh, T. 2015, ApJ, 800, 73 Fazio, G. Hora, J. L., | 26, 523
Christiansen, J. L., Ballard, S., ChaRbonneau, D., eT al. 2010, Apj, 710, 97
CoOpeR, C. s., & ShoWman, a. P. 2005, ApJ, 629, 45
Coughlin, J. l., & l[ó]{}peZ-Morales, M. 2012, AJ, 143, 39
Cowan, N. B., & Agol, e. 2008, ApJ, 678, L129
coWAn, N. B., & aGoL, E. 2011, ApJ, 729, 54
cowan, N. B., mAcHALek, p., CRoLl, B., Et AL. 2012, APJ, 747, 82
CroSsfIeld, I. J. M., hansen, B. M. S., HArrInGton, J., et al. 2010, ApJ, 723, 1436
dEmIng, D., KnutsoN, H. A., kammer, J., et al. 2015, APJ, 805, 132
DEmory, B.-o., dE WiT, j., LewiS, N., eT al. 2013, Apj, 776, L25
DiamONd-Lowe, h., StevensoN, K. b., bean, J. L., lIne, M. R., & FoRTNeY, J. J. 2014, APJ, 796, 66
Dobbs-Dixon, I., & Agol, e. 2013, mNraS, 435, 3159
Dragomir, D., KanE, S. R., PilYaVSkY, g., Et aL. 2011, AJ, 142, 115
esteves, L. J., DE MOoij, E. j. w., & JayawaRDhANA, r. 2013, Apj, 772, 51
esteves, L. J., De MoOij, E. J. W., & JayawARdhAna, R. 2015, Apj, 804, 150
FAigLEr, S., & MazEh, T. 2015, Apj, 800, 73
FAZio, g. G., Hora, J. L., | 26, 523
Christiansen, J.L., Ballar d, S. , C har bo nnea u, D ., et al. 2010 , ApJ , 710, 97
Cooper, C.S., & S h owma n ,A. P. 2005,A pJ , 629 ,45
C ou g hl in, J . L ., & L[ ó]{}pez-Mo ral es , M. 2012, A J ,143, 39
C owa n, N. B., &Ago l, E.20 08, ApJ,678 , L12 9
Cow a n, N.B., & Ago l, E. 201 1 , ApJ,7 2 9, 54
Cowan, N. B., Ma c ha l ek, P., Croll, B., e ta l. 2 012 , A pJ, 747, 8 2
Cros s field,I .J . M., Hansen, B. M. S., Harrin g ton , J.,et a l . 2010 , ApJ ,7 23, 1436
Demi ng,D., Knuts on, H. A., Kam m er, J., et al . 2 015 , Ap J ,80 5,13 2
D e mo ry, B.- O., de W it ,J., L ewis , N . , et al . 20 13, A pJ, 776, L25
Di amon d -Lo we, H ., St even so n, K. B., B ean,J. L., Line, M. R ., & Fortney, J. J . 2 01 4, Ap J , 796, 66
D obbs-Di xon, I. , &Ag o l , E . 2013, MNRAS, 435 ,3 1 59
Dragom ir, D. , K an e , S. R., P ily avsk y , G.,et a l .2011, AJ , 142, 11 5
Esteve s, L. J. ,DeMoo ij, E . J.W., &Jayaward hana, R. 2013, ApJ,7 72, 51
Estev e s, L .J ., D e M ooij, E. J. W., & Ja yawa r dh ana , R. 2 015,Ap J ,8 04, 150
Faigler, S ., & Maz eh, T . 2015, ApJ,800, 73
F a z i o, G. G. , Ho r a, J. L., | 26, 523
Christiansen,_J. L., Ballard,_S., Charbonneau, D., et al._2010, ApJ,_710,_97
Cooper, C. S.,_&_Showman, A. P. 2005,_ApJ, 629, 45
Coughlin,_J. L., & L[ó]{}pez-Morales, M._2012, AJ, 143,_39
Cowan,_N. B., & Agol, E. 2008, ApJ, 678, L129
Cowan, N. B., & Agol, E. 2011, ApJ,_729,_54
Cowan, N. B.,_Machalek,_P.,_Croll, B., et al. 2012, ApJ,_747, 82
Crossfield, I. J. M., Hansen, B. M. S.,_Harrington, J.,_et al. 2010, ApJ, 723, 1436
Deming, D., Knutson,_H. A.,_Kammer, J., et al._2015, ApJ, 805, 132
Demory, B.-O., de Wit, J., Lewis,_N., et al. 2013, ApJ, 776, L25
Diamond-Lowe,_H., Stevenson, K. B.,_Bean,_J. L.,_Line, M. R., & Fortney,_J. J. 2014, ApJ, 796, 66
Dobbs-Dixon, I.,_& Agol, E. 2013, MNRAS, 435,_3159
Dragomir, D., Kane, S. R., Pilyavsky, G., et al._2011, AJ, 142, 115
Esteves, L. J., De_Mooij, E. J. W., & Jayawardhana, R._2013, ApJ,_772, 51
Esteves, L. J., De Mooij,_E. J. W., & Jayawardhana,_R. 2015,_ApJ, 804, 150
Faigler,_S., & Mazeh, T. 2015, ApJ,_800, 73
Fazio, G. G.,_Hora, J. L., |
berson, Su Dong, Tommaso Dorigo, William Dunwoodie, Estia Eichten, Bryan Fulsom, Tim Gershon, Christopher Hill (OSU), Pat Lukens, Namit Mahajan, Gautier Hamel de Monchenault, Jane Nachtman, Stephen Olsen, Fabrizio Palla, Frank Porter, Weiguo Li, Xiang Liu, Xinchou Lou, Rob Roser, Jonathan Rosner, Randy Ruchti, Paraskevas Sphicas, Christopher Thomas, Jim Russ, Joao Varela, Mikhail Voloshin, Hermine Woehri, E. Asli Yetkin, Shilin Zhu as well as CMS quarkonium group for useful discussions. However, all opinions and comments expressed, and any error committed, are solely responsibility of the author himself.
[0]{}
J. Beringer [*et al.*]{} (Particle Data Group), [*Phys. Rev. D*]{} [**86**]{}, 010001 (2012).
F. Abe [*et al*]{}, CDF Collaboration, [*Phys. Rev. Lett.*]{} [**81**]{}, 2432 (1998).
F. Abe [*et al*]{}, CDF Collaboration, [*Phys. Rev. D*]{} [**58**]{}, 112004 (1998).
A. Abulencia [*et al.*]{}, CDF Collaboration, [*Phys. Rev. Lett.*]{} [**96**]{} 082002 (2006).
T. Aaltonen [*et al.*]{}, CDF Collaboration, [*Phys. Rev. Lett.*]{} [**100**]{} 182002 (2008).
V.M. Abazov [*et al.*]{}, D0 Collaboration, [*Phys. Rev. Lett.*]{} [**101**]{} 012001 (2008).
R. L. Jaffe, [*Phys. Rev. D*]{} [**15**]{} 267 (1977).
M. Gell-Mann, [*Phys. Lett.*]{} [**8**]{}, 214 (1964).
G.S. Bali, [*et al.*]{}, [*Phys. Lett. B*]{} [**309**]{}, 378 (1993).
C. J. Morningstar and M. Peardon, [*Phys. Rev. D*]{} [**60**]{}, 034509 (1999).
C. Morningstar and M. Pe | berson, Su Dong, Tommaso Dorigo, William Dunwoodie, Estia Eichten, Bryan Fulsom, Tim Gershon, Christopher Hill (OSU), Pat Lukens, Namit Mahajan, Gautier Hamel de Monchenault, Jane Nachtman, Stephen Olsen, Fabrizio Palla, Frank Porter, Weiguo Li, Xiang Liu, Xinchou Lou, Rob Roser, Jonathan Rosner, Randy Ruchti, Paraskevas Sphicas, Christopher Thomas, Jim Russ, Joao Varela, Mikhail Voloshin, Hermine Woehri, E. Asli Yetkin, Shilin Zhu as well as CMS quarkonium group for useful discussions. However, all opinions and gossip carry, and any error committed, are solely province of the author himself.
[ 0 ] { }
J. Beringer [ * et al. * ] { } (Particle Data Group), [ * Phys. Rev. D * ] { } [ * * 86 * * ] { }, 010001 (2012).
F. Abe [ * et al * ] { }, CDF Collaboration, [ * Phys. Rev. Lett. * ] { } [ * * 81 * * ] { }, 2432 (1998).
F. Abe [ * et al * ] { }, CDF Collaboration, [ * Phys. Rev. D * ] { } [ * * 58 * * ] { }, 112004 (1998).
A. Abulencia [ * et al. * ] { }, CDF Collaboration, [ * Phys. Rev. Lett. * ] { } [ * * 96 * * ] { } 082002 (2006).
T. Aaltonen [ * et al. * ] { }, CDF Collaboration, [ * Phys. Rev. Lett. * ] { } [ * * 100 * * ] { } 182002 (2008).
V.M. Abazov [ * et al. * ] { }, D0 Collaboration, [ * Phys. Rev. Lett. * ] { } [ * * 101 * * ] { } 012001 (2008).
R. L. Jaffe, [ * Phys. Rev. D * ] { } [ * * 15 * * ] { } 267 (1977).
M. Gell - Mann, [ * Phys. Lett. * ] { } [ * * 8 * * ] { }, 214 (1964).
G.S. Bali, [ * et al. * ] { }, [ * Phys. Lett. B * ] { } [ * * 309 * * ] { }, 378 (1993).
C. J. Morningstar and M. Peardon, [ * Phys. Rev. D * ] { } [ * * 60 * * ] { }, 034509 (1999).
C. Morningstar and M. Pe | berdon, Su Dong, Tommaso Dorino, William Dunwoodie, Esvia Eicgten, Bryxn Fulsom, Tim Gershon, Christlpyer Hull (OSU), Pat Lukens, Namkt Mahajaj, Gautiee Hanwl de Moncisnault, Mcne Nzghtmau, Wtephen Olsen, Nabrizio Panla, Frank Portar, Wziguo Li, Xiang Liu, Xinchou Lou, Rob Rjser, Jomahhan Rosner, Ragdy Gushti, Iavaskevas Sphicas, Christopher Thojas, Jim Russ, Joao Vatela, Mikhail Voloshin, Hermlne Aoehri, E. Asli Yetkln, Shilin Zyu af well as CMS quarkonium group for oseful discussions. However, all ooinious and commgurs ffpressed, anv any vrror committcc, are volely tesponsibility of thw author himself.
[0]{}
J. Bermnger [*et al.*]{} (Particle Data Grogp), [*Phys. Rev. D*]{} [**86**]{}, 010001 (2012).
F. Ave [*et an*]{}, CDX Couoaburauioi, [*Pgys. Rev. Petv.*]{} [**81**]{}, 2432 (1998).
F. Abe [*ef al*]{}, CDF Coolaboration, [*Phys. Rev. C*]{} [**58**]{}, 112004 (1998).
A. Abulencia [*st al.*]{}, SDS Collaboration, [*Phys. Rev. Lett.*]{} [**96**]{} 082002 (2006).
T. Aaltoten [*et al.*]{}, CDF Collaboratiob, [*Phys. Rev. Lett.*]{} [**100**]{} 182002 (2008).
V.M. Wbazov [*et al.*]{}, D0 Collaboration, [*Phys. Rev. Lett.*]{} [**101**]{} 012001 (2008).
R. L. Jaffe, [*Phyv. Rev. A*]{} [**15**]{} 267 (1977).
M. Ewlp-Mann, [*Phys. Lett.*]{} [**8**]{}, 214 (1964).
G.S. Bali, [*et al.*]{}, [*Phys. Lett. B*]{} [**309**]{}, 378 (1993).
D. K. Korningstar anb M. Peardon, [*Phys. Rfv. Q*]{} [**60**]{}, 034509 (1999).
C. Morningrtar aus J. Pe | berson, Su Dong, Tommaso Dorigo, William Dunwoodie, Bryan Tim Gershon, Hill (OSU), Pat de Jane Nachtman, Stephen Fabrizio Palla, Frank Weiguo Li, Xiang Liu, Xinchou Lou, Roser, Jonathan Rosner, Randy Ruchti, Paraskevas Sphicas, Christopher Thomas, Jim Russ, Joao Varela, Voloshin, Hermine Woehri, E. Asli Yetkin, Shilin Zhu as well as CMS quarkonium for discussions. all and comments expressed, and any error committed, are solely responsibility of the author himself. [0]{} J. [*et al.*]{} (Particle Data Group), [*Phys. Rev. D*]{} 010001 (2012). F. Abe al*]{}, CDF Collaboration, [*Phys. Rev. [**81**]{}, (1998). F. [*et CDF [*Phys. Rev. D*]{} 112004 (1998). A. Abulencia [*et al.*]{}, CDF Collaboration, [*Phys. Rev. Lett.*]{} [**96**]{} 082002 (2006). T. Aaltonen [*et CDF Collaboration, Lett.*]{} [**100**]{} (2008). Abazov al.*]{}, D0 Collaboration, Lett.*]{} [**101**]{} 012001 (2008). R. L. D*]{} [**15**]{} 267 (1977). M. Gell-Mann, [*Phys. Lett.*]{} 214 (1964). Bali, [*et al.*]{}, [*Phys. Lett. B*]{} 378 (1993). C. J. Morningstar and M. Peardon, Rev. D*]{} [**60**]{}, 034509 (1999). C. Morningstar and M. Pe | berson, Su Dong, Tommaso Dorigo, william DunWoodiE, EsTia eiChteN, BryAn Fulsom, Tim GerSHon, CHristopher Hill (OSU), Pat LuKens, NAmIT MahAJaN, GautIer HameL De mONchEnAuLt, JAnE naChtmaN, StEphen OlSen, FabriziO PaLlA, Frank Porter, wEiGuo Li, Xiang liu, xinchou Lou, RoB RoSer, JonAtHan rOsner, ranDy RucHti, ParASkevas sphicas, ChRiSTopher tHomas, JiM rUsS, JoaO Varela, Mikhail VolOShIN, Hermine Woehri, e. Asli YEtKIn, sHIliN ZhU as well as CmS QuarkONium groUP fOR USefUL discussions. HOwever, all opINioNs and cOmMenTS expreSsed, aNd ANy eRror committEd, arE solely reSponsiBIlity of THe authoR himseLf.
[0]{}
J. berIngeR [*Et Al.*]{} (parTiCLe DATa groUP), [*PhYs. Rev. D*]{} [**86**]{}, 010001 (2012).
F. ABe [*Et Al*]{}, CDF collABORAtioN, [*PhYs. ReV. Lett.*]{} [**81**]{}, 2432 (1998).
f. Abe [*et al*]{}, CDF CoLlaBoraTIon, [*phys. REv. D*]{} [**58**]{}, 112004 (1998).
A. ABuleNcIa [*et aL.*]{}, CDF CoLlaboRaTion, [*Phys. Rev. Lett.*]{} [**96**]{} 082002 (2006).
t. AalTonen [*et al.*]{}, cDF coLlaBoRatioN, [*phys. ReV. LeTt.*]{} [**100**]{} 182002 (2008).
V.m. Abazov [*Et al.*]{}, D0 CoLLabOrATIOn, [*phys. Rev. Lett.*]{} [**101**]{} 012001 (2008).
R. L. Jaffe, [*phYS. reV. D*]{} [**15**]{} 267 (1977).
M. Gell-MAnn, [*PhyS. leTt.*]{} [**8**]{}, 214 (1964).
g.s. Bali, [*et aL.*]{}, [*PHys. lett. b*]{} [**309**]{}, 378 (1993).
c. j. MornIngsTAr And M. PearDon, [*PhyS. reV. D*]{} [**60**]{}, 034509 (1999).
c. MorninGsTar and m. PE | berson, Su Dong, Tommaso D origo, Wil liamDun woo di e, E stia Eichten, Brya n Ful som, Tim Gershon, Chri stoph er Hill (O SU),Pat Luk e ns , Nam it M aha ja n ,Gauti erHamel d e Monchena ult ,Jane Nachtma n ,Stephen Ol sen , Fabrizio P all a, Fra nk Po r ter,Wei guo L i, Xia n g Liu, XinchouLo u , RobR oser, J o n at hanRosner, Randy Ruc h ti , Paraskevas Sp hicas, C h ri s t oph erThomas, Ji mRuss, Joao Va r el a , Mik h ail Voloshin, Hermine Wo e hri , E. A sl i Y e tkin,Shili nZ huas well asCMSquarkoniu m grou p for us e ful dis cussio ns. Ho weve r ,al l o pi n ion s a ndc omm ents exp re ss ed, a nd a n y e rror co mmit ted,are solely re spo nsib i lit y ofthe a utho rhimse lf.
[ 0]{}
J . Beringer [*et al. *]{} (Par tic le Da ta Grou p ), [*P hys . Rev. D *]{} [* * 86* *] { } , 0 10001 (2012).
F.Ab e [* et al*]{ }, CDF Co ll a boration ,[*P hys. R ev. L ett. * ]{ } [**81* *]{},2 43 2(1998).
F. Abe [ *et al *]{}, CDFCollab oration, [*Ph y s. Rev. D*]{}[ **58**]{}, 11 2 00 4 (1 9 98).
A . Abulencia [*e t al. *]{} , C DFC ollab orati on , [ * Phys. Rev. Lett.* ]{ } [**9 6**]{ } 082002 (200 6).
T. Aa l t o nen [*et al. * ]{ } , CDF Collabor ation , [*Phys.Rev. Le tt.*] {} [**10 0**]{} 18 2 0 02 (2008 ).
V. M.Aba z o v[*et al.*]{}, D 0 Co ll aborati on, [*Phys . Rev . Let t. *]{} [**1 01**]{}01 20 01 ( 200 8).
R . L. Jaf fe , [ *P hys . Re v . D*] {} [* *15* *] {} 267 (1977) .
M . Gel l- Ma nn,[*P hy s. L ett. * ]{} [**8** ]{}, 214(19 6 4).
G .S . Bali, [*et al.*]{} ,[*Phys. L et t. B*]{} [ **309**] {}, 378 (1993).
C. J.M ornings tar andM. P eardon, [ *Ph ys. R ev. D*]{} [**60 **]{} ,034 5 0 9 (19 9 9 ).
C .Morningsta r and M. P e | berson, Su_Dong, Tommaso_Dorigo, William Dunwoodie, Estia_Eichten, Bryan_Fulsom,_Tim Gershon,_Christopher_Hill (OSU), Pat_Lukens, Namit Mahajan,_Gautier Hamel de Monchenault,_Jane Nachtman, Stephen_Olsen,_Fabrizio Palla, Frank Porter, Weiguo Li, Xiang Liu, Xinchou Lou, Rob Roser, Jonathan Rosner,_Randy_Ruchti, Paraskevas_Sphicas,_Christopher_Thomas, Jim Russ, Joao Varela,_Mikhail Voloshin, Hermine Woehri, E._Asli Yetkin,_Shilin Zhu as well as CMS quarkonium group_for_useful discussions. However,_all opinions and comments expressed, and any error committed,_are solely responsibility of the author_himself.
[0]{}
J. Beringer [*et_al.*]{}_(Particle_Data Group), [*Phys. Rev._ D*]{} [**86**]{}, 010001 (2012).
F. Abe [*et_al*]{}, CDF Collaboration, [*Phys. Rev. Lett.*]{} [**81**]{}, 2432_(1998).
F. Abe [*et al*]{}, CDF Collaboration, [*Phys. Rev. D*]{}_[**58**]{}, 112004 (1998).
A. Abulencia [*et al.*]{},_CDF Collaboration, [*Phys. Rev. Lett.*]{}_[**96**]{} 082002_(2006).
T. Aaltonen [*et al.*]{}, CDF_Collaboration, [*Phys. Rev._ Lett.*]{} [**100**]{}_182002 (2008).
V.M. Abazov_[*et al.*]{}, D0 Collaboration, [*Phys. Rev._ Lett.*]{} [**101**]{} 012001_(2008).
R. L. Jaffe, [*Phys. Rev. D*]{}_[**15**]{}_267 (1977).
M. Gell-Mann,_[*Phys._ Lett.*]{}_[**8**]{}, 214_(1964).
G.S. Bali, [*et_al.*]{},_[*Phys. Lett._ B*]{}_[**309**]{}, 378 (1993).
C. J. Morningstar and_M._Peardon, [*Phys. Rev. D*]{} [**60**]{}, 034509 (1999).
C._Morningstar and M. Pe |
s in open clusters is indeed similar as in the field, the reason why some CV searches have not been successful is likely because of the relatively small number of cluster members surveyed [e.g. @kafkea04]. A few candidate CVs in open clusters have been identified through dwarf-nova–like outbursts (V57 in in the 2–3 Gyr old NGC2158, [@mochea06]; 15877\_2 in the $\sim$3.5 Gyr old NGC6253 [@demaea10]), or their X-ray spectral properties and UV/optical colors (X2 in NGC6819, @gosnea12). More follow-up is needed to establish the nature and membership status of these candidates. Some of the faint, blue candidate counterparts to [*Chandra*]{} or [*XMM-Newton*]{} sources may also turn out to be CVs (or other compact accreting binaries), although, at least for those in M67 and NGC6791, it is estimated that they are mainly background galaxies.
Active binaries {#sec_abs}
---------------
Active binaries, including both detached systems and contact binaries, are the most common open-cluster X-ray sources. They are among the brightest sources, but found down to the detection limit of the observations ($\sim$2$\times$10$^{28}$ erg s$^{-1}$, 0.3–7 keV, for M67). The study of coronal activity of M67 sources by [@pasqbell] found no relation between the [*ROSAT*]{} X-ray luminosity and stellar parameters like optical magnitude or orbital period. As we now know, this was the result of wrong or incomplete information on orbital or rotation periods, by the inclusion of sources with likely more complex evolutionary histories than regular ABs, and by the high limiting flux of the initial [*ROSAT*]{} pointing (which, as it turns out, was only sensitive enough to detect regular ABs with orbital periods between $\sim$0.5 and 1.5 d). Later studies with larger, and cleaner, AB samples [*do*]{} reveal an activity-rotation relationship in the sense that the coronal X-ray luminosity decreases as a function of orbital period up to the limiting period for tidal circularization [@vdbergea04; @vdbergea12ba]; for M67 this period lies around 12 d. The explanation is that the stellar rotation, and hence the level of activity, is lower in tidally-circularized ( | s in open clusters is indeed similar as in the field, the rationality why some curriculum vitae searches have not been successful is likely because of the relatively humble phone number of cluster members review [ for example @kafkea04 ]. A few candidate CVs in open clusters have been identified through gnome - nova – like outbursts (V57 in in the 2–3 Gyr old NGC2158, [ @mochea06 ]; 15877\_2 in the $ \sim$3.5 Gyr old NGC6253 [ @demaea10 ]), or their X - re spectral properties and UV / optical colors (X2 in NGC6819, @gosnea12). More keep up - up is needed to establish the nature and membership status of these candidate. Some of the faint, blue candidate counterparts to [ * Chandra * ] { } or [ * XMM - Newton * ] { } beginning may also turn out to be CVs (or other compact accrete binaries), although, at least for those in M67 and NGC6791, it is estimated that they are mainly background galaxies.
Active binaries { # sec_abs }
---------------
Active binaries, including both detached systems and contact binaries, are the most common open - cluster X - ray reference. They are among the brightest sources, but discover down to the signal detection limit of the observations ($ \sim$2$\times$10$^{28}$ erg s$^{-1}$, 0.3–7 keV, for M67). The study of coronal activity of M67 sources by [ @pasqbell ] found no relation between the [ * ROSAT * ] { } X - re luminosity and stellar parameters like optical magnitude or orbital period. As we now sleep together, this was the result of wrong or incomplete information on orbital or rotation periods, by the inclusion body of sources with probably more complex evolutionary histories than regular ABs, and by the high limiting flux of the initial [ * ROSAT * ] { } point (which, as it turns out, was only sensitive enough to detect even ABs with orbital periods between $ \sim$0.5 and 1.5 d). Later studies with bombastic, and cleaner, AB sample distribution [ * do * ] { } reveal an activity - rotation kinship in the sense that the coronal X - ray luminosity decreases as a affair of orbital period up to the specify period for tidal circularization [ @vdbergea04; @vdbergea12ba ]; for M67 this period lie around 12 d. The explanation is that the stellar rotation, and therefore the level of activity, is lower in tidally - circularize ( | s ij open clusters is indeea similar as in the fixld, the reason dhy some CV searches have nov bewn suxcessful is likely becxuse of tje relatuvelb small number oh clustev memgcrs sbrteyed [e.g. @kafkea04]. A few cangidate CVs in mpdn clusters have been identified throtgh dwatf-jova–like outbutsts (N57 yn ih the 2–3 Gyr old NGC2158, [@mochea06]; 15877\_2 in the $\aim$3.5 Gyr old NGC6253 [@demara10]), or their X-ray spectral oropfrties and UV/opticwl colors (X2 in GTC6819, @gosnea12). Mofe follow-uk ns needed tk establish the nature and membdrshi' status of tyesf candidates. Some of the fainb, blue wandidaye counterpartx tm [*Cyandra*]{} or [*XMM-Newton*]{} smurces may also tutn out to te CVs (or other compaxt acwrethng cunafiea), eltgough, wt meast for fhose in M67 qnd NGC6791, it is estimstqe that they ade maiglr background galaxies.
Active binaries {#seb_abs}
---------------
Zctive binaries, includibg both detached systgms and cogtact binaries, are the most common open-cluster X-rdy sonrzes. Tmey xee among the brightest sources, but found down tj tne detection limlt of the observatoojs ($\fim$2$\times$10$^{28}$ erg r$^{-1}$, 0.3–7 keV, fod M67). The study of clronal wctivuty of M67 fourves by [@pasqbell] found no reoation betwevn tye [*ROSAT*]{} X-ray lumiuosity and scellar paraketers like optical maguitude or orbital period. Aa we now know, thir wss the result of wrong or insomplete mnforkation un otbital jr rotatioj perlmds, by the inclusiln of smurces witj likely more complex evolutionecy histories jhat rvgular ABf, and by the high jimiting flux pf the nnitiau [*ROSAT*]{} popnting (whmch, as it tuwns out, was otpy sensitive enough eo dwtecr regulxf ABs with orbotal periods betweeb $\sim$0.5 and 1.5 d). Later stodjes with larger, ane cleaner, AB sakplds [*qo*]{} rxveal dn activity-rmtatkon telatkonship in bhe senxe that the coronal F-ray luminosity decreaxef as a fonction os orbital perood up to the limiuing pxriod hor tical circularization [@vdbergea04; @vdbedgea12ba]; fog M67 this period liew around 12 d. Che explanation is that the stellar rotavion, and hence the levgl of activity, is lowgr ln tidally-ciccularyzed ( | s in open clusters is indeed similar the the reason some CV searches likely of the relatively number of cluster surveyed [e.g. @kafkea04]. A few candidate in open clusters have been identified through dwarf-nova–like outbursts (V57 in in the Gyr old NGC2158, [@mochea06]; 15877\_2 in the $\sim$3.5 Gyr old NGC6253 [@demaea10]), or X-ray properties UV/optical (X2 in NGC6819, @gosnea12). More follow-up is needed to establish the nature and membership status of candidates. Some of the faint, blue candidate counterparts [*Chandra*]{} or [*XMM-Newton*]{} sources also turn out to be (or compact accreting although, least those in M67 NGC6791, it is estimated that they are mainly background galaxies. Active binaries {#sec_abs} --------------- Active binaries, including detached systems binaries, are most open-cluster sources. They are brightest sources, but found down to of the observations ($\sim$2$\times$10$^{28}$ erg s$^{-1}$, 0.3–7 keV, M67). The of coronal activity of M67 sources [@pasqbell] found no relation between the [*ROSAT*]{} X-ray and stellar parameters like optical magnitude or orbital period. As we now know, this was of wrong or incomplete on orbital or periods, the of with likely complex evolutionary histories than regular ABs, and by the high limiting of the initial [*ROSAT*]{} pointing (which, as it turns out, sensitive to detect regular with orbital periods between and d). Later studies with cleaner, samples activity-rotation in sense that the coronal luminosity decreases as a function orbital period up to circularization [@vdbergea04; @vdbergea12ba]; for M67 this period lies 12 d. The explanation is that the rotation, and hence the level of activity, is lower in tidally-circularized ( | s in open clusters is indeed siMilar as in tHe fieLd, tHe rEaSon wHy soMe CV searches haVE not Been successful is likely BecauSe OF the RElAtiveLy small NUmBER of ClUsTer MeMBeRs surVeyEd [e.g. @kafKea04]. A few canDidAtE CVs in open clUStErs have beeN idEntified throUgh Dwarf-nOvA–liKE outbUrsTs (V57 in In the 2–3 GYR old NGc2158, [@mochea06]; 15877\_2 in ThE $\Sim$3.5 Gyr OLd NGC6253 [@deMAEa10]), Or thEir X-ray spectral prOPeRTies and UV/opticAl coloRs (x2 In ngc6819, @goSneA12). More folloW-uP is neEDed to esTAbLISH thE Nature and membErship statuS Of tHese caNdIdaTEs. Some Of the FaINt, bLue candidatE couNterparts To [*ChanDRa*]{} or [*XMM-nEwton*]{} soUrces mAy aLso Turn OUt To Be Cvs (OR otHEr ComPAct AccretinG bInAries), AlthOUGH, At leAst For tHose iN M67 and NGC6791, it is eStiMateD ThaT they Are maInly BaCkgroUnd galAxies.
acTive binaries {#sec_Abs}
---------------
ACtive binaRieS, iNclUdIng boTH detacHed SysTems and Contact BInaRiES, ARe The most common open-cLuSTEr x-ray sourCes. TheY ArE aMOng the brIgHteSt soURCes, buT fouND dOwn to the DetectIOn LiMit of thE oBservaTiOns ($\Sim$2$\Times$10$^{28}$ ERg s$^{-1}$, 0.3–7 kEV, for M67). the study Of corONal activity of M67 SOurces by [@pasqbELl] FOUnD No reLatIon between tHe [*ROsaT*]{} X-rAy luMInOsiTY and sTellaR pARaMEters like optical magNiTude or OrbitAl period. As we nOw know, this WAS The resulT of wROnG Or incomplete inFormaTion on orbiTAl or rotaTion pEriods, by The inclusION of sourcEs wIth LikEly MORe Complex evolutIONary HiStories ThaN regulaR ABS, anD by The HiGh limitinG flux of tHe InItIaL [*ROsAT*]{} poINting (whiCh, As iT tUrnS out, wAS only sEnsitIve eNoUgH To dEtect reGUlAR aBs wItH oRbitAl pErIods bEtweEN $\siM$0.5 and 1.5 d). LaTer studieS wiTH larGeR, aNd cleanEr, AB samples [*do*]{} ReVeal an actiViTy-rOtatioN RElationsHip in the sense that the corONal X-ray LumInosiTy deCreases as A fuNction Of oRBital pEriod uP to thE lImiTINg perIOD fOr tIdAl circularIZAtiOn [@vdbErGea04; @vDbergea12Ba]; for M67 this period liES arOund 12 d. The explaNatIon iS THaT thE StELlaR rOTatION, and hence the levEl of activiTy, IS lOwer in tidaLLy-cIrCularizEd ( | s in open clusters is inde ed similar as i n t hefi eld, the reason why so m e CV searches have not bee n suc ce s sful is like ly beca u se o f t he r ela ti v el y sma llnumberof cluster me mb ers surveyed [e .g. @kafke a04 ]. A few can did ate CV sino pen c lus tershave b e en ide ntified t hr o ugh dw a rf-nova – l ik e ou tbursts (V57 in i n t h e 2–3 Gyr oldNGC215 8, [@ m o che a06 ]; 15877\_ 2in th e $\sim$ 3 .5 G y r o l d NGC6253 [@d emaea10]),o r t heir X -r ays pectra l pro pe r tie s and UV/op tica l colors(X2 in NGC6819 , @gosne a12).Mor e f ollo w -u pisne e ded to es t abl ish thena tu re an d me m b e r ship st atus of t hese candidat es. Som e of thefaint , bl ue cand idatecount er parts to [*Chan dra* ]{} or [* XMM -N ewt on *]{}s ources ma y a lso tur n out t o be C V s (o r other compact ac cr e t in g binari es), a l th ou g h, at le as t f or t h o se in M67 an d NGC679 1, iti ses timated t hat th ey ar e m ainly back ground galaxie s.
A c tive binaries{ #sec_abs}
--- - -- - - -- - ----
A ctive binar ies, incl udin g b oth detac hed s ys t em s and contact binari es , arethe m ost common op en-cluster X - ray sour ces. Th e y are among th e bri ghtest sou r ces, but foun d down t o the det e c tion lim itofthe ob s e rv ations ($\sim $ 2 $\ti me s$10$^{ 28} $ erg s $^{ -1} $,0.3 –7 keV, for M67). T he s tu dy of coro n al activ it y o fM67 sour c es by[@pas qbel l] f o und no rel a ti o n bet we en the [* RO SAT*] {} X - ray lumino sity andste l larpa ra meterslike opticalma gnitude or o rbi tal pe r i od. As w e now know, this was th e result of wron g or incomple teinform ati o n on o rbital or r ot ati o n peri o d s, by t he inclusi o n of sour ce s wi th like ly more complex ev o lut ionary histor ies tha n re gul a rA Bs, a n d b y the high limiti ng flux of t h einitial [* R OSA T* ]{} poi nting ( which , as itturns out , was onl ysens i t ive enough to detectregular A B s wit h o rbita l p eriods b etw een $ \sim$0 . 5 a nd 1. 5 d).La ter st udies w ith larg er, and cleaner, AB sam ples [ *do*] {}reveal an ac t ivi ty-rotati on r elationshi p i n t he se nse thatthec or ona l X-ra y lu m inosity d e cr eas e s a s a functio n o f o rbita l p e riod u p to the limiting per i od for tidal c ircu l a riz ati o n [@ vd bergea04; @vdb erg ea 1 2 ba]; for M 67 this per iod lies a r ound12 d.The ex planati o n i s thattheste llar rota tio n, and hen ce t h e leve l of a ctivit y, isl ower i n tidally-circul arize d ( | s in_open clusters_is indeed similar as_in the_field,_the reason_why_some CV searches_have not been_successful is likely because_of the relatively_small_number of cluster members surveyed [e.g. @kafkea04]. A few candidate CVs in open clusters_have_been identified_through_dwarf-nova–like_outbursts (V57 in in the_2–3 Gyr old NGC2158, [@mochea06];_15877\_2 in_the $\sim$3.5 Gyr old NGC6253 [@demaea10]), or their_X-ray_spectral properties and_UV/optical colors (X2 in NGC6819, @gosnea12). More follow-up is_needed to establish the nature and_membership status of_these_candidates._Some of the faint,_blue candidate counterparts to [*Chandra*]{} or_[*XMM-Newton*]{} sources may also turn out_to be CVs (or other compact accreting_binaries), although, at least for those_in M67 and NGC6791, it_is estimated_that they are mainly background_galaxies.
Active binaries {#sec_abs}
---------------
Active_binaries, including_both detached systems_and contact binaries, are the most_common open-cluster X-ray_sources. They are among the brightest_sources,_but found down_to_the_detection limit_of the observations_($\sim$2$\times$10$^{28}$_erg s$^{-1}$,_0.3–7_keV, for M67). The study of_coronal_activity of M67 sources by [@pasqbell] found_no relation between the_[*ROSAT*]{}_X-ray luminosity and stellar_parameters like optical magnitude or_orbital period. As we now know,_this was_the result_of wrong or incomplete information on orbital or rotation periods, by_the inclusion of sources with likely_more complex evolutionary histories_than regular_ABs,_and by the_high_limiting flux_of the initial [*ROSAT*]{} pointing (which, as_it turns_out, was only sensitive enough to_detect regular ABs with_orbital_periods between $\sim$0.5 and 1.5 d)._Later studies with larger, and cleaner,_AB samples [*do*]{} reveal an_activity-rotation_relationship_in the sense that the_coronal X-ray luminosity decreases as a_function of orbital_period up to the limiting period for_tidal_circularization [@vdbergea04; @vdbergea12ba]; for M67 this_period_lies around 12 d. The explanation_is_that_the stellar rotation, and hence_the level of activity, is lower_in tidally-circularized ( |
hypothesis.
Let $x = {\textsc{RMin}}(A[1..n])$ and $y = {\textsc{RMax}}(A[1..n])$ be the indices of the minimum and maximum elements in the array, respectively. If $x\in\{1,n\}$ or $y\in\{1,n\}$ we can compare $A[1]$ and $A[n]$, so assume $x,y\in [2,n-1]$. Without loss of generality we consider the case where $x < y$: the opposite case is symmetric (i.e., replacing ${3\text{-}14\text{-}2}$ with ${2\text{-}41\text{-}3}$), and $x\neq y$ because $n\geq
2$. Consider an arbitrary index $i \in [x,...,y]$, and the result of comparing $A[1]$ to $A[i]$ and $A[i]$ to $A[n]$ (that can be done by the induction hypothesis, as $i\in[2,n-1]$). The result is a partial order on three elements, and is either:
1. One of the two chains $A[1] < A[i] < A[n]$ or $A[n] < A[i] <
A[1]$, in which case we are done since $A[1]$ and $A[n]$ can be compared; or
2. A partial order in which $A[i]$ is the minimum or maximum element, and $A[1]$ is incomparable with $A[n]$.
If we are in the latter case for all $i \in [x,y]$, then let $f(i) =
0$ if $A[i]$ is the minimum element in this partial order, and $f(i) =
1$ otherwise. Because of how $x$ and $y$ were chosen, $f(x) = 0$ and $f(y) = 1$. If we consider the values of $f(i)$ for all $i \in [x,y]$, there must exist two indices $i,i+1 \in [x,y]$ such that $f(i)=0$ and $f(i+1)=1$. Therefore, the indices $1, i, i+1, n | hypothesis.
Let $ x = { \textsc{RMin}}(A[1.. n])$ and $ y = { \textsc{RMax}}(A[1.. n])$ be the indices of the minimum and maximal chemical element in the array, respectively. If $ x\in\{1,n\}$ or $ y\in\{1,n\}$ we can compare $ A[1]$ and $ A[n]$, so assume $ x, y\in [ 2,n-1]$. Without personnel casualty of generalization we consider the case where $ x < y$: the opposite font is symmetric (i.e., replacing $ { 3\text{-}14\text{-}2}$ with $ { 2\text{-}41\text{-}3}$), and $ x\neq y$ because $ n\geq
2$. see an arbitrary index $ i \in [ x,... ,y]$, and the result of compare $ A[1]$ to $ A[i]$ and $ A[i]$ to $ A[n]$ (that can be done by the induction guess, as $ i\in[2,n-1]$). The result is a partial order on three elements, and is either:
1. One of the two chains $ A[1 ] < A[i ] < A[n]$ or $ A[n ] < A[i ] <
A[1]$, in which case we are done since $ A[1]$ and $ A[n]$ can be compare; or
2. A partial order in which $ A[i]$ is the minimum or maximal element, and $ A[1]$ is incomparable with $ A[n]$.
If we are in the latter case for all $ i \in [ x, y]$, then permit $ f(i) =
0 $ if $ A[i]$ is the minimum component in this partial order, and $ f(i) =
1 $ otherwise. Because of how $ x$ and $ y$ were chosen, $ f(x) = 0 $ and $ f(y) = 1$. If we consider the value of $ f(i)$ for all $ i \in [ x, y]$, there must exist two indices $ i, i+1 \in [ x, y]$ such that $ f(i)=0 $ and $ f(i+1)=1$. Therefore, the indices $ 1, i, i+1, n | hyoothesis.
Let $x = {\textsc{RMik}}(A[1..n])$ and $y = {\textse{EMax}}(A[1..n])$ be ths indicer of the minimum and maximum eoemenuf in the array, resoectively. If $x\in\{1,n\}$ or $b\in\{1,n\}$ we can compede $A[1]$ and $A[n]$, sk assbmx $x,y\in [2,n-1]$. Withouj loss of geterality we cotskdzr the case where $x < y$: the opposite sase is sjmmetric (i.e., reklacimd ${3\tesn{-}14\ttxt{-}2}$ with ${2\text{-}41\text{-}3}$), and $x\neq y$ becahse $n\gex
2$. Consider an arbitrary index $i \in [x,...,y]$, ajd tje result of compaging $A[1]$ to $A[u]$ anq $A[i]$ to $A[n]$ (thxt can be done by the jnduction hypothesis, as $i\in[2,n-1]$). Thd resblt is a patcual mrder on thcee elvments, and is either:
1. One of the two chainx $A[1] < A[u] < A[n]$ or $A[n] < A[i] <
A[1]$, in which case we are dmnz since $A[1]$ and $A[n]$ can ve compdred; or
2. Q pxrtjak krder ln xhich $A[i]$ is the minimun or maximum elemenu, age $A[1]$ is incompzrable wyth $A[n]$.
If we are in the latter case for dll $i \in [x,y]$, then let $f(i) =
0$ id $A[i]$ is the minimum epement in this partial order, and $f(i) =
1$ otherwise. Because of vow $x$ xnd $y$ dwrf chosen, $f(x) = 0$ and $f(y) = 1$. If we consider the valtss on $f(i)$ for all $i \ik [x,y]$, there must exosh yro indices $i,i+1 \in [x,y]$ sudh that $f(i)=0$ and $f(i+1)=1$. Hherefote, the indices $1, i, i+1, n | hypothesis. Let $x = {\textsc{RMin}}(A[1..n])$ and $y be indices of minimum and maximum If or $y\in\{1,n\}$ we compare $A[1]$ and so assume $x,y\in [2,n-1]$. Without loss generality we consider the case where $x < y$: the opposite case is (i.e., replacing ${3\text{-}14\text{-}2}$ with ${2\text{-}41\text{-}3}$), and $x\neq y$ because $n\geq 2$. Consider an index \in and result of comparing $A[1]$ to $A[i]$ and $A[i]$ to $A[n]$ (that can be done by the hypothesis, as $i\in[2,n-1]$). The result is a partial on three elements, and either: 1. One of the chains < A[i] A[n]$ $A[n] A[i] < A[1]$, which case we are done since $A[1]$ and $A[n]$ can be compared; or 2. A partial order which $A[i]$ minimum or element, $A[1]$ incomparable with $A[n]$. are in the latter case for [x,y]$, then let $f(i) = 0$ if $A[i]$ the minimum in this partial order, and $f(i) 1$ otherwise. Because of how $x$ and $y$ chosen, $f(x) = 0$ and $f(y) = 1$. If we consider the values of $f(i)$ $i \in [x,y]$, there exist two indices \in such $f(i)=0$ $f(i+1)=1$. Therefore, indices $1, i, i+1, n | hypothesis.
Let $x = {\textsc{RMin}}(A[1..N])$ and $y = {\textsC{RMax}}(a[1..n])$ bE thE iNdicEs of The minimum and mAXimuM elements in the array, resPectiVeLY. If $x\IN\{1,n\}$ Or $y\in\{1,N\}$ we can cOMpARE $A[1]$ aNd $a[n]$, So aSsUMe $X,y\in [2,n-1]$. witHout losS of generalIty We Consider the cASe Where $x < y$: the OppOsite case is sYmmEtric (i.E., rEplACing ${3\tExt{-}14\Text{-}2}$ wIth ${2\texT{-}41\Text{-}3}$), anD $x\neq y$ becAuSE $n\geq
2$. CONsider aN ARbItraRy index $i \in [x,...,y]$, and thE ReSUlt of comparing $a[1]$ to $A[i]$ aNd $a[I]$ tO $a[N]$ (thAt cAn be done by ThE induCTion hypOThESIS, as $I\In[2,n-1]$). The result iS a partial orDEr oN three ElEmeNTs, and iS eithEr:
1. oNe oF the two chaiNs $A[1] < A[I] < A[n]$ or $A[n] < A[i] <
a[1]$, in whiCH case we ARe done sInce $A[1]$ aNd $A[N]$ caN be cOMpArEd; oR
2. A PArtIAl OrdER in Which $A[i]$ iS tHe MinimUm or MAXIMum eLemEnt, aNd $A[1]$ is Incomparable wIth $a[n]$.
If WE arE in thE lattEr caSe For alL $i \in [x,y]$, Then lEt $F(i) =
0$ if $A[i]$ is the miniMum eLement in tHis PaRtiAl Order, ANd $f(i) =
1$ otHerWisE. BecausE of how $x$ ANd $y$ WeRE CHoSen, $f(x) = 0$ and $f(y) = 1$. If we consIdER ThE values oF $f(i)$ for ALl $I \iN [X,y]$, there mUsT exIst tWO IndicEs $i,i+1 \IN [x,Y]$ such thaT $f(i)=0$ and $F(I+1)=1$. THeRefore, tHe IndiceS $1, i, I+1, n | hypothesis.
Let $x = {\t extsc{RMin }}(A[ 1.. n]) $and$y = {\textsc{RMax } }(A[ 1..n])$ be the indices of t he mini m um andmaximum el e m ent sin th ea rr ay, r esp ectivel y. If $x\i n\{ 1, n\}$ or $y\i n \{ 1,n\}$ wecan compare $A[ 1]$ and $ A[ n]$ , so a ssu me $x ,y\in[ 2,n-1] $. Withou tl oss of general i t ywe c onsider the casew he r e $x < y$: the oppos it e c a s e i s s ymmetric ( i. e., r e placing ${ 3 \ t ext { -}14\text{-}2 }$ with ${2 \ tex t{-}41 \t ext { -}3}$) , and $ x \ne q y$ becaus e $n \geq
2$.Consid e r an ar b itraryindex$i\in [x, . .. ,y ]$, a n d t h eres u ltof compa ri ng $A[1 ]$ t o $ A [i]$ an d $A [i]$to $A[n]$ (th atcanb e d one b y the ind uc tionhypoth esis, a s $i\in[2,n-1]$ ). T he result is a pa rt ial o r der on th ree elemen ts, and isei t h e r:
1. One of the t wo c ha ins $A[1 ] < A[ i ]<A [n]$ or$A [n] < A [ i ] <
A[1]$, i n whic h c as e we ar edone s in ce$A[ 1]$ a n d $A [n]$ c an be co mpare d ; or
2. A pa r tial order in wh i c h$ A[i] $ i s the minim um o r max imum el eme n t, an d $A[ 1] $ i s incomparable with$A [n]$.
If w e are in thelatter cas e f or all $ i \i n [ x ,y]$, then let $f(i ) =
0$ if$ A[i]$ is theminimumelement i n this par tia l o rde r,a n d$f(i) =
1$ ot h e rwis e. Becaus e o f how $ x$and $y $ w er e chosen, $f(x) = 0 $an d$f( y) =1 $. If we c ons id erthe v a lues o f $f( i)$fo ra ll$i \in[ x, y ] $, t he re mus t e xi st tw o in d ice s $i,i+ 1 \in [x, y]$ such t ha t $f(i) =0$ and $f(i+ 1) =1$. There fo re, the i n d ices $1, i, i+1, n | hypothesis.
Let_$x =_{\textsc{RMin}}(A[1..n])$ and $y =_{\textsc{RMax}}(A[1..n])$ be_the_indices of_the_minimum and maximum_elements in the_array, respectively. If $x\in\{1,n\}$_or $y\in\{1,n\}$ we_can_compare $A[1]$ and $A[n]$, so assume $x,y\in [2,n-1]$. Without loss of generality we consider_the_case where_$x_<_y$: the opposite case is_symmetric (i.e., replacing ${3\text{-}14\text{-}2}$ with_${2\text{-}41\text{-}3}$), and_$x\neq y$ because $n\geq
2$. Consider an arbitrary index_$i_\in [x,...,y]$, and_the result of comparing $A[1]$ to $A[i]$ and $A[i]$_to $A[n]$ (that can be done_by the induction_hypothesis,_as_$i\in[2,n-1]$). The result is_a partial order on three elements,_and is either:
1. One of_the two chains $A[1] < A[i] <_A[n]$ or $A[n] < A[i] <
_ _ _A[1]$, in which case we_are done since_$A[1]$ and_$A[n]$ can be_compared; or
2. A partial order_in which $A[i]$_is the minimum or maximum element,_and_$A[1]$ is incomparable_with_$A[n]$.
If_we are_in the latter_case_for all_$i_\in [x,y]$, then let $f(i) =
0$_if_$A[i]$ is the minimum element in this_partial order, and $f(i)_=
1$_otherwise. Because of how_$x$ and $y$ were chosen,_$f(x) = 0$ and $f(y) =_1$. If_we consider_the values of $f(i)$ for all $i \in [x,y]$, there must_exist two indices $i,i+1 \in [x,y]$_such that $f(i)=0$ and_$f(i+1)=1$. Therefore,_the_indices $1, i,_i+1,_n |
In [@N-Z Problem 5.1] it was asked whether the combinatorial $(2k-1)$-spheres $\Delta^{2k-1}_n$ are shellable for all $k\geq 2$ and $n\geq2k$. Here we answer this question in the $3$-dimensional case: we verify that the spheres $\Delta^3_n$ are shellable, by showing that they possess a symmetric shelling. A shelling order of a cs simplicial complex is called [*symmetric*]{} if it is the form $(F_1, F_2,\dots, F_m, -F_m, -F_{m-1},\dots,-F_1)$.
\[thm:shellable\] Let $n\geq 4$. There exists a symmetric shelling order of $\Delta^{3}_n$.
Our strategy is as follows: use equation (\[eq:decomposition\]) to separate one half of the facets of $\Delta^{3}_n$ in the $n-2$ blocks described below; list these blocks in the shelling we are about to construct in the following order:
- The facets of $B^{3,1}_{n}$.
- The facets of $\partial B^{3,1}_{k-1}*k$ that are not in $\pm B^{3,1}_{k}$. Here $5\leq k\leq n$. These $n-4$ blocks will be listed in the decreasing order of $k$, i.e., from $k=n$ to $k=5$;
- Three of the six facets of $\Delta^{3}_4\backslash\pm B^{3,1}_4$ (which three will be specified later).
We will now discuss the order inside each of these blocks. Then we will list the other half of the facets to make the ordering symmetric.
The ball $B^{3,1}_n$ is a stacked ball, hence shellable. We list its facets in any shelling order of $B^{3,1}_n$. Now, for $k=n, n-1, \dots, 5$, the facets of $\partial B^{3,1}_{k-1}*k$ that are not in $\pm B^{3,1}_{k}$ are described in Lemma \[lm: facets\]: they consist of $$F_{k,1}:=(-k+3)(- | In [ @N - Z Problem 5.1 ] it was asked whether the combinatorial $ (2k-1)$-spheres $ \Delta^{2k-1}_n$ are shellable for all $ k\geq 2 $ and $ n\geq2k$. Here we answer this question in the $ 3$-dimensional case: we control that the sphere $ \Delta^3_n$ are shellable, by showing that they possess a symmetric barrage. A shelling order of a cs simplicial complex is call [ * symmetric * ] { } if it is the form $ (F_1, F_2,\dots, F_m, -F_m, -F_{m-1},\dots,-F_1)$.
\[thm: shellable\ ] permit $ n\geq 4$. There exists a symmetric shell order of $ \Delta^{3}_n$.
Our strategy is as follow: use equation (\[eq: decomposition\ ]) to separate one half of the facets of $ \Delta^{3}_n$ in the $ n-2 $ blocks describe below; list these blocks in the barrage we are about to construct in the following order:
- The aspect of $ B^{3,1}_{n}$.
- The facets of $ \partial B^{3,1}_{k-1}*k$ that are not in $ \pm B^{3,1}_{k}$. Here $ 5\leq k\leq n$. These $ n-4 $ block will be listed in the decreasing order of $ k$, i.e., from $ k = n$ to $ k=5 $;
- Three of the six facets of $ \Delta^{3}_4\backslash\pm B^{3,1}_4 $ (which three will be assign later).
We will now discuss the order inside each of these blocks. Then we will list the other half of the facets to make the ordering symmetric.
The ball $ B^{3,1}_n$ is a stacked ball, hence shellable. We list its facets in any blast order of $ B^{3,1}_n$. Now, for $ thousand = n, n-1, \dots, 5 $, the facet of $ \partial B^{3,1}_{k-1}*k$ that are not in $ \pm B^{3,1}_{k}$ are described in Lemma \[lm: facets\ ]: they consist of $ $ F_{k,1}:=(-k+3)(- | In [@J-Z Problem 5.1] it was asked whether the combinatocial $(2k-1)$-slheres $\Ddlta^{2k-1}_n$ are shellable for all $k\teq 2$ qnd $n\geq2k$. Here we answdr this qlestion ib tht $3$-dimensional casx: we verlyy thzb the wpheres $\Delta^3_n$ are shelldble, by showinc ghct they possess a symmetric shelling. A shelkijg order of a ss spm[licjal complex is called [*symmetric*]{} if it is uhe form $(F_1, F_2,\dots, F_k, -F_m, -F_{m-1},\dots,-F_1)$.
\[thm:shellable\] Lft $n\heq 4$. There exists w symmetric shejoing order ow $\Delta^{3}_n$.
Oug strategy js as follows: use equation (\[eq:dezompoxition\]) to weparwje one half if thv facets of $\Dckta^{3}_n$ it the $n-2$ blocks descrined bxlow; list these blocks in the shelling we ate about tm eonstruct in the folliwung otder:
- Thd fazeta pf $B^{3,1}_{n}$.
- Hhe facets of $\partial B^{3,1}_{k-1}*j$ that are not in $\pk F^{3,1}_{j}$. Here $5\leq k\lsq n$. Trefe $n-4$ blocks will be listed in the decredsihg order of $k$, i.e., from $k=b$ to $k=5$;
- Three of the six facees of $\Delta^{3}_4\backslash\pm B^{3,1}_4$ (which three will be spechfied uattr).
Qe wiuo jow discuss the order inside each of these bljdkx. Nhen we will list the other napf jf the facets to make fhe ordering symmehric.
The ball $B^{3,1}_n$ is a ftacled ball, hence shellable. We list its fabets in any shelling oxder of $B^{3,1}_n$. Nuw, fpr $k=n, n-1, \dots, 5$, the facets of $\'artiam B^{3,1}_{k-1}*k$ that wre not ih $\pm B^{3,1}_{k}$ are descrkbec hn Lemma \[lm: facets\]: they confist of $$F_{j,1}:=(-k+3)(- | In [@N-Z Problem 5.1] it was asked combinatorial $\Delta^{2k-1}_n$ are for all $k\geq answer question in the case: we verify the spheres $\Delta^3_n$ are shellable, by that they possess a symmetric shelling. A shelling order of a cs simplicial is called [*symmetric*]{} if it is the form $(F_1, F_2,\dots, F_m, -F_m, -F_{m-1},\dots,-F_1)$. Let 4$. exists symmetric shelling order of $\Delta^{3}_n$. Our strategy is as follows: use equation (\[eq:decomposition\]) to separate one of the facets of $\Delta^{3}_n$ in the $n-2$ described below; list these in the shelling we are to in the order: The of $B^{3,1}_{n}$. - facets of $\partial B^{3,1}_{k-1}*k$ that are not in $\pm B^{3,1}_{k}$. Here $5\leq k\leq n$. These $n-4$ blocks be listed decreasing order $k$, from to $k=5$; - the six facets of $\Delta^{3}_4\backslash\pm B^{3,1}_4$ be specified later). We will now discuss the inside each these blocks. Then we will list other half of the facets to make the symmetric. The ball $B^{3,1}_n$ is a stacked ball, hence shellable. We list its facets in order of $B^{3,1}_n$. Now, $k=n, n-1, \dots, the of B^{3,1}_{k-1}*k$ are not $\pm B^{3,1}_{k}$ are described in Lemma \[lm: facets\]: they consist of | In [@N-Z Problem 5.1] it was asked whetHer the combInatoRiaL $(2k-1)$-sPhEres $\deltA^{2k-1}_n$ are shellablE For aLl $k\geq 2$ and $n\geq2k$. Here we anSwer tHiS QuesTIoN in thE $3$-dimensIOnAL CasE: wE vEriFy THaT the sPheRes $\DeltA^3_n$ are shellAblE, bY showing that THeY possess a sYmmEtric shellinG. A sHellinG oRdeR Of a cs SimPliciAl compLEx is caLled [*symmeTrIC*]{} if it iS The form $(f_1, f_2,\DoTs, F_m, -f_m, -F_{m-1},\dots,-F_1)$.
\[thm:shellABlE\] let $n\geq 4$. There exIsts a sYmMEtRIC shEllIng order of $\deLta^{3}_n$.
OUR strateGY iS AS FolLOws: use equatioN (\[eq:decomposITioN\]) to sepArAte ONe half Of the FaCEts Of $\Delta^{3}_n$ in tHe $n-2$ bLocks descRibed bELow; list THese bloCks in tHe sHelLing WE aRe AboUt TO coNStRucT In tHe followInG oRder:
- THe faCETS Of $B^{3,1}_{n}$.
- the FaceTs of $\pArtial B^{3,1}_{k-1}*k$ that Are Not iN $\Pm B^{3,1}_{K}$. Here $5\Leq k\lEq n$. THeSe $n-4$ blOcks wiLl be lIsTed in the decreasIng oRder of $k$, i.e., FroM $k=N$ to $K=5$;
- THree oF The six FacEts Of $\Delta^{3}_4\BackslaSH\pm b^{3,1}_4$ (wHICH tHree will be specifieD lATEr).
we will noW discuSS tHe ORder insiDe EacH of tHESe bloCks. THEn We will liSt the oTHeR hAlf of thE fAcets tO mAke The OrderINg syMmetriC.
The ball $b^{3,1}_n$ is a STacked ball, hencE Shellable. We liST iTS FaCEts iN anY shelling orDer oF $b^{3,1}_n$. NoW, for $K=N, n-1, \DotS, 5$, The faCets oF $\pARtIAl B^{3,1}_{k-1}*k$ that are not in $\pm b^{3,1}_{k}$ Are desCribeD in Lemma \[lm: facEts\]: they conSIST of $$F_{k,1}:=(-k+3)(- | In [@N-Z Problem 5.1] it w as asked w hethe r t heco mbin ator ial $(2k-1)$-s p here s $\Delta^{2k-1}_n$ ar e she ll a blef or all$k\geq2 $a n d $ n\ ge q2k $. He re we an swer th is questio n i nthe $3$-dime n si onal case: we verify that th e sphe re s $ \ Delta ^3_ n$ ar e shel l able,by showin gt hat th e y posse s s a sym metric shelling.A s h elling order o f a cs s i mp l i cia l c omplex isca lled[ *symmet r ic * ] { } i f it is the fo rm $(F_1, F _ 2,\ dots,F_ m,- F_m, - F_{m- 1} , \do ts,-F_1)$.
\[t hm:shella ble\]L et $n\g e q 4$. T here e xis tsa sy m me tr icsh e lli n gord e r o f $\Delt a^ {3 }_n$.
Ou r s t rate gyis a s fol lows: use equ ati on ( \ [eq :deco mposi tion \] ) tosepara te on ehalf of the fac etsof $\Delt a^{ 3} _n$ i n the $n-2$blo cks descri bed bel o w;li s t th ese blocks in thesh e l li ng we ar e abou t t oc onstruct i n t he f o l lowin g or d er :
- T he fac e ts o f $B^{3 ,1 }_{n}$ .
- T he fa c etsof $\p artial B ^{3,1 } _{k-1}*k$ that are not in $\ p mB ^ {3 , 1}_{ k}$ . Here $5\l eq k \ leqn$.T he se$ n-4$block sw il l be listed in the d ec reasin g ord er of $k$, i. e., from $ k = n $ to $k= 5$;- Three of thesix f acets of $ \ Delta^{3 }_4\b ackslash \pm B^{3, 1 } _4$ (whi chthr eewil l be specified la t e r).
W e willnow discus s t heord erin side each of thes ebl oc ks . T hen w e will li st th eoth er ha l f of t he fa cets t om ake the or d er i n g sy mm et ric.
T he ball $B^ { 3,1 }_n$ is a stacke d b a ll,he nc e shell able. We list i ts facetsin an y shel l i ng order of $B^{3,1}_n$. Now, f o r $k=n, n- 1, \d ots, 5$, thefac ets of $\ p artial B^{3, 1}_{k -1 }*k $ thata r enot i n $\pm B^{ 3 , 1}_ {k}$ar e de scribed in Lemma \[lm: fa c ets \]: they cons ist of$ $ F_ {k, 1 }: = (-k +3 ) (- | In [@N-Z_Problem 5.1]_it was asked whether_the combinatorial_$(2k-1)$-spheres_$\Delta^{2k-1}_n$ are_shellable_for all $k\geq_2$ and $n\geq2k$._Here we answer this_question in the_$3$-dimensional_case: we verify that the spheres $\Delta^3_n$ are shellable, by showing that they possess_a_symmetric shelling._A_shelling_order of a cs simplicial_complex is called [*symmetric*]{} if_it is_the form $(F_1, F_2,\dots, F_m, -F_m, -F_{m-1},\dots,-F_1)$.
\[thm:shellable\] Let_$n\geq_4$. There exists_a symmetric shelling order of $\Delta^{3}_n$.
Our strategy is as_follows: use equation (\[eq:decomposition\]) to separate_one half of_the_facets_of $\Delta^{3}_n$ in the_$n-2$ blocks described below; list these_blocks in the shelling we are_about to construct in the following order:
-_ The facets of $B^{3,1}_{n}$.
-_ The facets of_$\partial B^{3,1}_{k-1}*k$_that are not in $\pm_B^{3,1}_{k}$. Here $5\leq_k\leq n$._These $n-4$ blocks_will be listed in the decreasing_order of $k$,_i.e., from $k=n$ to $k=5$;
- __Three of the_six_facets_of $\Delta^{3}_4\backslash\pm_B^{3,1}_4$ (which three_will_be specified_later).
We_will now discuss the order inside_each_of these blocks. Then we will list_the other half of_the_facets to make the_ordering symmetric.
The ball $B^{3,1}_n$ is_a stacked ball, hence shellable. We_list its_facets in_any shelling order of $B^{3,1}_n$. Now, for $k=n, n-1, \dots, 5$,_the facets of $\partial B^{3,1}_{k-1}*k$ that_are not in $\pm_B^{3,1}_{k}$ are_described_in Lemma \[lm: facets\]:_they_consist of_$$F_{k,1}:=(-k+3)(- |
)}
- \beta_{i,j} n^{(i)} n^{(j)} + h_{i j}{}' n^{(i)} n^{(j)} \notag\\
& \quad
- \alpha \Theta' - (\beta^i + h^i{}_j n^{(j)}) \Theta_{, i}
- \Bigl[ \alpha_{,i} - (\beta_{j,i} - h_{i j}{}') n^{(j)}
+ (h_{k i, j} - h_{j k, i}) n^{(j)} n^{(k)} \Bigr] D^i \Theta \notag\\
& \quad
- \Bigl( \alpha \alpha_{,i} + \alpha_{, j} h^j{}_i \Bigr) n^{(i)}
- \Bigl[ \alpha (- \beta_{i,j} + h_{i j}{}') + \beta^k h_{i j, k}
+ (\beta_{k, i} - \beta_{i, k} + 2 h_{i k}{}') h^k{}_j \Bigr] n^{(i)} n^{(j)}
\notag\\
& \quad
- \Theta n^{(i)} \Bigl( - \alpha_{,i} + \beta_{i,j} n^{(j)} - h_{i j}{}' n^{(j)} \Bigr)
= {{\cal C}}^\Theta \,.\end{aligned}$$
Temperature and spectral distortion of collision terms {#ssec:Collision}
------------------------------------------------------
The expression of the collision term has been derived by taking into account only the intensity in [@Dodelson1993; @Bartolo:2006cu] and then it was extended to include the effect of polarisation in [@Pitrou:2008hy; @Pitrou:2008ut; @Beneke:2010eg]. Here we summarize the result obtained by Beneke et al. [@Beneke:2010eg] applying the decomposition of the distribution function into intensity and linear polarisation.
The complete expression of the collision term for intensity is given by $$\begin{aligned}
{{\cal C}}^\Theta
&= a \, \bar{n}_e \sigma_T \left( - \Theta + \langle \Theta \rangle
- \frac{3}{4} S^{(i) (j)} \ | ) }
- \beta_{i, j } n^{(i) } n^{(j) } + h_{i j } { }' n^{(i) } n^{(j) } \notag\\
& \quad
- \alpha \Theta' - (\beta^i + h^i{}_j n^{(j) }) \Theta _ {, i }
- \Bigl [ \alpha_{,i } - (\beta_{j, i } - h_{i j } { }') n^{(j) }
+ (h_{k i, j } - h_{j k, i }) n^{(j) } n^{(k) } \Bigr ] D^i \Theta \notag\\
& \quad
- \Bigl (\alpha \alpha_{,i } + \alpha _ {, j } h^j{}_i \Bigr) n^{(i) }
- \Bigl [ \alpha (- \beta_{i, j } + h_{i j } { }') + \beta^k h_{i j, k }
+ (\beta_{k, i } - \beta_{i, k } + 2 h_{i k } { }') h^k{}_j \Bigr ] n^{(i) } n^{(j) }
\notag\\
& \quad
- \Theta n^{(i) } \Bigl (- \alpha_{,i } + \beta_{i, j } n^{(j) } - h_{i j } { }' n^{(j) } \Bigr)
= { { \cal C}}^\Theta \,.\end{aligned}$$
Temperature and apparitional distorted shape of collision terms { # ssec: Collision }
------------------------------------------------------
The expression of the collision condition has been derive by taking into account entirely the saturation in [ @Dodelson1993; @Bartolo:2006cu ] and then it was extended to admit the consequence of polarisation in [ @Pitrou:2008hy; @Pitrou:2008ut; @Beneke:2010eg ]. Here we sum up the result obtained by Beneke et al. [ @Beneke:2010eg ] applying the decomposition of the distribution affair into intensity and linear polarisation.
The complete expression of the collision condition for intensity is given by $ $ \begin{aligned }
{ { \cal C}}^\Theta
& = a \, \bar{n}_e \sigma_T \left (- \Theta + \langle \Theta \rangle
- \frac{3}{4 } S^{(i) (j) } \ | )}
- \bfta_{i,j} n^{(i)} n^{(j)} + h_{i j}{}' n^{(i)} n^{(j)} \kotag\\
& \quad
- \alpkq \Thete' - (\beta^j + h^i{}_j n^{(g)}) \Theta_{, i}
- \Bigl[ \alpha_{,i} - (\beta_{o,i} - y_{i j}{}') b^{(j)}
+ (h_{k i, j} - h_{j k, i}) n^{(j)} v^{(k)} \Bigr] D^p \Theta \nitag\\
& \quad
- \Bigl( \alpha \alpha_{,i} + \alpka_{, j} h^j{}_i \Bigr) n^{(i)}
- \Bigl[ \alpva (- \beta_{i,j} + h_{i j}{}') + \yeta^k h_{i j, k}
+ (\beta_{k, i} - \beta_{i, k} + 2 h_{i h}{}') h^k{}_j \Bogg] n^{(i)} n^{(j)}
\notag\\
& \quac
- \Thsna n^{(i)} \Bigl( - \alpha_{,i} + \beta_{i,j} n^{(j)} - g_{i j}{}' n^{(j)} \Bigr)
= {{\cal C}}^\Yheta \,.\end{aligned}$$
Temperaturf anf spectral distortlon of collusiog terms {#ssec:Cullision}
------------------------------------------------------
Tht zxpression kf the collision term has been aerivzd by takint untl account onoy thv intensity ik [@Codelsmn1993; @Bartplo:2006cu] and then it waw extended to include the effect of polwrisation iu [@Pitrou:2008hy; @Pitrou:2008ut; @Bebeje:2010eg]. Vere we rymmxriee vhe resulh outained by Geneke et ao. [@Beneke:2010eg] applying trv decompositikn of ehq distribution function into intensity dnd linear polarisation.
The complete expression lf the cojlision term for intensity is given by $$\begin{alignad}
{{\cam C}}^\Tkcba
&= x \, \har{n}_e \sigma_T \left( - \Theta + \langle \Theta \rangle
- \ftab{3}{4} S^{(i) (j)} \ | )} - \beta_{i,j} n^{(i)} n^{(j)} + h_{i n^{(j)} & \quad \alpha \Theta' - i} \Bigl[ \alpha_{,i} - - h_{i j}{}') + (h_{k i, j} - h_{j i}) n^{(j)} n^{(k)} \Bigr] D^i \Theta \notag\\ & \quad - \Bigl( \alpha \alpha_{,i} \alpha_{, j} h^j{}_i \Bigr) n^{(i)} - \Bigl[ \alpha (- \beta_{i,j} + h_{i j}{}') \beta^k j, + i} - \beta_{i, k} + 2 h_{i k}{}') h^k{}_j \Bigr] n^{(i)} n^{(j)} \notag\\ & \quad - n^{(i)} \Bigl( - \alpha_{,i} + \beta_{i,j} n^{(j)} - j}{}' n^{(j)} \Bigr) = C}}^\Theta \,.\end{aligned}$$ Temperature and spectral of terms {#ssec:Collision} The of collision term has derived by taking into account only the intensity in [@Dodelson1993; @Bartolo:2006cu] and then it was extended to the effect in [@Pitrou:2008hy; @Beneke:2010eg]. we the result obtained et al. [@Beneke:2010eg] applying the decomposition function into intensity and linear polarisation. The complete of the term for intensity is given by {{\cal C}}^\Theta &= a \, \bar{n}_e \sigma_T \left( \Theta + \langle \Theta \rangle - \frac{3}{4} S^{(i) (j)} \ | )}
- \beta_{i,j} n^{(i)} n^{(j)} + h_{i j}{}' n^{(i)} n^{(j)} \notag\\
& \quaD
- \alpha \ThetA' - (\beta^I + h^i{}_J n^{(j)}) \thEta_{, i}
- \bigl[ \Alpha_{,i} - (\beta_{j,i} - h_{i J}{}') N^{(j)}
+ (h_{k I, j} - h_{j k, i}) n^{(j)} n^{(k)} \Bigr] D^i \Theta \nOtag\\
& \qUaD
- \bigl( \ALpHa \alpHa_{,i} + \alphA_{, J} h^J{}_I \bigR) n^{(I)}
- \BIgl[ \AlPHa (- \Beta_{i,J} + h_{i J}{}') + \beta^k h_{I j, k}
+ (\beta_{k, i} - \bEta_{I, k} + 2 H_{i k}{}') h^k{}_j \Bigr] n^{(i)} N^{(J)}
\nOtag\\
& \quad
- \ThEta N^{(i)} \Bigl( - \alpha_{,i} + \BetA_{i,j} n^{(j)} - h_{I j}{}' N^{(j)} \BIGr)
= {{\cal c}}^\ThEta \,.\enD{alignED}$$
TempeRature and SpECtral dIStortioN OF cOlliSion terms {#ssec:CollISiON}
------------------------------------------------------
The expression Of the cOlLIsION teRm hAs been deriVeD by taKIng into ACcOUNT onLY the intensity In [@Dodelson1993; @BARtoLo:2006cu] anD tHen IT was exTendeD tO IncLude the effeCt of PolarisatIon in [@PITrou:2008hy; @PITrou:2008ut; @BEneke:2010eG]. HeRe wE sumMArIzE thE rESulT ObTaiNEd bY Beneke eT aL. [@BEneke:2010Eg] apPLYINg thE deCompOsitiOn of the distriButIon fUNctIon inTo intEnsiTy And liNear poLarisAtIon.
The complete eXpreSsion of thE coLlIsiOn Term fOR intenSitY is Given by $$\Begin{alIGneD}
{{\cAL c}}^\thEta
&= a \, \bar{n}_e \sigma_T \lefT( - \THETa + \Langle \ThEta \ranGLe
- \FrAC{3}{4} S^{(i) (j)} \ | )}
- \beta_{i,j} n^{(i)}n^{(j)} +h_{ij}{ }'n^ {(i) } n^ {(j)} \notag\\ & \ quad
- \alpha \Theta' - (\ be t a^i+ h ^i{}_ j n^{(j ) }) \ The ta _{ , i }-\Bigl [ \ alpha_{ ,i} - (\be ta_ {j ,i} - h_{i j } {} ') n^{(j)}
+ (h_{k i, j} -h_{j k ,i}) n^{(j )}n^{(k )} \Bi g r] D^i \Theta \ no t ag\\
& \quad- \ Bigl ( \alpha \alpha_{ , i} + \alpha_{, j} h^j{} _i \B i g r)n^{ (i)}
- \B ig l[ \a l pha (-\ be t a _ {i, j } + h_{i j}{} ') + \beta^ k h_ {i j,k}
+ (\beta _{k,i} - \ beta_{i, k} + 2 h_{i k}{ }') h^ k {}_j \B i gr] n^{ (i)} n ^{( j)}
\n o ta g\ \
&\ qua d
- \ T het a n^{(i) }\B igl(- \a l p h a _{,i } + \be ta_{i ,j} n^{(j)} - h_ {i j } {}' n^{( j)} \ Bigr )
= { {\calC}}^\ Th eta \,.\end{ali gned }$$
Temp era tu rean d spe c tral d ist ort ion ofcollisi o n t er m s {# ssec:Collision}
-- -- - - -- -------- ------ - -- -- - -------- -- --- ---- - - ----- --
T he express ion of th ecollisi on termha s b een deri v ed b y taki ng intoaccou n t only the int e nsity in [@Do d el s o n1 9 93;@Ba rtolo:2006c u] a n d th en i t w ase xtend ed to i n cl u de the effect of po la risati on in [@Pitrou:200 8hy; @Pitr o u : 2008ut;@Ben e ke : 2010eg]. Herewe su mmarize th e resultobtai ned by B eneke eta l . [@Bene ke: 201 0eg ] a p p ly ing the decom p o siti on of the di stribut ion fu nct ion i nto inten sity and l in ea rpol arisa t ion.
Th ecom pl ete expr e ssionof th e co ll is i onterm fo r i n t ensi ty i s gi ven b y $$\ begi n {al igned} {{\cal C }}^ \ Thet a & = a \,\bar{n}_e \si gm a_T \left( - \T heta + \ langle \ Theta \rangle
- \frac{ 3 }{4} S^ {(i ) (j) } \ | )}
-_\beta_{i,j} n^{(i)}_n^{(j)} + h_{i j}{}'_n^{(i)} n^{(j)}_\notag\\
_& \quad
_-_\alpha \Theta' -_(\beta^i + h^i{}_j_n^{(j)}) \Theta_{, i}
-_\Bigl[ \alpha_{,i} -_(\beta_{j,i}_- h_{i j}{}') n^{(j)}
+ (h_{k i, j} - h_{j k, i}) n^{(j)} n^{(k)}_\Bigr]_D^i \Theta_\notag\\
_&_\quad
- \Bigl( \alpha \alpha_{,i}_+ \alpha_{, j} h^j{}_i \Bigr)_n^{(i)}
-_\Bigl[ \alpha (- \beta_{i,j} + h_{i j}{}') +_\beta^k_h_{i j, k}
_+ (\beta_{k, i} - \beta_{i, k} + 2 h_{i_k}{}') h^k{}_j \Bigr] n^{(i)} n^{(j)}
\notag\\
_& \quad
-_\Theta_n^{(i)}_\Bigl( - \alpha_{,i} +_\beta_{i,j} n^{(j)} - h_{i j}{}' n^{(j)}_\Bigr)
= {{\cal C}}^\Theta \,.\end{aligned}$$
Temperature_and spectral distortion of collision terms {#ssec:Collision}
------------------------------------------------------
The_expression of the collision term has_been derived by taking into_account only_the intensity in [@Dodelson1993; @Bartolo:2006cu] and_then it was_extended to_include the effect_of polarisation in [@Pitrou:2008hy; @Pitrou:2008ut; @Beneke:2010eg]. Here_we summarize the_result obtained by Beneke et al._[@Beneke:2010eg]_applying the decomposition_of_the_distribution function_into intensity and_linear_polarisation.
The complete_expression_of the collision term for intensity_is_given by $$\begin{aligned}
{{\cal C}}^\Theta
&= a_\, \bar{n}_e \sigma_T \left(_-_\Theta + \langle \Theta_\rangle
- \frac{3}{4} S^{(i) (j)}_\ |
done. If $W^{\prime }$ contains $F_n$, according to Lemma \[aiju\], we have that $W^{\prime}$ is a linear combination of words in the form $$W_1R_nW_2$$ where $R_n = E_n$, $F_n$ or $E_nF_n$ and $W_i$ are words in $1, E_1, \dots, E_{n-1}, F_1, \dots, F_{n-1}$. If $R_n=E_n$, according to Lemma \[FE=EF\], we can move $R_n$ to the front position and then using the induction hypothesis we are done. Suppose $R_n = F_n$, we note that by induction hypothesis $W_2$ is a linear combination of words in the form $$E_JV(F_nF_{n-1}\cdots F_{j_k})$$ where now $J\in\mathsf{P}_{n-1}$, $V$ is a word reduced in $1, F_1, \dots, F_{n-2}$ (notice that $F_nF_{n-1}\cdots F_{j_k}$ could be empty). Hence $W^{\prime}$ is a linear combination of words of the form $$W_1F_n E_JV(F_nF_{n-1}\cdots F_{j_k})$$ Now, $F_n E_J= E_{s_nJ}F_n $, so using (\[ptlF4\]) and (\[ptlF2\]) follows that $W^{\prime}$ can be written as a linear combination $
(1 + \delta )N_1 - \delta N_2
$ with $N_1:= E_{J^{\prime}}V^{\prime}(F_nF_{n-1}\dots F_{j_k})$ and $N_2:= E_{J^{\prime}}V^{\prime}(E_nF_nF_{n-1}\dots F_{j_k})$, where $J^{\prime}\in\mathsf{P}_n$ and $V^{\prime}$ is a word in $1, F_1, \dots F_{n-1}$. Again we note that in $N_2$, $E_n$ can move to the front position, so $N_ | done. If $ W^{\prime } $ contains $ F_n$, according to Lemma \[aiju\ ], we have that $ W^{\prime}$ is a analogue combination of son in the form $ $ W_1R_nW_2$$ where $ R_n = E_n$, $ F_n$ or $ E_nF_n$ and $ W_i$ are words in $ 1, E_1, \dots, E_{n-1 }, F_1, \dots, F_{n-1}$. If $ R_n = E_n$, harmonize to Lemma \[FE = EF\ ], we can go $ R_n$ to the front position and then using the trigger guess we are done. Suppose $ R_n = F_n$, we note that by induction hypothesis $ W_2 $ is a analogue combination of words in the form $ $ E_JV(F_nF_{n-1}\cdots F_{j_k})$$ where now $ J\in\mathsf{P}_{n-1}$, $ V$ is a word reduce in $ 1, F_1, \dots, F_{n-2}$ (notice that $ F_nF_{n-1}\cdots F_{j_k}$ could be empty). therefore $ W^{\prime}$ is a linear combination of word of the kind $ $ W_1F_n E_JV(F_nF_{n-1}\cdots F_{j_k})$$ Now, $ F_n E_J= E_{s_nJ}F_n $, so using (\[ptlF4\ ]) and (\[ptlF2\ ]) be that $ W^{\prime}$ can be written as a linear combination $
(1 + \delta) N_1 - \delta N_2
$ with $ N_1:= E_{J^{\prime}}V^{\prime}(F_nF_{n-1}\dots F_{j_k})$ and $ N_2:= E_{J^{\prime}}V^{\prime}(E_nF_nF_{n-1}\dots F_{j_k})$, where $ J^{\prime}\in\mathsf{P}_n$ and $ V^{\prime}$ is a word in $ 1, F_1, \dots F_{n-1}$. Again we note that in $ N_2 $, $ E_n$ can travel to the front position, so $ N _ | doje. If $W^{\prime }$ contains $F_k$, according to Lgmna \[aijn\], we habe that $D^{\prime}$ is a linear combinatiln of wirds in the form $$W_1R_nW_2$$ dhere $R_n = E_n$, $F_n$ oe $E_nH_n$ and $W_i$ are wocss in $1, C_1, \dota, E_{n-1}, Y_1, \vots, F_{n-1}$. If $R_n=E_n$, according to Lemma \[FE=EF\], wd ean move $R_n$ to the front position anq then islng the inductyon nrpotgvsls we are done. Suppose $R_n = F_n$, we note tiat by inductiom hypothesis $W_2$ is a linear comhination of words ln the form $$E_JV(S_bF_{n-1}\cdots F_{j_k})$$ where now $J\in\mathsf{P}_{n-1}$, $V$ is a word reduced in $1, F_1, \dots, F_{n-2}$ (nptice that $F_bF_{n-1}\fgots F_{j_k}$ conld be empty). Hence $W^{\prime}$ hs a limear combinatipn mf qords of the form $$W_1F_n E_JV(F_nF_{n-1}\cdots F_{j_k})$$ Now, $F_n A_J= E_{s_nJ}F_n $, so using (\[prlF4\]) atd (\[pdlF2\]) willuws tiat $W^{\primf}$ cen be writtsn as a linwar combination $
(1 + \dtltw )N_1 - \delta N_2
$ wjth $N_1:= Q_{J^{\[rime}}V^{\prime}(F_nF_{n-1}\dots F_{j_k})$ and $N_2:= E_{J^{\prime}}V^{\pgime}(S_nF_nF_{n-1}\dots F_{j_k})$, where $J^{\peime}\in\mathsf{P}_n$ and $V^{\ptime}$ is a rord in $1, F_1, \dots F_{n-1}$. Again we note that in $N_2$, $E_n$ can move go uhc weojt position, so $N_ | done. If $W^{\prime }$ contains $F_n$, according \[aiju\], have that is a linear form where $R_n = $F_n$ or $E_nF_n$ $W_i$ are words in $1, E_1, E_{n-1}, F_1, \dots, F_{n-1}$. If $R_n=E_n$, according to Lemma \[FE=EF\], we can move to the front position and then using the induction hypothesis we are done. $R_n F_n$, note by induction hypothesis $W_2$ is a linear combination of words in the form $$E_JV(F_nF_{n-1}\cdots F_{j_k})$$ where $J\in\mathsf{P}_{n-1}$, $V$ is a word reduced in $1, \dots, F_{n-2}$ (notice that F_{j_k}$ could be empty). Hence is linear combination words the $$W_1F_n E_JV(F_nF_{n-1}\cdots F_{j_k})$$ $F_n E_J= E_{s_nJ}F_n $, so using (\[ptlF4\]) and (\[ptlF2\]) follows that $W^{\prime}$ can be written as a combination $ \delta )N_1 \delta $ $N_1:= E_{J^{\prime}}V^{\prime}(F_nF_{n-1}\dots F_{j_k})$ E_{J^{\prime}}V^{\prime}(E_nF_nF_{n-1}\dots F_{j_k})$, where $J^{\prime}\in\mathsf{P}_n$ and $V^{\prime}$ in $1, F_1, \dots F_{n-1}$. Again we note in $N_2$, can move to the front position, $N_ | done. If $W^{\prime }$ contains $F_n$, accOrding to LeMma \[aiJu\], wE haVe That $w^{\priMe}$ is a linear comBInatIon of words in the form $$W_1R_nw_2$$ wherE $R_N = e_n$, $F_n$ OR $E_NF_n$ anD $W_i$ are wORdS IN $1, E_1, \dOtS, E_{N-1}, F_1, \dOtS, f_{n-1}$. if $R_n=E_N$, acCording To Lemma \[FE=Ef\], we CaN move $R_n$ to the FRoNt position And Then using the IndUction HyPotHEsis wE arE done. supposE $r_n = F_n$, we Note that bY iNDuctioN HypotheSIS $W_2$ Is a lInear combination oF WoRDs in the form $$E_JV(f_nF_{n-1}\cdOtS f_{j_K})$$ WHerE noW $J\in\mathsf{p}_{n-1}$, $v$ is a wORd reducED iN $1, f_1, \DOts, f_{N-2}$ (notice that $F_nf_{n-1}\cdots F_{j_k}$ cOUld Be emptY). HEncE $w^{\prime}$ Is a liNeAR coMbination of WordS of the forM $$W_1F_n E_Jv(f_nF_{n-1}\cdoTS F_{j_k})$$ Now, $f_n E_J= E_{s_NJ}F_N $, so UsinG (\[PtLF4\]) And (\[PtLf2\]) foLLoWs tHAt $W^{\Prime}$ can Be WrItten As a lINEAR comBinAtioN $
(1 + \deltA )N_1 - \delta N_2
$ with $N_1:= e_{J^{\pRime}}v^{\PriMe}(F_nF_{N-1}\dots f_{j_k})$ aNd $n_2:= E_{J^{\prIme}}V^{\prIme}(E_nf_nf_{n-1}\dots F_{j_k})$, where $J^{\PrimE}\in\mathsf{p}_n$ aNd $v^{\prImE}$ is a wORd in $1, F_1, \dOts f_{n-1}$. AGain we nOte that IN $N_2$, $E_N$ cAN MOvE to the front positioN, sO $n_ | done. If $W^{\prime }$ co ntains $F_ n$, a cco rdi ng toLemm a \[aiju\], we have that $W^{\prime}$ isa lin ea r com b in ation of wor d si n th efo rm$$ W _1 R_nW_ 2$$ where$R_n = E_n $,$F _n$ or $E_nF _ n$ and $W_i$ ar e words in $ 1,E_1, \ do ts, E_{n- 1}, F_1, \dots , F_{n- 1}$. If $ R_ n =E_n$, accordi n g t o Le mma \[FE=EF\], we ca n move $R_n$ to the f ro n tp o sit ion and thenus ing t h e induc t io n h ypo t hesis we aredone. Suppo s e $ R_n =F_ n$, we not e tha tb y i nduction hy poth esis $W_2 $ is a linearc ombinat ion of wo rds int he f orm $ $ E_J V (F _nF _ {n- 1}\cdots F _{ j_k}) $$ w h e r e now $J \in\ maths f{P}_{n-1}$,$V$ isa wo rd re duced in$1 , F_1 , \dot s, F_ {n -2}$ (notice th at $ F_nF_{n-1 }\c do tsF_ {j_k} $ could be em pty). H ence $W ^ {\p ri m e } $is a linear combin at i o nof words of th e f or m $$W_1F_ n E_ JV(F _ n F_{n- 1}\c d ot s F_{j_k })$$ N o w, $ F_n E_ J= E_{s_ nJ }F_ n $ , sou sing (\[pt lF4\]) a nd (\ [ ptlF2\]) follo w s that $W^{\p r im e } $c an b e w ritten as a lin e ar c ombi n at ion $
(1+ \de lt a ) N _1 - \delta N_2
$ w it h $N_1 := E_ {J^{\prime}}V ^{\prime}( F _ n F_{n-1}\ dots F_ { j_k})$ and $N_ 2:= E _{J^{\prim e }}V^{\pr ime}( E_nF_nF_ {n-1}\dot s F_{j_k}) $,whe re$J^ { \ pr ime}\in\maths f { P}_n $and $V^ {\p rime}$isa w ord in $ 1, F_1, \ dots F_{ n- 1} $. A gai n wen ote that i n $ N_ 2$, $E_n $ can m ove t o th efr o ntpositio n ,s o $N_ | done._If $W^{\prime_}$ contains $F_n$, according_to Lemma_\[aiju\],_we have_that_$W^{\prime}$ is a_linear combination of_words in the form_$$W_1R_nW_2$$ where $R_n_=_E_n$, $F_n$ or $E_nF_n$ and $W_i$ are words in $1, E_1, \dots, E_{n-1}, F_1,_\dots,_F_{n-1}$. If_$R_n=E_n$,_according_to Lemma \[FE=EF\], we can_move $R_n$ to the front_position and_then using the induction hypothesis we are done._Suppose_$R_n = F_n$,_we note that by induction hypothesis $W_2$ is a_linear combination of words in the_form $$E_JV(F_nF_{n-1}\cdots F_{j_k})$$_where_now_$J\in\mathsf{P}_{n-1}$, $V$ is a_word reduced in $1, F_1, \dots,_F_{n-2}$ (notice that $F_nF_{n-1}\cdots F_{j_k}$ could_be empty). Hence $W^{\prime}$ is a linear_combination of words of the form_$$W_1F_n E_JV(F_nF_{n-1}\cdots F_{j_k})$$ Now,_$F_n _E_J= E_{s_nJ}F_n $, so using_(\[ptlF4\]) and (\[ptlF2\])_follows that_$W^{\prime}$ can be_written as a linear combination $
(1_+ \delta )N_1_- \delta N_2
$ with $N_1:= E_{J^{\prime}}V^{\prime}(F_nF_{n-1}\dots_F_{j_k})$_and $N_2:= E_{J^{\prime}}V^{\prime}(E_nF_nF_{n-1}\dots_F_{j_k})$,_where_$J^{\prime}\in\mathsf{P}_n$ and_$V^{\prime}$ is a_word_in $1,_F_1,_\dots F_{n-1}$. Again we note that_in_$N_2$, $E_n$ can move to the front_position, so $N_ |
eq:D1\])\], is caused by the diverging length scale below jamming, i.e. $\xi\sim|\Delta\phi|^{-1}$ \[Eq. (\[eq:xi1\])\]. As shown in Fig. \[fig:xi\](b), the linear relation (dotted line) well explains our results if the length scale $\xi$ is smaller than $10d_0$. If the system is above jamming, the length scale increases (more than $10d_0$) with the increase of $\phi$. However, the diffusivity over the shear rate $D/\dot{\gamma}$ starts to deviate from the linear relation (dotted line) and the length scale reaches the system size $L/2\simeq 44d_0$ (shaded region). We conclude that this deviation is caused by finite-size effects and further studies of different system sizes are necessary (as in Refs. [@diff_shear_md3; @diff_shear_md4]) to figure out the relation between $D/\dot{\gamma}$ and $\xi$ in this regime, which we postpone as a future work.
Discussions {#sec:disc}
===========
In this study, we have numerically investigated rheological and transport properties of soft athermal particles under shear. Employing MD simulations of two-dimensional disks, we have clarified how the rheology, self-diffusion, and size of rigid clusters vary with the control parameters,i.e. the externally imposed shear rate $\dot{\gamma}$ and packing fraction of the disks $\phi$. Our main result is the critical scaling of the diffusivity (Sec. \[sub:diff\]) and size of rigid clusters (Sec. \[sub:rigid\]), where their dependence on both $\dot{\gamma}$ and $\phi$ is reported \[Eqs. (\[eq:D1\]), (\[eq:D2\]), (\[eq:xi1\]), and (\[eq:xi2\])\]. The diffusivity has been calculated on both sides of jamming (by a single numerical protocol) to unify the understanding of self-diffusion of soft particulate systems: We found that (i) the diffusivity below jamming exhibits a crossover from the linear scaling $D\sim\dot{\gamma}$ to the power-law $D\sim\dot{\gamma}^{3/4}$. Such a crossover can be also seen in previous simulations [@diff_she | eq: D1\])\ ], is caused by the diverging length scale below jamming, i.e. $ \xi\sim|\Delta\phi|^{-1}$ \[Eq. (\[eq: xi1\])\ ]. As testify in Fig. \[fig: xi\](b), the analogue relation (dotted course) well explain our results if the length plate $ \xi$ is smaller than $ 10d_0$. If the system is above jamming, the length plate increase (more than $ 10d_0 $) with the increase of $ \phi$. However, the diffusivity over the shear pace $ D/\dot{\gamma}$ starts to deviate from the linear relation (dotted lineage) and the length scale reaches the system size $ L/2\simeq 44d_0 $ (shaded region). We reason that this deviation is caused by finite - size effects and further study of different system size are necessary (as in Refs. [ @diff_shear_md3; @diff_shear_md4 ]) to figure out the sexual intercourse between $ D/\dot{\gamma}$ and $ \xi$ in this regime, which we postpone as a future work.
Discussions { # sec: disc }
= = = = = = = = = = =
In this study, we have numerically investigate rheological and transport properties of soft athermal particles under shear. Employing MD simulations of two - dimensional disks, we have clarified how the rheology, self - diffusion, and size of rigid clusters vary with the control parameters, i.e. the externally imposed shear pace $ \dot{\gamma}$ and tamp down fraction of the disks $ \phi$. Our independent resultant role is the critical scaling of the diffusivity (Sec. \[sub: diff\ ]) and size of rigid clusters (Sec. \[sub: rigid\ ]), where their dependence on both $ \dot{\gamma}$ and $ \phi$ is report \[Eqs. (\[eq: D1\ ]), (\[eq: D2\ ]), (\[eq: xi1\ ]), and (\[eq: xi2\])\ ]. The diffusivity has been calculate on both sides of jamming (by a single numeric protocol) to unify the understanding of self - diffusion of soft particulate systems: We found that (i) the diffusivity below jamming exhibits a crossover from the linear scaling $ D\sim\dot{\gamma}$ to the power - jurisprudence $ D\sim\dot{\gamma}^{3/4}$. Such a crossover can be besides watch in previous simulations [ @diff_she | eq:D1\])\], is caused by the diverglng length scale below oamming, j.e. $\xi\sim|\Ddlta\phi|^{-1}$ \[Eq. (\[eq:xi1\])\]. As shown in Fmg. \[fit:xi\](b), uke linear relation (dutted linv) well explaiis our results ih the lekyth sdwle $\ri$ is smaller thsn $10d_0$. If tha system is abmvd lamming, the length scale increases (mjre tham $10f_0$) with the inctease jf $\pgp$. Mowever, the diffusivity over the shear gate $D/\dot{\gamma}$ stsrts to deviate from the llneag relation (dotted pine) and thg legtth scale rexches the system size $M/2\simeq 44d_0$ (shaded region). We concljde tkat this deciqtilt is caused by fpnite-size effcbts and furthet studies of dlfferxnt wystem sizes are necevsary (as in Refs. [@disf_shear_md3; @dnff_shear_md4]) to figure iur the reldtiov begwetn $V/\dof{\gamma}$ anv $\xi$ in thia regime, whuch we postpone as s slyure work.
Disdussiogs {#sec:disc}
===========
In this study, we have numericalny jnvestigated rheologicao and transport propetties of sjft athermal particles under shear. Employing MD shmulaviuns on twu-eilensional disks, we have clarified how the rhejmoby, self-diffusion, and size og gibyd clusters vxry wicg fhe control paramehers,i.e. tre exrernally ympoxed shear rate $\dot{\gamma}$ and packing frabtiob of the disks $\phi$. Our main rzsult os thr critical scaling of tke difrusivity (Sef. \[sub:diff\]) zvd size of rigid cllstess (Sec. \[sub:rigid\]), where their dependenre on both $\dut{\gakma}$ anq $\phi$ is rfportcg \[Eqs. (\[eq:D1\]), (\[eq:D2\]), (\[eq:xi1\]), wnd (\[ee:xh2\])\]. The diffksivity has been calculated on upth sides of jdmmpng (by a finglc numerical proeocol) to unify the unberstavding of svlf-diffusmon of soft [articulate sfdtems: We fouid that (i) the difdusivith below jamming exhibits a crossovwr from the linear scxming $D\sim\dot{\gamna}$ uo the power-law $C\sio\doe{\gwmka}^{3/4}$. Vuch a crossmver cav be auso seen in orevoous simulations [@difx_she | eq:D1\])\], is caused by the diverging length jamming, $\xi\sim|\Delta\phi|^{-1}$ \[Eq. As shown in (dotted well explains our if the length $\xi$ is smaller than $10d_0$. If system is above jamming, the length scale increases (more than $10d_0$) with the of $\phi$. However, the diffusivity over the shear rate $D/\dot{\gamma}$ starts to deviate the relation line) the length scale reaches the system size $L/2\simeq 44d_0$ (shaded region). We conclude that this deviation caused by finite-size effects and further studies of system sizes are necessary in Refs. [@diff_shear_md3; @diff_shear_md4]) to out relation between and in regime, which we as a future work. Discussions {#sec:disc} =========== In this study, we have numerically investigated rheological and transport of soft under shear. MD of disks, we have the rheology, self-diffusion, and size of with the control parameters,i.e. the externally imposed shear $\dot{\gamma}$ and fraction of the disks $\phi$. Our result is the critical scaling of the diffusivity \[sub:diff\]) and size of rigid clusters (Sec. \[sub:rigid\]), where their dependence on both $\dot{\gamma}$ and reported \[Eqs. (\[eq:D1\]), (\[eq:D2\]), and (\[eq:xi2\])\]. The has calculated both of jamming a single numerical protocol) to unify the understanding of self-diffusion of particulate systems: We found that (i) the diffusivity below jamming crossover the linear scaling to the power-law $D\sim\dot{\gamma}^{3/4}$. a can be also seen simulations | eq:D1\])\], is caused by the diverging Length scalE beloW jaMmiNg, I.e. $\xi\Sim|\DElta\phi|^{-1}$ \[Eq. (\[eq:xi1\])\]. AS ShowN in Fig. \[fig:xi\](b), the linear rElatiOn (DOtteD LiNe) welL explaiNS oUR ResUlTs If tHe LEnGth scAle $\Xi$ is smaLler than $10d_0$. IF thE sYstem is above JAmMing, the lenGth Scale increasEs (mOre thaN $10d_0$) WitH The inCreAse of $\Phi$. HowEVer, the DiffusiviTy OVer the SHear ratE $d/\DoT{\gamMa}$ starts to deviate FRoM The linear relatIon (dotTeD LiNE) And The Length scalE rEacheS The systEM sIZE $l/2\siMEq 44d_0$ (shaded regiOn). We concludE ThaT this dEvIatIOn is caUsed bY fINitE-size effectS and Further stUdies oF DiffereNT system Sizes aRe nEceSsarY (As In refS. [@dIFf_sHEaR_md3; @DIff_Shear_md4]) tO fIgUre ouT the RELATion BetWeen $d/\dot{\gAmma}$ and $\xi$ in thIs rEgimE, WhiCh we pOstpoNe as A fUture Work.
DiScussIoNs {#sec:disc}
===========
In this StudY, we have nuMerIcAllY iNvestIGated rHeoLogIcal and TranspoRT prOpERTIeS of soft athermal parTiCLEs Under sheAr. EmplOYiNg md simulatIoNs oF two-DIMensiOnal DIsKs, we have ClarifIEd HoW the rheOlOgy, selF-dIffUsiOn, and SIze oF rigid Clusters Vary wITh the control paRAmeters,i.e. the eXTeRNAlLY impOseD shear rate $\dOt{\gaMMa}$ anD pacKInG frACtion Of the DiSKs $\PHi$. Our main result is thE cRiticaL scalIng of the diffuSivity (Sec. \[sUB:DIff\]) and siZe of RIgID clusters (Sec. \[suB:rigiD\]), where theiR DependenCe on bOth $\dot{\gaMma}$ and $\phi$ IS Reported \[eqs. (\[Eq:D1\]), (\[Eq:D2\]), (\[Eq:xI1\]), ANd (\[Eq:xi2\])\]. The diffusIVIty hAs Been calCulAted on bOth SidEs oF jaMmIng (by a sinGle numerIcAl PrOtOcoL) to unIFy the undErStaNdIng Of selF-DiffusIon of Soft PaRtICulAte systEMs: wE FounD tHaT (i) thE diFfUsiviTy beLOw jAmming eXhibits a cRosSOver FrOm The lineAr scaling $D\sim\DoT{\gamma}$ to thE pOweR-law $D\sIM\Dot{\gamma}^{3/4}$. such a crossover can be also SEen in prEviOus siMulaTions [@diff_She | eq:D1\])\], is caused by t he divergi ng le ngt h s ca le b elow jamming, i.e. $\xi \sim|\Delta\phi|^{-1}$ \[Eq .( \[eq : xi 1\])\ ]. As s h ow n inFi g. \[ fi g :x i\](b ),the lin ear relati on(d otted line)w el l explains ou r results if th e leng th sc a le $\ xi$ is s maller than $ 10d_0$. I ft he sys t em is a b o ve jam ming, the lengths ca l e increases (m ore th an $1 0 d _0$ ) w ith the in cr easeo f $\phi $ .H o w eve r , the diffusi vity over t h e s hear r at e $ D /\dot{ \gamm a} $ st arts to dev iate from the linea r relati o n (dott ed lin e)and the le ng thsc a ler ea che s th e system s iz e $L/ 2\si m e q 44d_ 0$(sha ded r egion). We co ncl udet hat this devi atio nis ca used b y fin it e-size effectsandfurther s tud ie s o fdiffe r ent sy ste m s izes ar e neces s ary ( a s in Refs. [@diff_shea r_ m d 3; @diff_s hear_m d 4] )t o figure o utther e latio n be t we en $D/\d ot{\ga m ma }$ and $\ xi $ in t hi s r egi me, w h ichwe pos tpone as a fu t ure work.
Dis c ussions {#sec : di s c }= ==== === ===
In thi s st u dy,we h a ve nu m erica lly i nv e st i gated rheological a nd trans portproperties of soft athe r m a l partic lesu nd e r shear. Emplo yingMD simulat i ons of t wo-di mensiona l disks,w e have cl ari fie d h owt h erheology, sel f - diff us ion, an d s ize ofrig idclu ste rs vary wit h the co nt ro lpa ram eters , i.e. the e xte rn all y imp o sed sh ear r ate$\ do t {\g amma}$a nd p acki ng f ract ion o f the dis k s $ \phi$.Our mainres u lt i sth e criti cal scaling o fthe diffus iv ity (Sec. \ [sub:dif f\]) and size of rigidc lusters (S ec. \ [sub :rigid\]) , w here t hei r depen denceon bo th $\ d o t{\ga m m a} $ a nd $\phi$ is r epo rted\[ Eqs. (\[eq: D1\]), (\[eq:D2\]) , (\ [eq:xi1\]), a nd(\[e q : xi 2\] ) \] . Th ed iff u s ivity has beencalculated o n b oth sideso f j am ming (b y a sin gle n u merical protocol ) to unif ytheu n der standing o f self-d iffusiono f sof t p artic ula te sys te ms: We f ound t h at(i) t he dif fu sivity belo wjammingexhibits a crossover fr om the line arscaling $ D\s i m\d ot{\gamma }$ t o the powe r-l aw$D\si m\d o t{\ga mma} ^ {3 /4} $ . Suc h ac rossoverc an be a ls o seen in p r e v iou s sim ula t ions [ @dif f_she | eq:D1\])\], is_caused by_the diverging length scale_below jamming, i.e. $\xi\sim|\Delta\phi|^{-1}$_\[Eq. (\[eq:xi1\])\]._As shown_in_Fig. \[fig:xi\](b), the linear_relation (dotted line)_well explains our results_if the length_scale_$\xi$ is smaller than $10d_0$. If the system is above jamming, the length scale_increases_(more than_$10d_0$)_with_the increase of $\phi$. However,_the diffusivity over the shear_rate $D/\dot{\gamma}$_starts to deviate from the linear relation (dotted_line)_and the length_scale reaches the system size $L/2\simeq 44d_0$ (shaded region)._We conclude that this deviation is_caused by finite-size_effects_and_further studies of different_system sizes are necessary (as in_Refs. [@diff_shear_md3; @diff_shear_md4]) to figure out the_relation between $D/\dot{\gamma}$ and $\xi$ in this_regime, which we postpone as a_future work.
Discussions {#sec:disc}
===========
In this study,_we have_numerically investigated rheological and transport_properties of soft_athermal particles_under shear. Employing_MD simulations of two-dimensional disks, we_have clarified how_the rheology, self-diffusion, and size of_rigid_clusters vary with_the_control_parameters,i.e. the externally_imposed shear rate_$\dot{\gamma}$_and packing_fraction_of the disks $\phi$. Our main_result_is the critical scaling of the diffusivity_(Sec. \[sub:diff\]) and size of_rigid_clusters (Sec. \[sub:rigid\]), where their_dependence on both $\dot{\gamma}$ and_$\phi$ is reported \[Eqs. (\[eq:D1\]), (\[eq:D2\]), (\[eq:xi1\]),_and (\[eq:xi2\])\]._The diffusivity_has been calculated on both sides of jamming (by a single_numerical protocol) to unify the understanding_of self-diffusion of soft_particulate systems:_We_found that (i)_the_diffusivity below_jamming exhibits a crossover from the linear_scaling $D\sim\dot{\gamma}$_to the power-law $D\sim\dot{\gamma}^{3/4}$. Such a_crossover can be also_seen_in previous simulations [@diff_she |
(\[19\]). For this aim we rewrite (\[44\]) in the second order form by means of an integration over the variable $\pi^\mu$. First, by using of $\gamma$-matrix identities as well as the identity $$\begin{aligned}
\label{45}
\epsilon^{dabc}\epsilon_{d\mu\nu\rho}=-\delta^a_\mu
(\delta^b_\nu\delta^c_\rho-\delta^c_\nu\delta^b_\rho)+ \cr
\delta^b_\mu(\delta^a_\nu\delta^c_\rho-\delta^c_\nu\delta^a_\rho)-
\delta^c_\mu(\delta^a_\nu\delta^b_\rho-\delta^b_\nu\delta^a_\rho),\end{aligned}$$ the second Casimir operator $W^2$ can be rewritten as $$\begin{aligned}
\label{46}
W^2=-\frac 18\pi^2(\bar\chi\Gamma^{\mu\nu}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)+
\frac
14(\bar\chi\Gamma^{\mu\rho}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)p_\rho p^\nu \cr
=-\frac 1{16}\pi^2(\bar\xi\chi)^2+
\frac 14\pi_\mu(\bar\xi\bar\sigma^\mu\chi)
\xi(\pi_\nu\sigma^\nu\bar\chi-m\chi)+ \cr
\frac{m}4(\bar\xi\chi)\bar\xi(\pi_\mu\bar\sigma^\mu\chi-m\chi)+
\frac 14(\bar\xi\bar\chi)(\xi\chi)(\pi^2+m^2),\end{aligned}$$ where the last three terms are written in two dimensional spinor notations, and $\sigma^\mu{}_{b\dot a},\bar\sigma^{\mu\dot a b}$ are $D=4$ matrices Pauli [@27]. After substitution of (\[46\]) into (\[44\]), the last three terms can be included into redefinition of the variables $e,\lambda^a,\bar\lambda_{\dot a}$. As a result, one obtains the action in the form $$\begin{aligned}
\label{47}
S=\int d\tau\ | (\[19\ ]). For this aim we rewrite (\[44\ ]) in the second order form by means of an consolidation over the variable star $ \pi^\mu$. foremost, by using of $ \gamma$-matrix identities as well as the identity $ $ \begin{aligned }
\label{45 }
\epsilon^{dabc}\epsilon_{d\mu\nu\rho}=-\delta^a_\mu
(\delta^b_\nu\delta^c_\rho-\delta^c_\nu\delta^b_\rho)+ \cr
\delta^b_\mu(\delta^a_\nu\delta^c_\rho-\delta^c_\nu\delta^a_\rho)-
\delta^c_\mu(\delta^a_\nu\delta^b_\rho-\delta^b_\nu\delta^a_\rho),\end{aligned}$$ the second Casimir hustler $ W^2 $ can be rewritten as $ $ \begin{aligned }
\label{46 }
W^2=-\frac 18\pi^2(\bar\chi\Gamma^{\mu\nu}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)+
\frac
14(\bar\chi\Gamma^{\mu\rho}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)p_\rho p^\nu \cr
= -\frac 1{16}\pi^2(\bar\xi\chi)^2 +
\frac 14\pi_\mu(\bar\xi\bar\sigma^\mu\chi)
\xi(\pi_\nu\sigma^\nu\bar\chi - m\chi)+ \cr
\frac{m}4(\bar\xi\chi)\bar\xi(\pi_\mu\bar\sigma^\mu\chi - m\chi)+
\frac 14(\bar\xi\bar\chi)(\xi\chi)(\pi^2+m^2),\end{aligned}$$ where the last three terms are write in two dimensional spinor notations, and $ \sigma^\mu{}_{b\dot a},\bar\sigma^{\mu\dot a b}$ are $ D=4 $ matrices Pauli [ @27 ]. After substitution of (\[46\ ]) into (\[44\ ]), the concluding three terms can be included into redefinition of the variable $ e,\lambda^a,\bar\lambda_{\dot a}$. As a result, one obtains the natural process in the mannequin $ $ \begin{aligned }
\label{47 }
S=\int d\tau\ | (\[19\]). Flr this aim we rewrite (\[44\]) ln the second orbwr fork by msans of xn integration over the varieble $\pi^\mu$. First, by using of $\gamoa$-matrix pdentitiew as qell as thx identibv $$\begjk{aliguev}
\label{45}
\epsilon^{danc}\epsilon_{d\mg\nu\rho}=-\delta^a_\mu
(\geutc^b_\nu\delta^c_\rho-\delta^c_\nu\delta^b_\rho)+ \cr
\delea^b_\mu(\dektw^a_\nu\delta^c_\rho-\dglta^c_\mt\delfa^a_\rho)-
\delta^c_\mu(\delta^a_\nu\delta^b_\rho-\delfa^b_\nu\denta^a_\rho),\end{alibned}$$ the second Casimir opfratlr $W^2$ can be rewrithen as $$\begib{alidbed}
\label{46}
W^2=-\fraz 18\pi^2(\bar\chi\Gamma^{\mu\nu}\xi)(\bzr\chi\Gamma_{\mu\nu}\xi)+
\frac
14(\bar\chi\Gamma^{\ou\rho}\ri)(\bar\chi\Gamna_{\nu\nk}\fi)p_\rho p^\nu \rr
=-\frac 1{16}\pi^2(\bar\xi\chi)^2+
\fvsc 14\pi_\mg(\bar\xi\bsr\sigma^\mu\chi)
\xl(\pi_\nu\vigna^\nu\bar\chi-m\chi)+ \cr
\frac{k}4(\bar\xi\chi)\bar\xi(\pi_\mu\far\sigma^\mg\cki-m\chi)+
\frac 14(\bar\xi\bar\chu)(\xu\chi)(\ph^2+m^2),\eng{aliebed}$$ whtre tge lash tiree terms zre written in two dimensional s[pmor notationa, and $\fidma^\mu{}_{b\dot a},\bar\sigma^{\mu\dot a b}$ are $D=4$ matrpces Pauli [@27]. After substitutuon of (\[46\]) into (\[44\]), the lasj three tewms can be included into redefinition of the varidbles $d,\lanbqx^q,\bwr\lambda_{\dot a}$. As a result, one obtains the actykn ik the form $$\begin{cligned}
\label{47}
S=\int d\hai\ | (\[19\]). For this aim we rewrite (\[44\]) second form by of an integration by of $\gamma$-matrix identities well as the $$\begin{aligned} \label{45} \epsilon^{dabc}\epsilon_{d\mu\nu\rho}=-\delta^a_\mu (\delta^b_\nu\delta^c_\rho-\delta^c_\nu\delta^b_\rho)+ \cr \delta^b_\mu(\delta^a_\nu\delta^c_\rho-\delta^c_\nu\delta^a_\rho)- the second Casimir operator $W^2$ can be rewritten as $$\begin{aligned} \label{46} W^2=-\frac 18\pi^2(\bar\chi\Gamma^{\mu\nu}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)+ 14(\bar\chi\Gamma^{\mu\rho}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)p_\rho p^\nu \cr =-\frac 1{16}\pi^2(\bar\xi\chi)^2+ \frac 14\pi_\mu(\bar\xi\bar\sigma^\mu\chi) \xi(\pi_\nu\sigma^\nu\bar\chi-m\chi)+ \cr \frac{m}4(\bar\xi\chi)\bar\xi(\pi_\mu\bar\sigma^\mu\chi-m\chi)+ \frac 14(\bar\xi\bar\chi)(\xi\chi)(\pi^2+m^2),\end{aligned}$$ where last terms written two dimensional spinor notations, and $\sigma^\mu{}_{b\dot a},\bar\sigma^{\mu\dot a b}$ are $D=4$ matrices Pauli [@27]. After substitution (\[46\]) into (\[44\]), the last three terms can included into redefinition of variables $e,\lambda^a,\bar\lambda_{\dot a}$. As a one the action the $$\begin{aligned} S=\int d\tau\ | (\[19\]). For this aim we rewrite (\[44\]) in the sEcond order Form bY meAns Of An inTegrAtion over the vaRIablE $\pi^\mu$. First, by using of $\gamMa$-matRiX IdenTItIes as Well as tHE iDENtiTy $$\BeGin{AlIGnEd}
\labEl{45}
\ePsilon^{dAbc}\epsilon_{D\mu\Nu\Rho}=-\delta^a_\mu
(\dELtA^b_\nu\delta^c_\Rho-\Delta^c_\nu\deltA^b_\rHo)+ \cr
\deLtA^b_\mU(\Delta^A_\nu\Delta^C_\rho-\deLTa^c_\nu\dElta^a_\rho)-
\dElTA^c_\mu(\deLTa^a_\nu\deLTA^b_\Rho-\dElta^b_\nu\delta^a_\rho),\eND{aLIgned}$$ the second casimiR oPErATOr $W^2$ Can Be rewritteN aS $$\begiN{Aligned}
\LAbEL{46}
w^2=-\FraC 18\Pi^2(\bar\chi\Gamma^{\Mu\nu}\xi)(\bar\chI\gamMa_{\mu\nu}\Xi)+
\FraC
14(\Bar\chi\gamma^{\Mu\RHo}\xI)(\bar\chi\GammA_{\mu\nU}\xi)p_\rho p^\nU \cr
=-\fraC 1{16}\Pi^2(\bar\xi\CHi)^2+
\frac 14\pI_\mu(\bar\Xi\bAr\sIgma^\MU\cHi)
\Xi(\pI_\nU\SigMA^\nU\baR\Chi-M\chi)+ \cr
\frAc{M}4(\bAr\xi\cHi)\baR\XI(\PI_\mu\bAr\sIgma^\Mu\chi-M\chi)+
\frac 14(\bar\xi\Bar\Chi)(\xI\Chi)(\Pi^2+m^2),\enD{aligNed}$$ wHeRe the Last thRee teRmS are written in twO dimEnsional sPinOr NotAtIons, aND $\sigma^\Mu{}_{b\Dot A},\bar\sigMa^{\mu\dot A B}$ arE $D=4$ MATRiCes Pauli [@27]. After substItUTIoN of (\[46\]) into (\[44\]), tHe last THrEe TErms can bE iNclUded INTo redEfinITiOn of the vAriablES $e,\LaMbda^a,\baR\lAmbda_{\dOt A}$. As A reSult, oNE obtAins thE action iN the fORm $$\begin{aligned}
\LAbel{47}
S=\int d\tau\ | (\[19\]). For this aim we rewrite ( \[44\ ])inth e se cond order form by mean s of an integration ov er th ev aria b le $\pi ^\mu$.F ir s t , b yus ing o f $ \gamm a$- matrixidentities as w ell as the i d en tity $$\be gin {aligned}
\l abe l{45}\e psi l on^{d abc }\eps ilon_{ d \mu\nu \rho}=-\d el t a^a_\m u
(\delt a ^ b_ \nu\ delta^c_\rho-\del t a^ c _\nu\delta^b_\ rho)+\c r
\ d e lta ^b_ \mu(\delta ^a _\nu\ d elta^c_ \ rh o - \ del t a^c_\nu\delta ^a_\rho)-
\ d elt a^c_\m u( \de l ta^a_\ nu\de lt a ^b_ \rho-\delta ^b_\ nu\delta^ a_\rho ) ,\end{a l igned}$ $ thesec ond Cas i mi rope ra t or$ W^ 2$c anbe rewri tt en as $ $\be g i n { alig ned }
\l abel{ 46}
W^2=-\fra c 1 8\pi ^ 2(\ bar\c hi\Ga mma^ {\ mu\nu }\xi)( \bar\ ch i\Gamma_{\mu\nu }\xi )+
\frac14( \b ar\ ch i\Gam m a^{\mu \rh o}\ xi)(\ba r\chi\G a mma _{ \ m u \n u}\xi)p_\rho p^\nu \ c r
= -\frac 1 {16}\p i ^2 (\ b ar\xi\ch i) ^2+
\fr a c 14\p i_\m u (\ bar\xi\b ar\sig m a^ \m u\chi)
\ xi(\pi _\ nu\ sig ma^\n u \bar \chi-m \chi)+ \ cr
\f r ac{m}4(\bar\xi \ chi)\bar\xi(\ p i_ \ m u\ b ar\s igm a^\mu\chi-m \chi ) +
\f rac1 4( \ba r \xi\b ar\ch i) ( \x i \chi)(\pi^2+m^2),\e nd {align ed}$$ where the la st three t e r m s are wr itte n i n two dimension al sp inor notat i ons, and $\si gma^\mu{ }_{b\dota } ,\bar\si gma ^{\ mu\ dot a b }$ are $D=4$m a tric es Pauli[@2 7]. Aft ersub sti tut io n of (\[4 6\]) int o(\ [4 4\ ]), thel ast thre eter ms ca n bei nclude d int o re de fi n iti on of t h ev a riab le s$e,\ lam bd a^a,\ bar\ l amb da_{\do t a}$. As ar esul t, o ne obta ins the actio nin the for m$$\ begin{ a l igned}
\ label{47}
S=\int d\tau\ | (\[19\])._For this_aim we rewrite (\[44\])_in the_second_order form_by_means of an_integration over the_variable $\pi^\mu$. First, by_using of $\gamma$-matrix_identities_as well as the identity $$\begin{aligned}
\label{45}
\epsilon^{dabc}\epsilon_{d\mu\nu\rho}=-\delta^a_\mu
(\delta^b_\nu\delta^c_\rho-\delta^c_\nu\delta^b_\rho)+ \cr
\delta^b_\mu(\delta^a_\nu\delta^c_\rho-\delta^c_\nu\delta^a_\rho)-
\delta^c_\mu(\delta^a_\nu\delta^b_\rho-\delta^b_\nu\delta^a_\rho),\end{aligned}$$ the second Casimir operator $W^2$ can be_rewritten_as $$\begin{aligned}
\label{46}
W^2=-\frac_18\pi^2(\bar\chi\Gamma^{\mu\nu}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)+
\frac
14(\bar\chi\Gamma^{\mu\rho}\xi)(\bar\chi\Gamma_{\mu\nu}\xi)p_\rho_p^\nu_\cr
=-\frac 1{16}\pi^2(\bar\xi\chi)^2+
\frac 14\pi_\mu(\bar\xi\bar\sigma^\mu\chi)
\xi(\pi_\nu\sigma^\nu\bar\chi-m\chi)+ \cr
\frac{m}4(\bar\xi\chi)\bar\xi(\pi_\mu\bar\sigma^\mu\chi-m\chi)+
\frac_14(\bar\xi\bar\chi)(\xi\chi)(\pi^2+m^2),\end{aligned}$$ where the last three_terms are_written in two dimensional spinor notations, and $\sigma^\mu{}_{b\dot_a},\bar\sigma^{\mu\dot_a b}$ are_$D=4$ matrices Pauli [@27]. After substitution of (\[46\]) into_(\[44\]), the last three terms can_be included into_redefinition_of_the variables $e,\lambda^a,\bar\lambda_{\dot a}$._As a result, one obtains the_action in the form $$\begin{aligned}
\label{47}
S=\int d\tau\ |
power-law $\Theta(k)$, the scalar variance diverges in the limit $\kappa\to0$. Finally, the present results are discussed in the light of existing literature$^{8,13,17,20}$ on the exponential decay behavior at large times, when the exponential decay rate becomes independent of $\kappa$ and remains non-zero.
Basic estimates
===============
This section briefly recalls a basic calculus inequality and related estimates, both being used in this study. In what follows, $\theta$ is bounded and assumed to be sufficiently smooth so that the quantities under consideration are well defined.
The Fourier representation of $\theta(\x,t)$ is $$\begin{aligned}
\label{fourier}
\theta(\x) = \sum_{\k}\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $\widehat{\theta}(\k)$ is the Fourier transform of $\theta(\x)$ and $\k$ is the wave vector. In (\[fourier\]), the time variable has been suppressed for convenience. For real $\alpha$, including negative values, the fractional derivative $(-\Delta)^{\alpha/2}$ is defined by $$\begin{aligned}
\label{fractional}
(-\Delta)^{\alpha/2}\theta(\x) =
\sum_{\k}k^\alpha\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $k=|\k|$ is the wave number. For $\alpha\le1$, the following interpolation-type inequality holds $$\begin{aligned}
\label{holder}
\langle|\nabla\theta|^2\rangle
&\le&
\langle|\Delta\theta|^2\rangle^{(1-\alpha)/(2-\alpha)}
\langle|(-\Delta)^{\alpha/2}\theta|^2\rangle^{1/(2-\alpha)}.\end{aligned}$$ This inequality reduces to the Cauchy–Schwarz inequality and an identity for $\alpha=0$ and $\alpha=1$, respectively. For $\alpha<1$, Eq. (\[holder\]) can be proved by an elementary method,$^{24,25}$ and the proof goes as follows. One has $$\begin{aligned}
\label{holder1}
\int k^2\Theta(k)\,dk &=&
\int(k^4\Theta(k))^{(1-\alpha)/(2-\alpha)}
| power - law $ \Theta(k)$, the scalar variance diverges in the limit $ \kappa\to0$. ultimately, the present result are discussed in the light of existing literature$^{8,13,17,20}$ on the exponential decay demeanor at large times, when the exponential decay rate become independent of $ \kappa$ and remains non - zero.
Basic estimate
= = = = = = = = = = = = = = =
This section briefly recall a basic calculus inequality and related appraisal, both being used in this study. In what follows, $ \theta$ is bounded and assumed to be sufficiently smooth so that the measure under consideration are well defined.
The Fourier representation of $ \theta(\x, t)$ is $ $ \begin{aligned }
\label{fourier }
\theta(\x) = \sum_{\k}\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $ \widehat{\theta}(\k)$ is the Fourier transform of $ \theta(\x)$ and $ \k$ is the wave vector. In (\[fourier\ ]), the prison term variable has been suppressed for convenience. For real $ \alpha$, include negative values, the fractional derivative $ (-\Delta)^{\alpha/2}$ is defined by $ $ \begin{aligned }
\label{fractional }
(-\Delta)^{\alpha/2}\theta(\x) =
\sum_{\k}k^\alpha\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $ k=|\k|$ is the wave number. For $ \alpha\le1 $, the follow interpolation - type inequality holds $ $ \begin{aligned }
\label{holder }
\langle|\nabla\theta|^2\rangle
& \le &
\langle|\Delta\theta|^2\rangle^{(1-\alpha)/(2-\alpha) }
\langle|(-\Delta)^{\alpha/2}\theta|^2\rangle^{1/(2-\alpha)}.\end{aligned}$$ This inequality reduces to the Cauchy – Schwarz inequality and an identity for $ \alpha=0 $ and $ \alpha=1 $, respectively. For $ \alpha<1 $, Eq. (\[holder\ ]) can be proved by an elementary method,$^{24,25}$ and the validation go as follows. One has $ $ \begin{aligned }
\label{holder1 }
\int k^2\Theta(k)\,dk & = &
\int(k^4\Theta(k))^{(1-\alpha)/(2-\alpha) } | poaer-law $\Theta(k)$, the scalar variance diveryws in vhe limjt $\kappa\go0$. Finally, the present resulvs aee diwcussed in the light ow existinh literarure$^{8,13,17,20}$ in the exponential decay nehavnoc at large timex, when the exponential dacxy rate becomes independent of $\kappa$ wnd remsijs non-zero.
Basis esuimwtes
===============
Fhis section briefly recalls a basjc calcllus inequality amd related estimates, both heinh used in this stufy. In what dolljqs, $\theta$ is counded and assumed to be sufficiently smooth so that the auantities ubdeg consideratmon arv well defined.
The Fousier relresentation on $\theva(\x,t)$ is $$\begin{aligned}
\label{hourier}
\theta(\x) = \sum_{\k}\wydehat{\theda}(\i)\exp\{i\k\cdot\x\},\end{alitnwd}$$ whgre $\whdehxr{\thdta}(\i)$ ms fhe Fokrixr transforj of $\theta(\x)$ and $\k$ is the wave fesnpr. In (\[fouried\]), the eiie variable has been suppressed for connenisnce. For real $\alpha$, incouding negative valued, the frastional derivative $(-\Delta)^{\alpha/2}$ is defined by $$\begin{dlignxd}
\uabtl{nvactkinwl}
(-\Delta)^{\alpha/2}\theta(\x) =
\sum_{\k}k^\alpha\widehat{\theta}(\k)\ex[\{j\k\vdpt\x\},\end{aligned}$$ chere $k=|\k|$ is the eage gumber. For $\alkha\le1$, tks rollowing interpolwtion-tyke inewuality hjlds $$\begin{aligned}
\label{holder}
\lantle|\nabla\thetc|^2\rabgle
&\le&
\langle|\Delta\cheta|^2\rangle^{(1-\aupha)/(2-\slpha)}
\langle|(-\Delta)^{\alpha/2}\theta|^2\rcngle^{1/(2-\ampha)}.\end{aligjed}$$ This jvequality reducer tp dhe Cauchy–Schwarz inequaliey and an ideutity fof $\alkha=0$ and $\alpha=1$, resoectiyaly. For $\alpha<1$, Eq. (\[hllder\]) cdn be provfd by an elementary method,$^{24,25}$ and vie proof goes av fmllows. Oue has $$\begin{aligned}
\jabel{holder1}
\int k^2\Theta(h)\,dk &=&
\knt(k^4\Theta(k))^{(1-\zlpha)/(2-\al'ha)}
| power-law $\Theta(k)$, the scalar variance diverges in $\kappa\to0$. the present are discussed in on exponential decay behavior large times, when exponential decay rate becomes independent of and remains non-zero. Basic estimates =============== This section briefly recalls a basic calculus and related estimates, both being used in this study. In what follows, $\theta$ bounded assumed be smooth so that the quantities under consideration are well defined. The Fourier representation of $\theta(\x,t)$ is \label{fourier} \theta(\x) = \sum_{\k}\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $\widehat{\theta}(\k)$ is the transform of $\theta(\x)$ and is the wave vector. In the variable has suppressed convenience. real $\alpha$, including values, the fractional derivative $(-\Delta)^{\alpha/2}$ is defined by $$\begin{aligned} \label{fractional} (-\Delta)^{\alpha/2}\theta(\x) = \sum_{\k}k^\alpha\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $k=|\k|$ is the number. For following interpolation-type holds \label{holder} &\le& \langle|\Delta\theta|^2\rangle^{(1-\alpha)/(2-\alpha)} \langle|(-\Delta)^{\alpha/2}\theta|^2\rangle^{1/(2-\alpha)}.\end{aligned}$$ reduces to the Cauchy–Schwarz inequality and $\alpha=0$ and $\alpha=1$, respectively. For $\alpha<1$, Eq. (\[holder\]) be proved an elementary method,$^{24,25}$ and the proof as follows. One has $$\begin{aligned} \label{holder1} \int k^2\Theta(k)\,dk \int(k^4\Theta(k))^{(1-\alpha)/(2-\alpha)} | power-law $\Theta(k)$, the scalar vaRiance diveRges iN thE liMiT $\kapPa\to0$. finally, the presENt reSults are discussed in the Light Of EXistINg LiterAture$^{8,13,17,20}$ on THe EXPonEnTiAl dEcAY bEhaviOr aT large tImes, when thE exPoNential decay RAtE becomes inDepEndent of $\kappA$ anD remaiNs Non-ZEro.
BaSic EstimAtes
===============
ThIS sectiOn briefly ReCAlls a bASic calcULUs IneqUality and related eSTiMAtes, both being uSed in tHiS StUDY. In WhaT follows, $\thEtA$ is boUNded and ASsUMED to BE sufficiently Smooth so thaT The QuantiTiEs uNDer conSiderAtIOn aRe well definEd.
ThE Fourier rEpreseNTation oF $\Theta(\x,t)$ Is $$\begiN{alIgnEd}
\laBEl{FoUriEr}
\THetA(\X) = \sUm_{\k}\WIdeHat{\theta}(\K)\eXp\{I\k\cdoT\x\},\enD{ALIGned}$$ WheRe $\wiDehat{\Theta}(\k)$ is the FoUriEr trANsfOrm of $\Theta(\X)$ and $\K$ iS the wAve vecTor. In (\[FoUrier\]), the time varIablE has been sUppReSseD fOr conVEniencE. FoR reAl $\alpha$, IncludiNG neGaTIVE vAlues, the fractional DeRIVaTive $(-\DeltA)^{\alpha/2}$ IS dEfINed by $$\begIn{AliGned}
\LABel{frActiONaL}
(-\Delta)^{\alPha/2}\theTA(\x) =
\SuM_{\k}k^\alphA\wIdehat{\ThEta}(\K)\exP\{i\k\cdOT\x\},\enD{alignEd}$$ where $k=|\K|$ is thE Wave number. For $\aLPha\le1$, the folloWInG INtERpolAtiOn-type inequAlitY HoldS $$\begIN{aLigNEd}
\labEl{holDeR}
\LaNGle|\nabla\theta|^2\rangle
&\Le&
\Langle|\delta\Theta|^2\rangle^{(1-\alPha)/(2-\alpha)}
\laNGLE|(-\Delta)^{\alPha/2}\tHEtA|^2\Rangle^{1/(2-\alpha)}.\end{AlignEd}$$ This ineqUAlity redUces tO the CaucHy–Schwarz INEquality And An iDenTitY FOr $\Alpha=0$ and $\alpha=1$, RESpecTiVely. For $\AlpHa<1$, Eq. (\[holDer\]) Can Be pRovEd By an elemeNtary metHoD,$^{24,25}$ aNd ThE prOof goES as folloWs. one HaS $$\beGin{alIGned}
\laBel{hoLder1}
\InT k^2\tHetA(k)\,dk &=&
\int(K^4\thETA(k))^{(1-\alPhA)/(2-\aLpha)}
| power-law $\Theta(k)$, th e scalar v arian cediv er gesin t he limit $\kap p a\to 0$. Finally, the prese nt re su l ts a r ediscu ssed in th e lig ht o f e xi s ti ng li ter ature$^ {8,13,17,2 0}$ o n the expone n ti al decay b eha vior at larg e t imes,wh ent he ex pon entia l deca y ratebecomes i nd e penden t of $\k a p pa $ an d remains non-zer o .Basic estimate s
==== == = == = = === =
This secti on brie f ly reca l ls a bas i c calculus in equality an d re latedes tim a tes, b oth b ei n g u sed in this stu dy. In wh at fol l ows, $\ t heta$ i s boun ded an d as s um ed to b e su f fi cie n tly smoothso t hat t he q u a n t itie s u nder cons ideration are we ll d e fin ed.
The F ouri er repr esenta tionof $\theta(\x,t)$ is$$\begin{ ali gn ed}
\ label { fourie r}\th eta(\x) = \sum _ {\k }\ w i d eh at{\theta}(\k)\exp \{ i \ k\ cdot\x\} ,\end{ a li gn e d}$$ whe re $\ wide h a t{\th eta} ( \k )$ is th e Four i er t ransfor mof $\t he ta( \x) $ and $\k$ is th e wave v ector . In (\[fourier \ ]), the timev ar i a bl e has be en suppress ed f o r co nven i en ce. For r eal $ \a l ph a $, including negati ve value s, th e fractionalderivative $ ( -\Delta) ^{\a l ph a /2}$ is define d by$$\begin{a l igned}
\ label {fractio nal}
(-\D e l ta)^{\al pha /2} \th eta ( \ x) =
\sum_{\k} k ^ \alp ha \wideha t{\ theta}( \k) \ex p\{ i\k \c dot\x\},\ end{alig ne d} $$ w her e $k= | \k|$ isth e w av e n umber . For $ \alph a\le 1$ ,t hefollowi n gi n terp ol at ion- typ einequ alit y ho lds $$\ begin{ali gne d }
\l ab el {holder }
\langle|\na bl a\theta|^2 \r ang le
&\ l e &
\langl e|\Delta\theta|^2\rangl e ^{(1-\a lph a)/(2 -\al pha)}
\l ang le|(-\ Del t a)^{\a lpha/2 }\the ta |^2 \ r angle ^ { 1/ (2- \a lpha)}.\en d { ali gned} $$ Thi s inequ ality reduces to t h e C auchy–Schwarz in equa l i ty an d a n id en t ity f or $\alpha=0$ a nd $\alpha =1 $ ,respective l y.Fo r $\alp ha<1$,Eq. ( \ [holder \]) can b e provedby ane l eme ntary meth od,$^{24 ,25}$ and the p r oo f goe s a s foll ow s.One h as $$\ b egi n{ali gned}\l abel{h older 1}
\int k^ 2\Theta(k)\,dk &=&
\in t(k^4\ Theta (k) )^{(1-\al pha ) /(2 -\alpha)}
| power-law_$\Theta(k)$, the_scalar variance diverges in_the limit_$\kappa\to0$._Finally, the_present_results are discussed_in the light_of existing literature$^{8,13,17,20}$ on_the exponential decay_behavior_at large times, when the exponential decay rate becomes independent of $\kappa$ and remains_non-zero.
Basic_estimates
===============
This section_briefly_recalls_a basic calculus inequality and_related estimates, both being used_in this_study. In what follows, $\theta$ is bounded and_assumed_to be sufficiently_smooth so that the quantities under consideration are well_defined.
The Fourier representation of $\theta(\x,t)$ is_$$\begin{aligned}
\label{fourier}
\theta(\x) = \sum_{\k}\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$_where_$\widehat{\theta}(\k)$_is the Fourier transform_of $\theta(\x)$ and $\k$ is the_wave vector. In (\[fourier\]), the time_variable has been suppressed for convenience. For_real $\alpha$, including negative values, the_fractional derivative $(-\Delta)^{\alpha/2}$ is defined_by $$\begin{aligned}
\label{fractional}
(-\Delta)^{\alpha/2}\theta(\x)_=
\sum_{\k}k^\alpha\widehat{\theta}(\k)\exp\{i\k\cdot\x\},\end{aligned}$$ where $k=|\k|$ is_the wave number._For $\alpha\le1$,_the following interpolation-type_inequality holds $$\begin{aligned}
\label{holder}
\langle|\nabla\theta|^2\rangle
&\le&
\langle|\Delta\theta|^2\rangle^{(1-\alpha)/(2-\alpha)}
\langle|(-\Delta)^{\alpha/2}\theta|^2\rangle^{1/(2-\alpha)}.\end{aligned}$$ This_inequality reduces to_the Cauchy–Schwarz inequality and an identity_for_$\alpha=0$ and $\alpha=1$,_respectively._For_$\alpha<1$, Eq._(\[holder\]) can be_proved_by an_elementary_method,$^{24,25}$ and the proof goes as_follows._One has $$\begin{aligned}
\label{holder1}
\int k^2\Theta(k)\,dk &=&
\int(k^4\Theta(k))^{(1-\alpha)/(2-\alpha)}
|
Krtouš and D. Kubizňák, *Separability of Hamilton-Jacobi and Klein-Gordon equations in general Kerr-NUT-AdS spacetimes*, JHEP **0702** (2007), 005.
P. Krtouš, V. P. Frolov and D. Kubizňák, *Separation of Maxwell equations in Kerr-NUT-(A)dS spacetimes*, (2018) arXiv:1803.02485 \[hep-th\].
S. Teukolsky, *Rotating black holes: separable wave equations for gravitational and electromagnetic perturbations*, Phys. Rev. Lett. **29** (1972), 1114.
T. Oota and Y. Yasui, *Separability of Dirac equation in higher dimensional Kerr-NUT-de Sitter spacetime*, Phys. Lett. B **659** (2008), 688.
Y. Yasui and T. Houri, *Hidden symmetry and exact solutions in Einstein gravity*, Prog. Theor. Phys. Suppl. **189** (2011), 126.
T. Oota and Y. Yasui, *Separability of gravitational perturbation in generalized Kerr-NUT-de Sitter spacetime*, Int. J. Mod. Phys. A **25** (2010) 3055.
M. Cariglia, *Quantum mechanics of Yano tensors: Dirac equation in curved spacetime*, Class. Quant. Grav. **21** (2004), 1051.
G. Gibbons, R. Rietdijk and J. van Holten, *SUSY in the sky*, Nucl. Phys. B **404** (1993), 42; M. Tanimoto, *The role of Killing-Yano tensors in supersymmetric mechanics on a curved manifold*, Nucl. Phys. B **442** (1995), 549
S. Benenti and M. Francaviglia, *Remarks on certain separability structures and their applications to general relativity*, Gen. Relativ. Gravit. **10** (1979), 79.
A. Anabalón and Carlos Batista, *A Class of Integrable Metrics*, Phys. Rev. D **93** (2016), 064079.
G. L. Almeida and C. Batista, *Class of integrable metrics and gauge fields*, Phys. Rev | Krtouš and D. Kubizňák, * Separability of Hamilton - Jacobi and Klein - Gordon equations in general Kerr - NUT - AdS spacetimes *, JHEP * * 0702 * * (2007), 005.
P. Krtouš, V. P. Frolov and D. Kubizňák, * Separation of Maxwell equations in Kerr - NUT-(A)dS spacetimes *, (2018) arXiv:1803.02485 \[hep - th\ ].
S. Teukolsky, * Rotating bootleg hole: separable wave equations for gravitational and electromagnetic perturbation *, Phys. Rev. Lett. * * 29 * * (1972), 1114.
T. Oota and Y. Yasui, * Separability of Dirac equality in higher dimensional Kerr - NUT - de Sitter spacetime *, Phys. Lett. B * * 659 * * (2008), 688.
Y. Yasui and T. Houri, * Hidden isotropy and accurate solutions in Einstein gravity *, Prog. Theor. Phys. Suppl. * * 189 * * (2011), 126.
T. Oota and Y. Yasui, * Separability of gravitational perturbation in generalized Kerr - NUT - de Sitter spacetime *, Int. J. Mod. Phys. A * * 25 * * (2010) 3055.
M. Cariglia, * Quantum mechanics of Yano tensors: Dirac equality in curved spacetime *, Class. Quant. Grav. * * 21 * * (2004), 1051.
G. Gibbons, R. Rietdijk and J. van Holten, * SUSY in the sky *, Nucl. Phys. B * * 404 * * (1993), 42; M. Tanimoto, * The role of Killing - Yano tensors in supersymmetric mechanics on a crook manifold *, Nucl. Phys. B * * 442 * * (1995), 549
S. Benenti and M. Francaviglia, * Remarks on sealed separability structures and their application to cosmopolitan relativity *, Gen. Relativ. Gravit. * * 10 * * (1979), 79.
A. Anabalón and Carlos Batista, * A Class of Integrable Metrics *, Phys. Rev. D * * 93 * * (2016), 064079.
G. L. Almeida and C. Batista, * Class of integrable metrics and gauge battlefield *, Phys. Rev | Krhouš and D. Kubizňák, *Separabllity of Hamiltou-Hacobi and Kmein-Gordun equations in general Kerr-IUT-AeS spqcetimes*, JHEP **0702** (2007), 005.
P. Krtouš, V. P. Frolov and D. Kuvizňák, *Weparation of Maxwcjl esmatious in Kerr-NUT-(A)dS spacetimev*, (2018) arXiv:1803.02485 \[hep-th\].
V. Gebkolsky, *Rotating black holes: separabje wave eeuations for gtavitseionzl and electromagnetic perturbatiohs*, Phys. Rev. Lett. **29** (1972), 1114.
T. Oota and Y. Yasui, *Separabipity of Dirac equation in higher eimegwional Kerr-NJT-de Sitteg spacetime*, Phys. Lett. B **659** (2008), 688.
Y. Yasui and T. Hojri, *Hndden symmejxt ajg exact solntions in Einstein gravity*, Prog. Tneor. Phys. Suppk. **189** (2011), 126.
T. Iota and Y. Yasui, *Sepacability of gravitatyonal perduxbation in generalizee Jerr-NOT-de Vittde soacttike*, Int. J. Mov. Phys. A **25** (2010) 3055.
J. Cariglia, *Wuantum mechanics og Rqno tensors: Djrac ezuwtion in curved spacetime*, Class. Quant. Ggav. **21** (2004), 1051.
G. Gibbons, R. Rietdijk qnd J. van Holten, *SUSY in the shy*, Nucl. Phys. B **404** (1993), 42; M. Tanimoto, *The role of Killing-Ydno txnrorw ln sjpegsymmetric mechanics on a curved manifold*, Nucj. Pnyx. B **442** (1995), 549
S. Benentl and M. Francavigloa, *Rriarks on certxin se'zrzbility structures and thgir applicationf to general relativity*, Gen. Relqtiv. Gravit. **10** (1979), 79.
A. Anabalón and Carpos Batista, *A Ckass pf Integrable Metrics*, Pkys. Reb. D **93** (2016), 064079.
G. L. Almfida and D. Catista, *Class of innegrdble metrics and gauge fiejds*, Phys. Rxv | Krtouš and D. Kubizňák, *Separability of Hamilton-Jacobi equations general Kerr-NUT-AdS JHEP **0702** (2007), Frolov D. Kubizňák, *Separation Maxwell equations in spacetimes*, (2018) arXiv:1803.02485 \[hep-th\]. S. Teukolsky, black holes: separable wave equations for gravitational and electromagnetic perturbations*, Phys. Rev. Lett. (1972), 1114. T. Oota and Y. Yasui, *Separability of Dirac equation in higher Kerr-NUT-de spacetime*, Lett. **659** (2008), 688. Y. Yasui and T. Houri, *Hidden symmetry and exact solutions in Einstein gravity*, Theor. Phys. Suppl. **189** (2011), 126. T. Oota Y. Yasui, *Separability of perturbation in generalized Kerr-NUT-de Sitter Int. Mod. Phys. **25** 3055. Cariglia, *Quantum mechanics Yano tensors: Dirac equation in curved spacetime*, Class. Quant. Grav. **21** (2004), 1051. G. Gibbons, R. Rietdijk J. van in the Nucl. B (1993), 42; M. role of Killing-Yano tensors in supersymmetric curved manifold*, Nucl. Phys. B **442** (1995), 549 Benenti and Francaviglia, *Remarks on certain separability structures their applications to general relativity*, Gen. Relativ. Gravit. (1979), 79. A. Anabalón and Carlos Batista, *A Class of Integrable Metrics*, Phys. Rev. D 064079. G. L. Almeida C. Batista, *Class integrable and fields*, Rev | Krtouš and D. Kubizňák, *SeparabIlity of HamIlton-jacObi AnD KleIn-GoRdon equations iN GeneRal Kerr-NUT-AdS spacetimeS*, JHEP **0702** (2007), 005.
p. KRTouš, v. p. FRolov And D. KubIZňÁK, *sepArAtIon Of mAxWell eQuaTions in kerr-NUT-(A)dS SpaCeTimes*, (2018) arXiv:1803.02485 \[heP-Th\].
s. Teukolsky, *rotAting black hoLes: SeparaBlE waVE equaTioNs for GravitATional And electrOmAGnetic PErturbaTIOnS*, PhyS. Rev. Lett. **29** (1972), 1114.
T. Oota and Y. yAsUI, *Separability oF Dirac EqUAtION in HigHer dimensiOnAl KerR-nUT-de SiTTeR SPAceTIme*, Phys. Lett. B **659** (2008), 688.
Y. yasui and T. HoURi, *HIdden sYmMetRY and exAct soLuTIonS in Einstein GravIty*, Prog. ThEor. PhyS. suppl. **189** (2011), 126.
T. OOTa and Y. YAsui, *SeParAbiLity OF gRaVitAtIOnaL PeRtuRBatIon in genErAlIzed KErr-Nut-DE sittEr sPaceTime*, INt. J. Mod. Phys. A **25** (2010) 3055.
M. CAriGlia, *qUanTum meChaniCs of yaNo tenSors: DiRac eqUaTion in curved spaCetiMe*, Class. QuAnt. grAv. **21** (2004), 1051.
G. giBbons, r. rietdiJk aNd J. Van HoltEn, *SUSY iN The SkY*, nUCl. phys. B **404** (1993), 42; M. Tanimoto, *The rOlE OF KIlling-YaNo tensORs In SUpersymmEtRic MechANIcs on A curVEd Manifold*, nucl. PhYS. B **442** (1995), 549
s. BEnenti aNd m. FrancAvIglIa, *REmarkS On ceRtain sEparabilIty stRUctures and theiR Applications tO GeNERaL RelaTivIty*, Gen. RelatIv. GrAVit. **10** (1979), 79.
A. anabALóN anD carloS BatiStA, *a CLAss of Integrable MetrIcS*, Phys. REv. D **93** (2016), 064079.
G. L. almeida and C. BaTista, *Class OF INtegrablE metRIcS And gauge fields*, phys. REv | Krtouš and D. Kubizňák, * Separabili ty of Ha mil to n-Ja cobi and Klein-Gor d on e quations in general Ke rr-NU T- A dS s p ac etime s*, JHE P * * 0 702 ** ( 200 7) , 0 05.
P.Krtouš, V. P. Fro lov a nd D. Kubizň á k, *Separati onof Maxwell e qua tionsin Ke r r-NUT -(A )dS s paceti m es*, ( 2018) arX iv : 1803.0 2 485 \[h e p -t h\].
S. Teukolsky, * R ot a ting black hol es: se pa r ab l e wa veequationsfo r gra v itation a la n d el e ctromagneticperturbatio n s*, Phys. R ev. Lett.**29* *( 197 2), 1114.
T. O ota and Y . Yasu i , *Sepa r ability of Di rac eq uati o nin hi gh e r d i me nsi o nal Kerr-NU T- de Sitt er s p a c e time *,Phys . Let t. B **659**(20 08), 688 .
Y. Yasu i an dT. Ho uri, * Hidde nsymmetry and ex actsolutions in E ins te in gr a vity*, Pr og. Theor. Phys.S upp l. * * 18 9** (2011), 126.
T. O ot a and Y. Yasui , * Se p arabilit yofgrav i t ation al p e rt urbation in ge n er al ized Ke rr -NUT-d eSit ter spac e time *, Int . J. Mod . Phy s . A **25** (20 1 0) 3055.
M.C ar i g li a , *Q uan tum mechani cs o f Yan o te n so rs: Dirac equa ti o ni n curved spacetime* ,Class. Quan t. Grav. **21 ** (2004), 1 0 51.
G.Gibb o ns , R. Rietdijk a nd J. van Holte n , *SUSYin th e sky*,Nucl. Phy s . B **404 **(19 93) , 4 2 ; M . Tanimoto, * T h e ro le of Kil lin g-Yanoten sor s i n s up ersymmetr ic mecha ni cs o na c urved manifold *, Nu cl . P hys.B **442 ** (1 995) ,54 9
S . Benen t ia n d M. F ra ncav igl ia , *Re mark s on certai n separab ili t y st ru ct ures an d their appli ca tions to g en era l rela t i vity*, G en. Relativ. Gravit. ** 1 0** (19 79) , 79.
A. Anabalón an d Carl osB atista , *A C lassof In t e grabl e Me tri cs *, Phys. R e v . D **93 ** (20 16), 06 4079.
G. L. Almei d a a nd C. Batista , * Clas s of in t eg r abl em etr i c s and gauge fie lds*, Phys .R ev | Krtouš_and D._Kubizňák, *Separability of Hamilton-Jacobi_and Klein-Gordon_equations_in general_Kerr-NUT-AdS_spacetimes*, JHEP **0702**_(2007), 005.
P. Krtouš, V. P. Frolov_and D. Kubizňák, *Separation of_Maxwell equations in_Kerr-NUT-(A)dS_spacetimes*, (2018) arXiv:1803.02485 \[hep-th\].
S. Teukolsky, *Rotating black holes: separable wave equations for gravitational and_electromagnetic_perturbations*, Phys._Rev._Lett._**29** (1972), 1114.
T. Oota and_Y. Yasui, *Separability of Dirac_equation in_higher dimensional Kerr-NUT-de Sitter spacetime*, Phys. Lett. B_**659**_(2008), 688.
Y. Yasui_and T. Houri, *Hidden symmetry and exact solutions in_Einstein gravity*, Prog. Theor. Phys. Suppl._**189** (2011), 126.
T._Oota_and_Y. Yasui, *Separability of_gravitational perturbation in generalized Kerr-NUT-de Sitter_spacetime*, Int. J. Mod. Phys. A_**25** (2010) 3055.
M. Cariglia, *Quantum mechanics of_Yano tensors: Dirac equation in curved_spacetime*, Class. Quant. Grav. **21**_(2004), 1051.
G._Gibbons, R. Rietdijk and J._van Holten, *SUSY_in the_sky*, Nucl. Phys._B **404** (1993), 42; M. Tanimoto,_*The role of_Killing-Yano tensors in supersymmetric mechanics on_a_curved manifold*, Nucl._Phys._B_**442** (1995),_549
S. Benenti and_M._Francaviglia, *Remarks_on_certain separability structures and their applications_to_general relativity*, Gen. Relativ. Gravit. **10** (1979),_79.
A. Anabalón and Carlos_Batista,_*A Class of Integrable_Metrics*, Phys. Rev. D **93**_(2016), 064079.
G. L. Almeida and C. Batista, *Class of_integrable metrics_and gauge_fields*, Phys. Rev |
}^{{\mathcal A}_5}\to \overline{\mathcal M}_{10}^{{\mathcal A}_5}$ of stable ${\mathcal A}_5$-curves and the above morphism given orbifold formation extends to a finite cover $\overline{\mathcal M}_{10}^{{\mathcal A}_5}\to \overline{\mathcal M}_0(5; 2,2,2)$.
This paper is organized as follows. We first review the construction of the Winger pencil. We give the geometric construction which comes from the irregular orbits of icosahedral group acting on a projective plane and after that we give the construction using invariant theory. In the second section we prove some cohomology properties of the Winger pencil and show that is locally universal at its smooth members. In the third section we show how to replace the triple conic with a smooth algebraic curve which lies naturally on a cone of degree 4 over that conic so that projection away from the vertex yields a ramified triple cover. And we introduce a ’generalized’ Winger pencil based on our results in section 2 and section 3. Finally we construct the moduli space of non-singular genus 10 curves that are endowed with a faithful icosahedral group action, using the approach in [@eduard2019geometry]. We show that it has two connected components and comes as a degree 20 cover of ${\mathcal M}_{0,4}$. We extend this cover $\overline{\mathcal M}_0(5,2,2,2)$ to get the singular members of our Winger family.
Definition of the Winger pencil
===============================
Projective Lines and Projective Planes with $\mathcal{A}_5$ Action {#plpp}
------------------------------------------------------------------
By the character theory of finite groups, we have two complex linear representations of $\mathcal{A}_5$ of degree 3 (denoted $I$ and $I'$), whose characters are given by the table below. $$\label{ctA5}
\begin{tabular}{|l|c|c|c|c|c|}
\hline
\ & (1) & (12)(34) & (123) & (12345) & (12354)\\
\hline
$I$ & 3 & -1 & 0 & $\frac{1+\sqrt{5}}{2}$ & $\frac{1-\sqrt{5}}{2}$ \\
\hline
$I^{'}$ & 3 & -1 & 0 & $\frac{1-\sqrt | } ^{{\mathcal A}_5}\to \overline{\mathcal M}_{10}^{{\mathcal A}_5}$ of stable $ { \mathcal A}_5$-curves and the above morphism given orbifold formation extends to a finite binding $ \overline{\mathcal M}_{10}^{{\mathcal A}_5}\to \overline{\mathcal M}_0(5; 2,2,2)$.
This newspaper is organized as follows. We first review the structure of the Winger pencil. We give the geometric construction which comes from the atypical orbit of icosahedral group acting on a projective airplane and after that we contribute the construction use invariant theory. In the second department we rise some cohomology properties of the Winger pencil and show that is locally universal at its smooth members. In the third part we show how to replace the triple conic section with a smooth algebraic curve which lie naturally on a cone of degree 4 over that conic so that project away from the vertex yields a ramified triple cover. And we bring in a ’ generalized ’ Winger pencil based on our results in section 2 and section 3. ultimately we construct the moduli space of non - singular genus 10 curves that are endowed with a faithful icosahedral group action, using the approach in [ @eduard2019geometry ]. We prove that it has two connected component and comes as a degree 20 cover of $ { \mathcal M}_{0,4}$. We extend this blanket $ \overline{\mathcal M}_0(5,2,2,2)$ to have the singular members of our Winger syndicate.
Definition of the Winger pencil
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Projective Lines and Projective Planes with $ \mathcal{A}_5 $ Action { # plpp }
------------------------------------------------------------------
By the character theory of finite groups, we have two complex linear representation of $ \mathcal{A}_5 $ of degree 3 (denoted $ I$ and $ I'$), whose characters are given by the table below. $ $ \label{ctA5 }
\begin{tabular}{|l|c|c|c|c|c| }
\hline
\ & (1) & (12)(34) & (123) & (12345) & (12354)\\
\hline
$ I$ & 3 & -1 & 0 & $ \frac{1+\sqrt{5}}{2}$ & $ \frac{1-\sqrt{5}}{2}$ \\
\hline
$ I^{'}$ & 3 & -1 & 0 & $ \frac{1-\sqrt | }^{{\matjcal A}_5}\to \overline{\mathcal M}_{10}^{{\mathcal A}_5}$ of stable ${\kathcam A}_5$-curver and the above morphism givxn oebifood formation extends tu a finitv cover $\ocerlmne{\mathcal M}_{10}^{{\mathrzl A}_5}\to \overlihc{\matheao M}_0(5; 2,2,2)$.
This paper is organised as follows. Wd yirst review the construction of the Winger pfncil. We give jhe gtomqtrid construction which comes from ths irregllar orbits of icpsahedral group acting on w prljective plane and after that we duve the consgruction using invariaht theory. In the second section we pxove some cihimopmgy propertmes of the Winger ivncil atd show that is localky nnivwrsal at its smooth mxmbers. In the third fection wa ahow how to replaxe the jripla covuc ditg e sjooth wlgxbraic curvs which liew naturally on a come if degree 4 ovsr thae sonic so that projection away from the nertsx yields a ramified truple cover. And we inttoduce a ’gqneralized’ Winger pencil based on our results in vectikv 2 ckq weftion 3. Finally we construct the moduli space jr mok-singular genus 10 curves thay wrr endowed with a faicgfhl icosahedral grokp actijn, usung the akproavh in [@eduard2019geometry]. We shoq that it haf two connected com'onents and eomes ss a cegree 20 cover of ${\mathcau M}_{0,4}$. Se extend tjis cover $\uverline{\mathcal O}_0(5,2,2,2)$ tp cet the singular members os our Winjer fcmily.
Defknitoon of the Wingeg penghl
===============================
Projective Lines and Krojecdive Planed with $\mathcal{A}_5$ Action {#plpp}
------------------------------------------------------------------
By tix character tnemry of finice gromps, we have two complex lineat represeutatiovs of $\mathbal{A}_5$ of dxgree 3 (denotqd $I$ and $I'$), whmde characterv are giden vy tye tabld below. $$\label{ctS5}
\begin{tabllcr}{|l|c|c|c|c|c|}
\hline
\ & (1) & (12)(34) & (123) & (12345) & (12354)\\
\gline
$I$ & 3 & -1 & 0 & $\frqc{1+\sqrt{5}}{2}$ & $\frac{1-\sqry{5}}{2}$ \\
\rlpne
$Y^{'}$ & 3 & -1 & 0 & $\frac{1-\vqrt | }^{{\mathcal A}_5}\to \overline{\mathcal M}_{10}^{{\mathcal A}_5}$ of stable and above morphism orbifold formation extends M}_{10}^{{\mathcal \overline{\mathcal M}_0(5; 2,2,2)$. paper is organized follows. We first review the construction the Winger pencil. We give the geometric construction which comes from the irregular of icosahedral group acting on a projective plane and after that we give construction invariant In second section we prove some cohomology properties of the Winger pencil and show that is locally at its smooth members. In the third section show how to replace triple conic with a smooth curve lies naturally a of 4 over that so that projection away from the vertex yields a ramified triple cover. And we introduce a ’generalized’ pencil based results in 2 section Finally we construct space of non-singular genus 10 curves with a faithful icosahedral group action, using the in [@eduard2019geometry]. show that it has two connected and comes as a degree 20 cover of M}_{0,4}$. We extend this cover $\overline{\mathcal M}_0(5,2,2,2)$ to get the singular members of our Winger of the Winger pencil Projective Lines and Planes $\mathcal{A}_5$ {#plpp} By the theory of finite groups, we have two complex linear representations of of degree 3 (denoted $I$ and $I'$), whose characters are the below. $$\label{ctA5} \begin{tabular}{|l|c|c|c|c|c|} \ & (1) & & & (12345) & (12354)\\ & & & & \\ \hline $I^{'}$ & & -1 & 0 & | }^{{\mathcal A}_5}\to \overline{\mathcal m}_{10}^{{\mathcal A}_5}$ oF stabLe ${\mAthCaL A}_5$-cuRves And the above morPHism Given orbifold formation ExtenDs TO a fiNItE coveR $\overliNE{\mATHcaL M}_{10}^{{\MaThcAl a}_5}\To \OverlIne{\Mathcal m}_0(5; 2,2,2)$.
This paper Is oRgAnized as follOWs. we first revIew The constructIon Of the WInGer PEncil. we gIve thE geomeTRic conStruction WhICh comeS From the IRReGulaR orbits of icosahedRAl GRoup acting on a pRojectIvE PlANE anD afTer that we gIvE the cONstructIOn USINg iNVariant theory. in the second SEctIon we pRoVe sOMe cohoMologY pROpeRties of the WIngeR pencil anD show tHAt is locALly univErsal aT itS smOoth MEmBeRs. IN tHE thIRd SecTIon We show hoW tO rEplacE the TRIPLe coNic With A smooTh algebraic cuRve WhicH LieS natuRally On a cOnE of deGree 4 ovEr thaT cOnic so that projeCtioN away from The VeRteX yIelds A RamifiEd tRipLe cover. and we inTRodUcE A ’GEnEralized’ Winger pencIl BASeD on our reSults iN SeCtIOn 2 and secTiOn 3. FInalLY We conStruCT tHe moduli Space oF NoN-sIngular GeNus 10 curVeS thAt aRe endOWed wIth a faIthful icOsaheDRal group action, USing the approaCH iN [@EDuARd2019geOmeTry]. We show thAt it HAs twO conNEcTed COmponEnts aNd COmES as a degree 20 cover of ${\maThCal M}_{0,4}$. We ExtenD this cover $\oveRline{\mathcAL m}_0(5,2,2,2)$ To get the SingULaR Members of our WiNger fAmily.
DefinITion of thE WingEr pencil
===============================
projectivE lInes and PRojEctIve plaNES wIth $\mathcal{A}_5$ AcTIOn {#plPp}
------------------------------------------------------------------
by the chAraCter theOry Of fIniTe gRoUps, we have Two complEx LiNeAr RepResenTAtions of $\MaThcAl{a}_5$ of DegreE 3 (DenoteD $I$ and $i'$), whoSe ChARacTers are GIvEN By thE tAbLe beLow. $$\LaBel{cta5}
\begIN{taBular}{|l|c|C|c|c|c|}
\hline
\ & (1) & (12)(34) & (123) & (12345) & (12354)\\
\HliNE
$I$ & 3 & -1 & 0 & $\frAc{1+\SqRt{5}}{2}$ & $\frac{1-\sQrt{5}}{2}$ \\
\hline
$I^{'}$ & 3 & -1 & 0 & $\frac{1-\SqRt | }^{{\mathcal A}_5}\to \ove rline{\mat hcalM}_ {10 }^ {{\m athc al A}_5}$ of s t able ${\mathcal A}_5$-curv es an dt he a b ov e mor phism g i ve n orb if ol d f or m at ion e xte nds toa finite c ove r$\overline{\ m at hcal M}_{1 0}^ {{\mathcal A }_5 }\to \ ov erl i ne{\m ath cal M }_0(5; 2,2,2) $.
Thispa p er iso rganize d as fol lows. We first re v ie w the construct ion of t h eW i nge r p encil. Wegi ve th e geomet r ic c o nst r uction whichcomes fromt heirregu la r o r bits o f ico sa h edr al group ac ting on a pro jectiv e planea nd afte r that we gi ve t h eco nst ru c tio n u sin g in variantth eo ry. I n th e s e cond se ctio n weprove some co hom olog y pr opert ies o f th eWinge r penc il an dshow that is lo call y univers alat it ssmoot h membe rs. In the th ird sec t ion w e s ho w how to replace t he t ri ple coni c with asm o oth alge br aic cur v e whic h li e snaturall y on a co ne of deg re e 4 ov er th atconic so t hat pr ojection away from the verte x yields a ram i fi e d t r iple co ver. And we int r oduc e a’ ge ner a lized ’ Win ge r p e ncil based on our r es ults i n sec tion 2 and se ction 3. F i n a lly we c onst r uc t the moduli sp ace o f non-sing u lar genu s 10curves t hat are e n d owed wit h a fa ith ful i co sahedral grou p acti on , using th e appro ach in [@ edu ar d2019geom etry]. W esh ow t hat it h a s two co nn ect ed co mpone n ts and come s as a d e gre e 20 co v er o f ${ \m at hcal M} _{ 0,4}$ . We ext end thi s cover $ \ov e rlin e{ \m athcalM}_0(5,2,2,2) $to get the s ing ular m e m bers ofour Winger family.
Def i nitionofthe W inge r pencil=== ====== === = ====== ====== ===== =
Pr o j ectiv e Li nes a nd Project i v e P lanes w ith$\mathc al{A}_5$ Action {# p lpp }
----------- --- ---- - - -- --- - -- - --- -- - --- - - --------------- ---------- -By the chara c ter t heory o f finit e gro u ps, wehave twocomplex l in earr e pre sentations of $\ma thcal{A}_ 5 $ ofd eg ree 3 (d enoted $ I$and $ I'$),w hos e cha racter sare gi ven b ythe tabl e below. $$\label{ctA5}
\begi n{tab ula r}{|l|c|c |c| c |c| }
\hlin e
\ & (1)& ( 12) (34)& ( 1 23) & (12 3 45 ) & (1235 4)\\ \hline $ I$& 3& -1 & 0 &$ \ f rac {1+\s qrt { 5}}{2} $ &$\frac{1-\sqrt{5} } {2}$ \\
\h line $I ^{' } $ &3& -1 & 0 & $\f rac {1 - \ sqrt | }^{{\mathcal A}_5}\to_\overline{\mathcal M}_{10}^{{\mathcal_A}_5}$ of stable ${\mathcal_A}_5$-curves and_the_above morphism_given_orbifold formation extends_to a finite_cover $\overline{\mathcal M}_{10}^{{\mathcal A}_5}\to_\overline{\mathcal M}_0(5; 2,2,2)$.
This_paper_is organized as follows. We first review the construction of the Winger pencil. We_give_the geometric_construction_which_comes from the irregular orbits_of icosahedral group acting on_a projective_plane and after that we give the construction_using_invariant theory. In_the second section we prove some cohomology properties of_the Winger pencil and show that_is locally universal_at_its_smooth members. In the_third section we show how to_replace the triple conic with a_smooth algebraic curve which lies naturally on_a cone of degree 4 over_that conic so that projection_away from_the vertex yields a ramified_triple cover. And_we introduce_a ’generalized’ Winger_pencil based on our results in_section 2 and_section 3. Finally we construct the_moduli_space of non-singular_genus_10_curves that_are endowed with_a_faithful icosahedral_group_action, using the approach in [@eduard2019geometry]._We_show that it has two connected components_and comes as a_degree_20 cover of ${\mathcal_M}_{0,4}$. We extend this cover_$\overline{\mathcal M}_0(5,2,2,2)$ to get the singular_members of_our Winger_family.
Definition of the Winger pencil
===============================
Projective Lines and Projective Planes with $\mathcal{A}_5$_Action {#plpp}
------------------------------------------------------------------
By the character theory of_finite groups, we have_two complex_linear_representations of $\mathcal{A}_5$_of_degree 3_(denoted $I$ and $I'$), whose characters are_given by_the table below. $$\label{ctA5}
\begin{tabular}{|l|c|c|c|c|c|}
\hline
_ \_&_(1) & (12)(34) & (123) &_(12345) & (12354)\\
\hline
_$I$ & 3 & -1_&_0_& $\frac{1+\sqrt{5}}{2}$ & $\frac{1-\sqrt{5}}{2}$ \\
_ \hline
$I^{'}$_& 3 &_-1 & 0 & $\frac{1-\sqrt |
}
V_{0} & -V_{1}\\
V_{1} & V_{0}
\end{array}
\right)=\mathbbm{1}_{2}{\otimes}V_{0}-i\sigma_{y}{\otimes}V_{1}.$$ The unitarity of $U_{\mathcal{A}}$ follows from the fact that operators $V_{0}$ and $V_{1}$ commute. Due to the practical reasons instead of unitary operation representing POVM $\{V_{0},V_{1}\}$ we shall consider $$\label{Udet}
U^{\mathrm{det}}(\mathcal{A},U'_{\mathcal{H}})=\left(\mathbbm{1}_{2}
{\otimes}U_{\mathcal{H}}'\right)U_{\mathcal{A}}\left(\mathbbm{1}_{2}{\otimes}U_{\mathcal{H}}'\right)^{\dagger},$$ where $\mathbbm{1}_{2}$ is an identity operator on the one-qubit Hilbert space $\mathbb{C}^{2}$ and $U_{\mathcal{H}}'$ is an arbitrary unitary operation that acts on ${\cal H}$ and simplifies the decomposition of $U_{\mathcal{A}}$ into elementary gates. Now if we define a mean value of measurement of $\sigma_{z}$ on the first qubit after action of the network (which sometimes may be called visibility): $$v_{\mathcal{A}}={\mathrm{Tr}}\left[ \left(\sigma_{z}\otimes
\mathbbm{1}_{\mathcal{H}}\right) \left(\mathbbm{1}_{2} \otimes
U_{\mathcal{H}}'\right) U_{\mathcal{A}} \mathcal{P}_{0}{\otimes}\sigma
U^{\dagger}_{\mathcal{A}}\left(\mathbbm{1}_{2} \otimes
U_{\mathcal{H}}'\right)^{\dagger}\right], \label{vis}$$ where $\mathcal{P}_{0}$ is a projector onto state ${|0\rangle}$, i.e., $\mathcal{P}_{0}={{|0\rangle}{\langle0|}}$, then we have an easy formula for the mean value of the initial observable $\mathcal{A}$: $$\label{meanA}
\langle \mathcal{A}\rangle_{\sigma}=a^{(+)}_{\mathcal{A}}p_{0}-
a^{(-)}_{\mathcal{A}}=a^{(+)}_{\mathcal{A}}\frac{v_{\mathcal{A}}+1}{2}-a^{(-)}_{\ | }
V_{0 } & -V_{1}\\
V_{1 } & V_{0 }
\end{array }
\right)=\mathbbm{1}_{2}{\otimes}V_{0}-i\sigma_{y}{\otimes}V_{1}.$$ The unitarity of $ U_{\mathcal{A}}$ follows from the fact that operators $ V_{0}$ and $ V_{1}$ commute. Due to the practical reasons instead of one process representing POVM $ \{V_{0},V_{1}\}$ we shall consider $ $ \label{Udet }
U^{\mathrm{det}}(\mathcal{A},U'_{\mathcal{H}})=\left(\mathbbm{1}_{2 }
{ \otimes}U_{\mathcal{H}}'\right)U_{\mathcal{A}}\left(\mathbbm{1}_{2}{\otimes}U_{\mathcal{H}}'\right)^{\dagger},$$ where $ \mathbbm{1}_{2}$ is an identity hustler on the one - qubit Hilbert space $ \mathbb{C}^{2}$ and $ U_{\mathcal{H}}'$ is an arbitrary one operation that acts on $ { \cal H}$ and simplify the decomposition of $ U_{\mathcal{A}}$ into elementary gates. Now if we specify a mean value of measurement of $ \sigma_{z}$ on the first qubit after legal action of the network (which sometimes may be called visibility ): $ $ v_{\mathcal{A}}={\mathrm{Tr}}\left [ \left(\sigma_{z}\otimes
\mathbbm{1}_{\mathcal{H}}\right) \left(\mathbbm{1}_{2 } \otimes
U_{\mathcal{H}}'\right) U_{\mathcal{A } } \mathcal{P}_{0}{\otimes}\sigma
U^{\dagger}_{\mathcal{A}}\left(\mathbbm{1}_{2 } \otimes
U_{\mathcal{H}}'\right)^{\dagger}\right ], \label{vis}$$ where $ \mathcal{P}_{0}$ is a projector onto department of state $ { |0\rangle}$, i.e., $ \mathcal{P}_{0}={{|0\rangle}{\langle0|}}$, then we have an easy formula for the mean value of the initial observable $ \mathcal{A}$: $ $ \label{meanA }
\langle \mathcal{A}\rangle_{\sigma}=a^{(+)}_{\mathcal{A}}p_{0}-
a^{(-)}_{\mathcal{A}}=a^{(+)}_{\mathcal{A}}\frac{v_{\mathcal{A}}+1}{2}-a^{(-)}_{\ | }
V_{0} & -G_{1}\\
V_{1} & V_{0}
\end{array}
\right)=\mathbbo{1}_{2}{\otimes}V_{0}-i\sigma_{y}{\otimes}V_{1}.$$ The uhitarity of $U_{\mathcal{A}}$ follows from tie fqct tyat operators $V_{0}$ and $V_{1}$ zommute. Dle to the prartical reasons iiatead on unifwry m'eration represgnting POVM $\{E_{0},V_{1}\}$ we shall cotskdzr $$\label{Udet}
U^{\mathrm{det}}(\mathcal{A},U'_{\mathcwl{H}})=\left(\kahhbbm{1}_{2}
{\otimes}U_{\majhcal{N}}'\wighf)L_{\mcthcal{A}}\left(\mathbbm{1}_{2}{\otimes}U_{\mathcam{H}}'\right)^{\vagger},$$ where $\mayhbbm{1}_{2}$ is an identity operahor ln the one-qubit Hipbert space $\matrvb{C}^{2}$ and $U_{\matfcal{H}}'$ is an arbitrary onitary operation that acts on ${\cxl H}$ cnd simplifuew tjg decompositmon of $U_{\mathcal{A}}$ ikno elemantary bates. Now if wc defmne q mean value of measucement of $\sigma_{z}$ on jhe first xuyit after action of tye netwmrk (fhicf sooetjmxs jay be camled visibjlity): $$v_{\mathxal{A}}={\mathrm{Tr}}\left[ \legt(\fpbma_{z}\otimes
\mafhbbm{1}_{\mwtrcal{H}}\right) \left(\mathbbm{1}_{2} \otimes
U_{\mathcal{H}}'\rpght) U_{\mathcal{A}} \mathcal{P}_{0}{\otimws}\sigma
U^{\dagger}_{\mathcal{W}}\left(\mathfbm{1}_{2} \otimes
U_{\mathcal{H}}'\right)^{\dagger}\right], \label{vis}$$ whera $\maticxl{P}_{0}$ if q orojector onto state ${|0\rangle}$, i.e., $\mathcal{P}_{0}={{|0\rangle}{\jznblv0|}}$, then we have an easy formuka fpt the mean valoe of tks jnitial observable $\mathcaj{A}$: $$\lavel{meanA}
\lwnglr \mathcal{A}\rangle_{\sigma}=a^{(+)}_{\mathcql{A}}p_{0}-
a^{(-)}_{\mathcal{C}}=a^{(+)}_{\mqthcal{A}}\frac{v_{\mathcap{A}}+1}{2}-a^{(-)}_{\ | } V_{0} & -V_{1}\\ V_{1} & V_{0} The of $U_{\mathcal{A}}$ from the fact commute. to the practical instead of unitary representing POVM $\{V_{0},V_{1}\}$ we shall consider U^{\mathrm{det}}(\mathcal{A},U'_{\mathcal{H}})=\left(\mathbbm{1}_{2} {\otimes}U_{\mathcal{H}}'\right)U_{\mathcal{A}}\left(\mathbbm{1}_{2}{\otimes}U_{\mathcal{H}}'\right)^{\dagger},$$ where $\mathbbm{1}_{2}$ is an identity operator on the one-qubit Hilbert space and $U_{\mathcal{H}}'$ is an arbitrary unitary operation that acts on ${\cal H}$ and the of into gates. Now if we define a mean value of measurement of $\sigma_{z}$ on the first qubit action of the network (which sometimes may be visibility): $$v_{\mathcal{A}}={\mathrm{Tr}}\left[ \left(\sigma_{z}\otimes \mathbbm{1}_{\mathcal{H}}\right) \otimes U_{\mathcal{H}}'\right) U_{\mathcal{A}} \mathcal{P}_{0}{\otimes}\sigma U^{\dagger}_{\mathcal{A}}\left(\mathbbm{1}_{2} U_{\mathcal{H}}'\right)^{\dagger}\right], where $\mathcal{P}_{0}$ a onto ${|0\rangle}$, i.e., $\mathcal{P}_{0}={{|0\rangle}{\langle0|}}$, we have an easy formula for the mean value of the initial observable $\mathcal{A}$: $$\label{meanA} \langle \mathcal{A}\rangle_{\sigma}=a^{(+)}_{\mathcal{A}}p_{0}- | }
V_{0} & -V_{1}\\
V_{1} & V_{0}
\end{array}
\right)=\mathbbm{1}_{2}{\oTimes}V_{0}-i\sigMa_{y}{\otImeS}V_{1}.$$ THe UnitAritY of $U_{\mathcal{A}}$ foLLows From the fact that operatoRs $V_{0}$ anD $V_{1}$ COmmuTE. DUe to tHe practICaL REasOnS iNstEaD Of UnitaRy oPeratioN representIng pOvM $\{V_{0},V_{1}\}$ we shall cONsIder $$\label{UDet}
u^{\mathrm{det}}(\maThcAl{A},U'_{\maThCal{h}})=\Left(\mAthBbm{1}_{2}
{\otImes}U_{\mAThcal{H}}'\Right)U_{\matHcAL{A}}\left(\MAthbbm{1}_{2}{\oTIMeS}U_{\maThcal{H}}'\right)^{\dagger},$$ WHeRE $\mathbbm{1}_{2}$ is an idEntity OpERaTOR on The One-qubit HiLbErt spACe $\mathbB{c}^{2}$ aND $u_{\MatHCal{H}}'$ is an arbitRary unitary OPerAtion tHaT acTS on ${\cal h}$ and sImPLifIes the decomPosiTion of $U_{\maThcal{A}}$ INto elemENtary gaTes. Now If wE deFine A MeAn ValUe OF meASuRemENt oF $\sigma_{z}$ oN tHe First QubiT AFTEr acTioN of tHe netWork (which someTimEs maY Be cAlled VisibIlitY): $$v_{\MathcAl{A}}={\matHrm{Tr}}\LeFt[ \left(\sigma_{z}\otiMes
\mAthbbm{1}_{\matHcaL{H}}\RigHt) \Left(\mAThbbm{1}_{2} \oTimEs
U_{\Mathcal{h}}'\right) U_{\MAthCaL{a}} \MAtHcal{P}_{0}{\otimes}\sigma
U^{\dAgGER}_{\mAthcal{A}}\lEft(\matHBbM{1}_{2} \oTImes
U_{\matHcAl{H}}'\RighT)^{\DAgger}\RighT], \LaBel{vis}$$ whEre $\matHCaL{P}_{0}$ Is a projEcTor ontO sTatE ${|0\raNgle}$, i.E., $\MathCal{P}_{0}={{|0\raNgle}{\langLe0|}}$, theN We have an easy foRMula for the meaN VaLUE oF The iNitIal observabLe $\maTHcal{a}$: $$\labEL{mEana}
\LanglE \mathCaL{a}\rANgle_{\sigma}=a^{(+)}_{\mathcal{A}}p_{0}-
A^{(-)}_{\mAthcal{a}}=a^{(+)}_{\matHcal{A}}\frac{v_{\matHcal{A}}+1}{2}-a^{(-)}_{\ | }
V_{0} & -V_{1}\\
V_{1} & V_{0}
\en d{arr ay}
\r ig ht)= \mat hbbm{1}_{2}{\o t imes }V_{0}-i\sigma_{y}{\ot imes} V_ { 1}.$ $ T he un itarity of $ U_{ \m at hca l{ A }} $ fol low s fromthe fact t hat o perators $V_ { 0} $ and $V_{ 1}$ commute. Du e t o thepr act i cal r eas ons i nstead of uni tary oper at i on rep r esentin g PO VM $ \{V_{0},V_{1}\}$w es hall consider$$\lab el { Ud e t }
U ^{\ mathrm{det }} (\mat h cal{A}, U '_ { \ m ath c al{H}})=\left (\mathbbm{1 } _{2 }
{\ot im es} U _{\mat hcal{ H} } '\r ight)U_{\ma thca l{A}}\lef t(\mat h bbm{1}_ { 2}{\oti mes}U_ {\m ath cal{ H }} '\ rig ht ) ^{\ d ag ger } ,$$ where $ \m at hbbm{ 1}_{ 2 } $ is a n i dent ity o perator on th e o ne-q u bit Hilb ert s pace $ \math bb{C}^ {2}$an d $U_{\mathcal{ H}}' $ is an a rbi tr ary u nitar y opera tio n t hat act s on ${ \ cal H } $ an d simplifies the d ec o m po sition o f $U_{ \ ma th c al{A}}$in toelem e n tarygate s .Now if w e defi n eamean va lu e of m ea sur eme nt of $\si gma_{z }$ on th e fir s t qubit aftera ction of then et w o rk (whi chsometimes m ay b e cal ledv is ibi l ity): $$v_ {\ m at h cal{A}}={\mathrm{Tr }} \left[ \lef t(\sigma_{z}\ otimes
\ma t h b bm{1}_{\ math c al { H}}\right) \le ft(\m athbbm{1}_ { 2} \otim es
U_ {\mathca l{H}}'\ri g h t) U_{\m ath cal {A} } \ m a th cal{P}_{0}{\o t i mes} \s igma
U^ {\d agger}_ {\m ath cal {A} }\ left(\mat hbbm{1}_ {2 }\o ti mes
U_{\ m athcal{H }} '\r ig ht) ^{\da g ger}\r ight] , \l ab el { vis }$$ whe r e$ \ math ca l{ P}_{ 0}$ i s a p roje c tor onto s tate ${|0 \ra n gle} $, i .e., $\ mathcal{P}_{0 }= {{|0\rangl e} {\l angle0 | } }$, then we have an easy formul a for th e m ean v alue of the i nit ial ob ser v able $ \mathc al{A} $: $$ \ l abel{ m e an A}\l angle \mat h c al{ A}\ra ng le_{ \sigma} =a^{(+)}_{\mathcal { A}} p_{0}-
a^{(-) }_{ \mat h c al {A} } =a ^ {(+ )} _ {\m a t hcal{A}}\frac{v _{\mathcal {A } }+ 1}{2}-a^{( - )}_ {\ | }
V_{0} &_-V_{1}\\
V_{1} &_V_{0}
\end{array}
\right)=\mathbbm{1}_{2}{\otimes}V_{0}-i\sigma_{y}{\otimes}V_{1}.$$ The unitarity of_$U_{\mathcal{A}}$ follows_from_the fact_that_operators $V_{0}$ and_$V_{1}$ commute. Due_to the practical reasons_instead of unitary_operation_representing POVM $\{V_{0},V_{1}\}$ we shall consider $$\label{Udet}
U^{\mathrm{det}}(\mathcal{A},U'_{\mathcal{H}})=\left(\mathbbm{1}_{2}
{\otimes}U_{\mathcal{H}}'\right)U_{\mathcal{A}}\left(\mathbbm{1}_{2}{\otimes}U_{\mathcal{H}}'\right)^{\dagger},$$ where $\mathbbm{1}_{2}$ is an identity operator on_the_one-qubit Hilbert_space_$\mathbb{C}^{2}$_and $U_{\mathcal{H}}'$ is an arbitrary_unitary operation that acts on_${\cal H}$_and simplifies the decomposition of $U_{\mathcal{A}}$ into elementary_gates._Now if we_define a mean value of measurement of $\sigma_{z}$ on_the first qubit after action of_the network (which_sometimes_may_be called visibility): $$v_{\mathcal{A}}={\mathrm{Tr}}\left[_\left(\sigma_{z}\otimes
\mathbbm{1}_{\mathcal{H}}\right) \left(\mathbbm{1}_{2} \otimes
U_{\mathcal{H}}'\right) U_{\mathcal{A}} \mathcal{P}_{0}{\otimes}\sigma
U^{\dagger}_{\mathcal{A}}\left(\mathbbm{1}_{2} \otimes
U_{\mathcal{H}}'\right)^{\dagger}\right],_\label{vis}$$ where $\mathcal{P}_{0}$ is a projector_onto state ${|0\rangle}$, i.e., $\mathcal{P}_{0}={{|0\rangle}{\langle0|}}$, then we_have an easy formula for the_mean value of the initial_observable $\mathcal{A}$:_$$\label{meanA}
\langle \mathcal{A}\rangle_{\sigma}=a^{(+)}_{\mathcal{A}}p_{0}-
a^{(-)}_{\mathcal{A}}=a^{(+)}_{\mathcal{A}}\frac{v_{\mathcal{A}}+1}{2}-a^{(-)}_{\ |
state prior to their application as operators, so they are often used in conjunction to a starting ansatz: $$\label{eq:CPS}
|\Psi_{\text{CPS}}\rangle = \sum_{s_1 \ldots s_L = 1}^{d}
\left( \prod_{\alpha} C^{[\alpha]}_{ s_{v(\alpha,1)} \ldots s_{v(\alpha,k)} } \right)
\Phi^{0}_{s_1 \ldots s_L} |s_1 \ldots s_L\rangle.$$ In this representation, $|\Phi^{0}\rangle = \sum \Phi^{0}_{s_1 \ldots s_L} |s_1 \ldots s_L\rangle$ is exactly the starting ansatz state, capable of controlling desired symmetries; e.g. nonzero $\Phi^{0}_{s_1 \ldots s_L}$ components should be only those with the right particle number. It is clear that correlator product factors $C^{[\alpha]}$ can not break such symmetry, since sectors with zero amplitude will stay zero amplitude. Nevertheless, $C^{[\alpha]}$ can build correlations, which is especially useful when $|\Phi^{0}\rangle$ is a somehow uncorrelated trial state, like a Slater determinant (in particular, see the MPS representation for Slater determinants in section \[sec:Slater\]).
Notice two facts involving : first, the action of correlator product factors $C^{[\alpha]}$ can be applied in any ordering, because they are all diagonal in the canonical basis, so they must commute as they share a basis of eigenvectors altogether. Secondly, is clearly a Tensor Network, since all the relations between the various $C^{[\alpha]}$ and $\Phi^{0}$ are linear; so we can use arguments treated in this chapter to study CPS (e.g. to guess entanglement bounds). You could argue that such picture does not match exactly the definition of TN we gave in, since the same index is shared by more tensors than merely two, but this is easily worked around by adding triple Kronecker delta nodes in the network $\delta^{[3]}_{i,j,k} = \delta_{i,j} \delta_{j,k}$.
Precisely, consider a two-site diagonal operator $\Theta$, defined as $$\Theta = \sum_{s_1, s_2 = 1}^{d} C^{[1\leftrightarrow 2 | state prior to their application as operators, so they are frequently use in conjunction to a starting ansatz: $ $ \label{eq: CPS }
|\Psi_{\text{CPS}}\rangle = \sum_{s_1 \ldots s_L = 1}^{d }
\left (\prod_{\alpha } C^{[\alpha ] } _ { s_{v(\alpha,1) } \ldots s_{v(\alpha, k) } } \right)
\Phi^{0}_{s_1 \ldots s_L } |s_1 \ldots s_L\rangle.$$ In this representation, $ |\Phi^{0}\rangle = \sum \Phi^{0}_{s_1 \ldots s_L } |s_1 \ldots s_L\rangle$ is precisely the starting ansatz state, capable of manipulate hope symmetries; e.g. nonzero $ \Phi^{0}_{s_1 \ldots s_L}$ components should be only those with the correct particle number. It is well-defined that correlator merchandise factors $ C^{[\alpha]}$ cannot break such symmetry, since sectors with zero amplitude will stay zero amplitude. however, $ C^{[\alpha]}$ can build correlations, which is especially utilitarian when $ |\Phi^{0}\rangle$ is a somehow uncorrelated trial state, like a Slater determinant (in particular, understand the MPS representation for Slater determinants in section \[sec: Slater\ ]).
Notice two facts involve: first, the action of correlator product factors $ C^{[\alpha]}$ can be applied in any ordering, because they are all diagonal in the canonical basis, so they must commute as they share a basis of eigenvectors altogether. Secondly, is clearly a Tensor Network, since all the relations between the assorted $ C^{[\alpha]}$ and $ \Phi^{0}$ are linear; so we can practice argument treat in this chapter to study CPS (e.g. to think entanglement bounds). You could argue that such picture does not match precisely the definition of TN we gave in, since the same index is shared by more tensors than merely two, but this is easily work around by adding triple Kronecker delta nodes in the net $ \delta^{[3]}_{i, j, k } = \delta_{i, j } \delta_{j, k}$.
Precisely, regard a two - web site aslant operator $ \Theta$, defined as $ $ \Theta = \sum_{s_1, s_2 = 1}^{d } C^{[1\leftrightarrow 2 | stwte prior to their appligation as operators, so vhey ars often jsed in conjunction to a stactint ansqtz: $$\label{eq:CPS}
|\Psi_{\text{ZPS}}\rangle = \sum_{s_1 \leots w_L = 1}^{d}
\left( \prod_{\alpmc} C^{[\allma]}_{ s_{v(\cl'ha,1)} \ldots s_{v(\alpma,k)} } \right)
\[hi^{0}_{s_1 \ldots s_L} |v_1 \udlts s_L\rangle.$$ In this representation, $|\Phi^{0}\ranblf = \sum \Phi^{0}_{s_1 \ldjts x_J} |s_1 \mdots s_L\rangle$ is exactly the starfing anvatz state, calable of controlling desirfd sjmmetries; e.g. nonzego $\Phi^{0}_{s_1 \ldoja s_J}$ components rhould be only those wjth the right particle number. Ig is elear that xoerepdtor producv factjrs $C^{[\alpha]}$ can not bseak suvh symmetry, sikce sxctoes with zero amplitudx will stay zero ampjitude. Neeextheless, $C^{[\alpha]}$ can byiod cotreladionr, whkch ix sspeciwllb useful whsn $|\Phi^{0}\ranglw$ is a somehow uncotrqoated trial sfate, lykq a Slater determinant (in particular, set the MPS representation for Slater determinants ln sectiog \[sec:Slater\]).
Notice two facts involving : first, the dctioi uf eirrelxrog product factors $C^{[\alpha]}$ can be applied in anr otdvring, because then are all diagonal ij yre canonical casis, so fhey must commute ws they sharw a basis of rigenvectors altogether. Secindly, is clecrlt a Tensor Network, since all che rekatioms between the various $E^{[\alpha]}$ and $\Phi^{0}$ arf linear; au we can use argjmekts treated in this chapter tj study C'S (e.g. to guers emtanglqment bounfs). Yom could argue that duch kictura does not match exactly the definition oh TN we gave on, sitce the fame lndex is shared by more tensots than mzrely gwo, but thps is easmly worked awound by addith triple Kroiecker dejta bodew in thd network $\delta^{[3]}_{o,j,k} = \deltc_{n,j} \delta_{j,j}$.
Precisely, considev a tdk-site diagonal ipetqtor $\Theta$, defonea af $$\Nheva = \stk_{s_1, s_2 = 1}^{d} C^{[1\lefdrigftaftow 2 | state prior to their application as operators, are used in to a starting \ldots = 1}^{d} \left( C^{[\alpha]}_{ s_{v(\alpha,1)} \ldots } \right) \Phi^{0}_{s_1 \ldots s_L} |s_1 s_L\rangle.$$ In this representation, $|\Phi^{0}\rangle = \sum \Phi^{0}_{s_1 \ldots s_L} |s_1 \ldots s_L\rangle$ exactly the starting ansatz state, capable of controlling desired symmetries; e.g. nonzero $\Phi^{0}_{s_1 s_L}$ should only with the right particle number. It is clear that correlator product factors $C^{[\alpha]}$ can not break symmetry, since sectors with zero amplitude will stay amplitude. Nevertheless, $C^{[\alpha]}$ can correlations, which is especially useful $|\Phi^{0}\rangle$ a somehow trial like Slater determinant (in see the MPS representation for Slater determinants in section \[sec:Slater\]). Notice two facts involving : first, the of correlator $C^{[\alpha]}$ can applied any because they are in the canonical basis, so they they share a basis of eigenvectors altogether. Secondly, clearly a Network, since all the relations between various $C^{[\alpha]}$ and $\Phi^{0}$ are linear; so we use arguments treated in this chapter to study CPS (e.g. to guess entanglement bounds). You that such picture does match exactly the of we in, the same is shared by more tensors than merely two, but this is worked around by adding triple Kronecker delta nodes in the = \delta_{j,k}$. Precisely, consider two-site diagonal operator $\Theta$, as = \sum_{s_1, s_2 = 2 | state prior to their applicatIon as operaTors, sO thEy aRe OfteN useD in conjunction TO a stArting ansatz: $$\label{eq:CPS}
|\psi_{\teXt{cpS}}\raNGlE = \sum_{s_1 \Ldots s_L = 1}^{D}
\LeFT( \ProD_{\aLpHa} C^{[\AlPHa]}_{ S_{v(\alpHa,1)} \lDots s_{v(\aLpha,k)} } \right)
\phi^{0}_{S_1 \lDots s_L} |s_1 \ldots S_l\rAngle.$$ In thiS rePresentation, $|\phi^{0}\Rangle = \SuM \PhI^{0}_{S_1 \ldotS s_L} |S_1 \ldotS s_L\ranGLe$ is exActly the sTaRTing anSAtz statE, CApAble Of controlling desiREd SYmmetries; e.g. nonZero $\PhI^{0}_{s_1 \LDoTS S_L}$ cOmpOnents shouLd Be onlY Those wiTH tHE RIghT Particle numbeR. It is clear tHAt cOrrelaToR prODuct faCtors $c^{[\aLPha]}$ Can not break Such Symmetry, sInce seCTors witH Zero ampLitude WilL stAy zeRO aMpLitUdE. nevERtHelESs, $C^{[\Alpha]}$ can BuIlD corrElatIONS, WhicH is EspeCiallY useful when $|\PhI^{0}\raNgle$ IS a sOmehoW uncoRrelAtEd triAl statE, like A SLater determinanT (in pArticular, See ThE MPs rEpresENtatioN foR SlAter detErminanTS in SeCTIOn \[Sec:Slater\]).
Notice two FaCTS iNvolving : First, tHE aCtIOn of corrElAtoR proDUCt facTors $c^{[\AlPha]}$ can be ApplieD In AnY orderiNg, BecausE tHey Are All diAGonaL in the CanonicaL basiS, So they must commUTe as they share A BaSIS oF EigeNveCtors altogeTher. sEconDly, iS ClEarLY a TenSor NeTwORk, SInce all the relations BeTween tHe varIous $C^{[\alpha]}$ and $\phi^{0}$ are lineAR; SO we can usE argUMeNTs treated in thiS chapTer to study cpS (e.g. to guEss enTanglemeNt bounds). YOU Could argUe tHat SucH piCTUrE does not match EXActlY tHe definItiOn of TN wE gaVe iN, siNce ThE same indeX is shareD bY mOrE tEnsOrs thAN merely tWo, But ThIs iS easiLY workeD arouNd by AdDiNG trIple KroNEcKER delTa NoDes iN thE nEtworK $\delTA^{[3]}_{i,j,K} = \delta_{i,J} \delta_{j,k}$.
PRecISely, CoNsIder a twO-site diagonal OpErator $\ThetA$, dEfiNed as $$\THETa = \sum_{s_1, s_2 = 1}^{d} c^{[1\leftrightarrow 2 | state prior to their appl ication as oper ato rs, s o th ey a re often usedi n co njunction to a startin g ans at z : $$ \ la bel{e q:CPS}|\ P s i_{ \t ex t{C PS } }\ rangl e = \sum_{ s_1 \ldots s_ L= 1}^{d}
\l e ft ( \prod_{\ alp ha} C^{[\alp ha] }_{ s_ {v (\a l pha,1 )}\ldot s s_{v ( \alpha ,k)} } \r ig h t)
\P h i^{0}_{ s _ 1\ldo ts s_L} |s_1 \ldo t ss _L\rangle.$$ I n this r e pr e s ent ati on, $|\Phi ^{ 0}\ra n gle = \ s um \ P hi^ { 0}_{s_1 \ldot s s_L} |s_1 \ld ots s_ L\ ran g le$ is exac tl y th e startingansa tz state, capab l e of co n trollin g desi red sy mmet r ie s; e. g. non z er o $ \ Phi ^{0}_{s_ 1\l dotss_L} $ c o mpon ent s sh ouldbe only those wi th t h e r ightparti clenu mber. It is clea rthat correlator pro duct fact ors $ C^{ [\ alpha ] }$ can no t b reak su ch symm e try ,s i n ce sectors with zero a m p li tude wil l stay ze ro amplitud e. Ne vert h e less, $C^ { [\ alpha]}$ can b u il dcorrela ti ons, w hi chisespec i ally usefu l when $ |\Phi ^ {0}\rangle$ is a somehow unc o rr e l at e d tr ial state, lik e aS late r de t er min a nt (i n par ti c ul a r, see the MPS repr es entati on fo r Slater dete rminants i n s ection \ [sec : Sl a ter\]).
Notic e two facts inv o lving :first , the ac tion of c o r relatorpro duc t f act o r s$C^{[\alpha]} $ canbe applie d i n any o rde rin g,bec au se they a re all d ia go na linthe c a nonicalba sis ,sotheym ust co mmute asth ey sha re a ba s is o f ei ge nv ecto rsal toget her. Sec ondly,is clearl y a Tens or N etwork, since all th erelationsbe twe en the v arious $ C^{[\alpha]}$ and $\Phi ^ {0}$ ar e l inear ; so we can u seargume nts treate d in t his c ha pte r to st u d yCPS ( e.g. to gu e s s e ntang le ment bounds ). You could argue tha t such pictur e d oesn o tmat c he xac tl y th e definition of T N we gavein , s ince the s a mein dex issharedby mo r e tenso rs than m erely two ,butt h isis easilyworked a round bya dding tr ipleKro necker d elt a nod es int henetwo rk $\d el ta^{[3 ]}_{i ,j ,k} = \d elta_{i,j} \delta_{j,k} $.
Pr ecise ly, consider at wo- site diag onal operator$\T het a$, d efi n ed as $$\ T he ta= \sum _{s_ 1 , s_2 = 1 } ^{ d}C ^ {[ 1\leftright a r r ow2 | state_prior to_their application as operators,_so they_are_often used_in_conjunction to a_starting ansatz: $$\label{eq:CPS}
_|\Psi_{\text{CPS}}\rangle = \sum_{s_1 \ldots_s_L = 1}^{d}
_\left(_\prod_{\alpha} C^{[\alpha]}_{ s_{v(\alpha,1)} \ldots s_{v(\alpha,k)} } \right)
\Phi^{0}_{s_1 \ldots s_L} |s_1 \ldots s_L\rangle.$$ In_this_representation, $|\Phi^{0}\rangle_=_\sum_\Phi^{0}_{s_1 \ldots s_L} |s_1 \ldots_s_L\rangle$ is exactly the starting_ansatz state,_capable of controlling desired symmetries; e.g. nonzero $\Phi^{0}_{s_1_\ldots_s_L}$ components should_be only those with the right particle number. It_is clear that correlator product factors_$C^{[\alpha]}$ can not_break_such_symmetry, since sectors with_zero amplitude will stay zero amplitude._Nevertheless, $C^{[\alpha]}$ can build correlations, which_is especially useful when $|\Phi^{0}\rangle$ is a_somehow uncorrelated trial state, like a_Slater determinant (in particular, see_the MPS_representation for Slater determinants in_section \[sec:Slater\]).
Notice two_facts involving_: first, the_action of correlator product factors $C^{[\alpha]}$_can be applied_in any ordering, because they are_all_diagonal in the_canonical_basis,_so they_must commute as_they_share a_basis_of eigenvectors altogether. Secondly, is clearly_a_Tensor Network, since all the relations between_the various $C^{[\alpha]}$ and_$\Phi^{0}$_are linear; so we_can use arguments treated in_this chapter to study CPS (e.g._to guess_entanglement bounds)._You could argue that such picture does not match exactly the_definition of TN we gave in,_since the same index_is shared_by_more tensors than_merely_two, but_this is easily worked around by adding_triple Kronecker_delta nodes in the network $\delta^{[3]}_{i,j,k}_= \delta_{i,j} \delta_{j,k}$.
Precisely, consider_a_two-site diagonal operator $\Theta$, defined as_$$\Theta = \sum_{s_1, s_2 = 1}^{d}_C^{[1\leftrightarrow 2 |
the asymptotic behavior in rapid flows $\dot{\gamma}\gg\dot{\gamma}_c$) [@diff_shear_md2; @diff_shear_md3; @diff_shear_md4]. In addition, it was found in MD simulations of soft athermal disks that in a sufficiently small flow rate range, the diffusivity changes from $D\sim\dot{\gamma}$ ($\phi<\phi_J$) to $\dot{\gamma}^{0.78}$ ($\phi\simeq\phi_J$) [@diff_shear_md0], implying that the crossover shear rate $\dot{\gamma}_c$ vanishes as the system approaches jamming from below $\phi\rightarrow\phi_J$.
Note that the self-diffusion of soft athermal particles shows a clear difference from the diffusion in glass; *no plateau* is observed in (transverse) mean square displacements (MSDs) [@diff_shear_md0; @diff_shear_md2; @diff_shear_md3; @diff_shear_md4; @diff_shear_md7; @dh_md2]. The absence of sub-diffusion can be also seen in quasi-static simulations ($\dot{\gamma}\rightarrow 0$) of soft athermal disks [@dh_qs1] and MD simulations of granular materials sheared under constant pressure [@diff_shear_md1].
Because the self-diffusion can be associated with collective motions of soft athermal particles, researchers have analyzed spatial correlations of velocity fluctuations [@rheol0] or non-affine displacements [@nafsc2] of the particles under shear. Characteristic sizes of collectively moving regions, i.e. *rigid clusters*, are then extracted as functions of $\dot{\gamma}$ and $\phi$, however, there is a lack of consensus on the scaling of the sizes. For example, the size of rigid clusters $\xi$ diverges as the shear rate goes to zero $\dot{\gamma}\rightarrow 0$ so that the power-law scaling $\xi\sim\dot{\gamma}^{-s}$ was suggested, where the exponent varies from $s=0.23$ to $0.5$ depending on numerical models and flow conditions [@dh_md2; @diff_shear_md1]. The dependence of the rigid cluster size on packing fraction is also controversial. If the system is below | the asymptotic behavior in rapid flows $ \dot{\gamma}\gg\dot{\gamma}_c$) [ @diff_shear_md2; @diff_shear_md3; @diff_shear_md4 ]. In addition, it was found in MD model of piano athermal disks that in a sufficiently small flow pace range, the diffusivity changes from $ D\sim\dot{\gamma}$ ($ \phi<\phi_J$) to $ \dot{\gamma}^{0.78}$ ($ \phi\simeq\phi_J$) [ @diff_shear_md0 ], implying that the crossing over shear rate $ \dot{\gamma}_c$ vanishes as the system approach jamming from below $ \phi\rightarrow\phi_J$.
Note that the self - dispersion of soft athermal particles testify a clear difference from the diffusion in glass; * no plateau * is observed in (transverse) entail square displacements (MSDs) [ @diff_shear_md0; @diff_shear_md2; @diff_shear_md3; @diff_shear_md4; @diff_shear_md7; @dh_md2 ]. The absence of sub - dispersion can be also seen in quasi - static simulations ($ \dot{\gamma}\rightarrow 0 $) of delicate athermal disks [ @dh_qs1 ] and MD simulations of granular materials sheared under changeless pressure [ @diff_shear_md1 ].
Because the self - diffusion can be associated with collective motions of soft athermal particles, researchers have analyzed spatial correlations of velocity fluctuations [ @rheol0 ] or non - affine displacements [ @nafsc2 ] of the particles under shear. Characteristic sizes of collectively affect regions, i.e. * rigid bunch *, are then educe as function of $ \dot{\gamma}$ and $ \phi$, however, there is a lack of consensus on the scaling of the sizes. For example, the size of rigid clusters $ \xi$ diverges as the shear rate goes to zero $ \dot{\gamma}\rightarrow 0 $ so that the baron - law scaling $ \xi\sim\dot{\gamma}^{-s}$ was suggested, where the exponent varies from $ s=0.23 $ to $ 0.5 $ depending on numeral models and flow conditions [ @dh_md2; @diff_shear_md1 ]. The dependence of the inflexible cluster size on packing material fraction is besides controversial. If the arrangement is below | thf asymptotic behavior in rapid flows $\doj{\gqmma}\gg\vot{\gammz}_c$) [@diff_sfear_md2; @diff_shear_md3; @diff_shear_ld4]. In aedition, it was found iv MD simupations if sidt athermal disks bkat ih a sbfhiciently small flow rate range, the difxuririty changes from $D\sim\dot{\gamma}$ ($\phi<\phy_J$) to $\dpt{\hamma}^{0.78}$ ($\phi\simeq\khi_J$) [@cyff_sgvav_md0], implying that the crossover ahear rete $\dot{\gamma}_c$ vsnishes as the system apprlachfs jamming from bepow $\phi\righjzrrjq\phi_J$.
Note thxt the self-diffusion or soft athermal particles shows a clzar differebcw fgmm the diffnsion pn glass; *no plateau* iv obserfed in (transvevse) mxan wquare displacements (KSDs) [@diff_shear_md0; @dyff_shear_mg2; @biff_shear_md3; @diff_shear_nd4; @diff_vheas_md7; @ay_md2]. Tht ausehce of suu-diffusion dan be also seen in quasi-stativ fpkulations ($\dof{\gamma}\widhtarrow 0$) of soft athermal disks [@dh_qs1] atd JD simulations of granuoar materials sheared under cogstant pressure [@diff_shear_md1].
Because the self-diffushon cen be africlated with collective motions of soft athermaj psrnicles, researcherf have analuzfd fpatial correuations or velocity fluctuahions [@rreol0] ir non-affyne cisplacements [@nafsc2] of the particles unber shear. Characteriscic sizes of colkectifely moving regions, i.e. *riyid clhsters*, are hhen extrzzted as functionr on $\dmt{\gamma}$ and $\phi$, however, thqre is a oack of conrensos on tre scaling of tma sizes. For examplf, the shze of rigld clusters $\xi$ diverges as the shear rate goex do dero $\dot{\gcmma}\rinhtarrow 0$ so thwt the power-lac scaliny $\xi\sio\dot{\gamma}^{-s}$ was sujgested, wherq the exponend varies from $s=0.23$ to $0.5$ dqpeneing on numdfical models amd flow conditions [@eh_md2; @diff_shear_md1]. Tme dekehdence of the rnyie cluster size pn oachijg fwdction is alvo cuntfpverskal. If uke whstek is below | the asymptotic behavior in rapid flows $\dot{\gamma}\gg\dot{\gamma}_c$) @diff_shear_md4]. addition, it found in MD that a sufficiently small rate range, the changes from $D\sim\dot{\gamma}$ ($\phi<\phi_J$) to $\dot{\gamma}^{0.78}$ [@diff_shear_md0], implying that the crossover shear rate $\dot{\gamma}_c$ vanishes as the system approaches from below $\phi\rightarrow\phi_J$. Note that the self-diffusion of soft athermal particles shows a difference the in *no plateau* is observed in (transverse) mean square displacements (MSDs) [@diff_shear_md0; @diff_shear_md2; @diff_shear_md3; @diff_shear_md4; @diff_shear_md7; @dh_md2]. absence of sub-diffusion can be also seen in simulations ($\dot{\gamma}\rightarrow 0$) of athermal disks [@dh_qs1] and MD of materials sheared constant [@diff_shear_md1]. the self-diffusion can associated with collective motions of soft athermal particles, researchers have analyzed spatial correlations of velocity fluctuations [@rheol0] non-affine displacements the particles shear. sizes collectively moving regions, clusters*, are then extracted as functions $\phi$, however, there is a lack of consensus the scaling the sizes. For example, the size rigid clusters $\xi$ diverges as the shear rate to zero $\dot{\gamma}\rightarrow 0$ so that the power-law scaling $\xi\sim\dot{\gamma}^{-s}$ was suggested, where the exponent $s=0.23$ to $0.5$ depending numerical models and conditions @diff_shear_md1]. dependence the rigid size on packing fraction is also controversial. If the system is | the asymptotic behavior in raPid flows $\doT{\gammA}\gg\Dot{\GaMma}_c$) [@Diff_Shear_md2; @diff_sheAR_md3; @dIff_shear_md4]. In addition, it Was foUnD In MD SImUlatiOns of soFT aTHErmAl DiSks ThAT iN a sufFicIently sMall flow raTe rAnGe, the diffusiVItY changes frOm $D\Sim\dot{\gamma}$ ($\pHi<\pHi_J$) to $\dOt{\GamMA}^{0.78}$ ($\phi\sImeQ\phi_J$) [@Diff_shEAr_md0], imPlying thaT tHE crossOVer sheaR RAtE $\dot{\Gamma}_c$ vanishes as tHE sYStem approaches JamminG fROm BELow $\Phi\Rightarrow\PhI_J$.
NotE That the SElF-DIFfuSIon of soft atheRmal particlES shOws a clEaR diFFerencE from ThE DifFusion in glaSs; *no Plateau* is ObservED in (tranSVerse) meAn squaRe dIspLaceMEnTs (mSDS) [@dIFf_sHEaR_md0; @DIff_Shear_md2; @dIfF_sHear_mD3; @difF_SHEAr_md4; @DifF_sheAr_md7; @dH_md2]. The absence Of sUb-diFFusIon caN be alSo seEn In quaSi-statIc simUlAtions ($\dot{\gamma}\rIghtArrow 0$) of soFt aThErmAl Disks [@DH_qs1] and mD sImuLations Of granuLAr mAtERIAlS sheared under constAnT PReSsure [@difF_shear_MD1].
BEcAUse the seLf-DifFusiON Can be AssoCIaTed with cOllectIVe MoTions of SoFt atheRmAl pArtIcles, REseaRchers Have analYzed sPAtial correlatiONs of velocity fLUcTUAtIOns [@rHeoL0] or non-affinE disPLaceMentS [@NaFsc2] OF the pArticLeS UnDEr shear. CharacteristIc Sizes oF collEctively movinG regions, i.e. *RIGId clusteRs*, arE ThEN extracted as fuNctioNs of $\dot{\gamMA}$ and $\phi$, hOweveR, there is A lack of coNSEnsus on tHe sCalIng Of tHE SiZes. For example, THE sizE oF rigid cLusTers $\xi$ dIveRgeS as The ShEar rate goEs to zero $\DoT{\gAmMa}\RigHtarrOW 0$ so that tHe PowEr-Law ScaliNG $\xi\sim\Dot{\gaMma}^{-s}$ WaS sUGgeSted, wheRE tHE ExpoNeNt VariEs fRoM $s=0.23$ to $0.5$ dEpenDIng On numerIcal modelS anD Flow CoNdItions [@dH_md2; @diff_shear_mD1]. THe dependenCe Of tHe rigiD CLuster siZe on packing fraction is alSO controVerSial. IF the System is bEloW | the asymptotic behavior i n rapid fl ows $ \do t{\ ga mma} \gg\ dot{\gamma}_c$ ) [@d iff_shear_md2; @diff_s hear_ md 3 ; @d i ff _shea r_md4]. In a ddi ti on , i tw as foun d i n MD si mulationsofso ft athermald is ks that in asufficiently sm all fl ow ra t e ran ge, thediffus i vity c hanges fr om $D\sim \ dot{\ga m m a} $ ($ \phi<\phi_J$) to$ \d o t{\gamma}^{0.7 8}$ ($ \p h i\ s i meq \ph i_J$) [@di ff _shea r _md0],i mp l y i ngt hat the cross over shearr ate $\dot {\ gam m a}_c$vanis he s as the system app roaches j amming from be l ow $\ph i\righ tar row \phi _ J$ .
No te tha t t hes elf -diffusi on o f sof t at h e r m al p art icle s sho ws a clear di ffe renc e fr om th e dif fusi on in g lass;*no p la teau* is observ ed i n (transv ers e) me an squa r e disp lac eme nts (MS Ds) [@d i ff_ sh e a r _m d0; @diff_shear_md 2; @ di ff_shear _md3;@ di ff _ shear_md 4; @d iff_ s h ear_m d7;@ dh _md2]. T he abs e nc eof sub- di ffusio ncan be also seen in qu asi-stat ic si m ulations ($\do t {\gamma}\righ t ar r o w0 $) o f s oft atherma l di s ks [ @dh_ q s1 ] a n d MDsimul at i on s of granular materi al s shea red u nder constant pressure[ @ d iff_shea r_md 1 ].
Because the s elf-d iffusion c a n be ass ociat ed withcollectiv e motionsofsof t a the r m al particles, r e s earc he rs have an alyzedspa tia l c orr el ations of velocit yfl uc tu ati ons [ @ rheol0]or no n- aff ine d i splace ments [@n af sc 2 ] o f the p a rt i c lesun de r sh ear .Chara cter i sti c sizes of colle cti v elymo vi ng regi ons, i.e. *ri gi d clusters *, ar e then e xtracted as functions of $\dot{ \ gamma}$ an d $\p hi$, however, th ere is al ack of conse nsuson th e scali n g o f t he sizes. Fo r exa mple, t he s ize ofrigid clusters $\x i $ d iverges as th e s hear r at e g o es toze r o $ \ d ot{\gamma}\righ tarrow 0$so th at the pow e r-l aw scalin g $\xi\ sim\d o t{\gamm a}^{-s}$was sugge st ed,w h ere the expon ent vari es from $ s =0.23 $ t o $0. 5$depend in g o n num erical mod els a nd flo wcondit ions[@ dh_md2;@diff_shear_md1]. The d epende nce o f t he rigidclu s ter size onpack ing fracti onisalsocon t rover sial . I f t h e sys temi s below | the_asymptotic behavior_in rapid flows $\dot{\gamma}\gg\dot{\gamma}_c$)_[@diff_shear_md2; @diff_shear_md3;_@diff_shear_md4]._In addition,_it_was found in_MD simulations of_soft athermal disks that_in a sufficiently_small_flow rate range, the diffusivity changes from $D\sim\dot{\gamma}$ ($\phi<\phi_J$) to $\dot{\gamma}^{0.78}$ ($\phi\simeq\phi_J$) [@diff_shear_md0], implying_that_the crossover_shear_rate_$\dot{\gamma}_c$ vanishes as the system_approaches jamming from below $\phi\rightarrow\phi_J$.
Note_that the_self-diffusion of soft athermal particles shows a clear_difference_from the diffusion_in glass; *no plateau* is observed in (transverse) mean_square displacements (MSDs) [@diff_shear_md0; @diff_shear_md2; @diff_shear_md3;_@diff_shear_md4; @diff_shear_md7; @dh_md2]._The_absence_of sub-diffusion can be_also seen in quasi-static simulations ($\dot{\gamma}\rightarrow_0$) of soft athermal disks [@dh_qs1]_and MD simulations of granular materials sheared_under constant pressure [@diff_shear_md1].
Because the self-diffusion_can be associated with collective_motions of_soft athermal particles, researchers have_analyzed spatial correlations_of velocity_fluctuations [@rheol0] or_non-affine displacements [@nafsc2] of the particles_under shear. Characteristic_sizes of collectively moving regions, i.e. *rigid clusters*,_are_then extracted as_functions_of_$\dot{\gamma}$ and_$\phi$, however, there_is_a lack_of_consensus on the scaling of the_sizes._For example, the size of rigid clusters_$\xi$ diverges as the_shear_rate goes to zero_$\dot{\gamma}\rightarrow 0$ so that the_power-law scaling $\xi\sim\dot{\gamma}^{-s}$ was suggested, where_the exponent_varies from_$s=0.23$ to $0.5$ depending on numerical models and flow conditions [@dh_md2;_@diff_shear_md1]. The dependence of the rigid_cluster size on packing_fraction is_also_controversial. If the_system_is below |
A12 $\rm erg \ cm^{-2}$ The lower limit of photon flux in $100-350$ keV.
100\_350kev\_hi A12 $\rm erg \ cm^{-2}$ The upper limit of photon flux in $100-350$ keV.
15\_150kev A12 $\rm erg \ cm^{-2}$ Photon flux in $15-150$ keV.
15\_150kev\_low A12 $\rm erg \ cm^{-2}$ The lower limit of photon flux in $15-150$ keV.
15\_150kev\_hi A12 $\rm erg \ cm^{-2}$ The upper limit of photon flux in $15-150$ keV.
15\_350kev A12 $\rm erg \ cm^{-2}$ Photon flux in $15-350$ keV.
15\_350kev\_low A12 $\rm erg \ cm^{-2}$ The lower limit of photon flux in $15-350$ keV.
15\_350kev\_hi A12 $\rm erg \ cm^{-2}$ The upper limit of photon flux in $15-350$ keV.
Exposure\_time A12 s The time interval of the spectrum.
T100\_start A12 s The start time of the spectrum,
relative to the BAT trigger time.
T100\_stop A12 s The end time of the spectrum,
relative to the BAT trigger time.
: \[tab:spec\_T100\_cutpow\_energy\_fluence\] The format of the table that presents the energy fluence in unit of $\rm erg \ s^{-1}$ from the CPL fit for the time-averaged ($T_{100}$) spectra. Note that this table includes the fit for every GRB, regardless of whether the fit is acceptable from the criteria listed in Section \[sect:event\_standard\]. A list of GRBs with acceptable fits can be found in Table \[tab:spec\_T100\_best\_model\]. (This table is available in its entirety in a machine-readable form in the online journal. Only the column formats are shown here for guidance regarding its form | A12 $ \rm erg \ cm^{-2}$ The lower limit of photon flux in $ 100 - 350 $ keV.
100\_350kev\_hi A12 $ \rm erg \ cm^{-2}$ The upper limit of photon magnetic field in $ 100 - 350 $ keV.
15\_150kev A12 $ \rm erg \ cm^{-2}$ Photon magnetic field in $ 15 - 150 $ keV.
15\_150kev\_low A12 $ \rm erg \ cm^{-2}$ The lower limit of photon flux in $ 15 - 150 $ keV.
15\_150kev\_hi A12 $ \rm erg \ cm^{-2}$ The upper terminus ad quem of photon flux in $ 15 - 150 $ keV.
15\_350kev A12 $ \rm erg \ cm^{-2}$ Photon flux in $ 15 - 350 $ keV.
15\_350kev\_low A12 $ \rm erg \ cm^{-2}$ The lower terminus ad quem of photon flux density in $ 15 - 350 $ keV.
15\_350kev\_hi A12 $ \rm erg \ cm^{-2}$ The upper limit of photon magnetic field in $ 15 - 350 $ keV.
Exposure\_time A12 s The time interval of the spectrum.
T100\_start A12 s The start prison term of the spectrum,
relative to the BAT trigger fourth dimension.
T100\_stop A12 s The goal time of the spectrum,
relative to the BAT trigger time.
: \[tab: spec\_T100\_cutpow\_energy\_fluence\ ] The format of the table that introduce the energy fluence in unit of $ \rm erg \ s^{-1}$ from the CPL fit for the prison term - averaged ($ T_{100}$) spectra. Note that this table include the fit for every GRB, regardless of whether the fit is acceptable from the criteria listed in Section \[sect: event\_standard\ ]. A tilt of GRBs with acceptable fits can be found in Table \[tab: spec\_T100\_best\_model\ ]. (This table is available in its entirety in a machine - readable mannequin in the online journal. Only the column formats are shown here for guidance regarding its form | A12 $\rm erg \ cm^{-2}$ The lower limit oh photoh flux iv $100-350$ keV.
100\_350kev\_hi A12 $\rm erg \ cm^{-2}$ The uppvr limit if pioton flux in $100-350$ kxB.
15\_150jev A12 $\rm erg \ wm^{-2}$ Photon flux in $15-150$ keV.
15\_150kev\_low A12 $\dm erg \ cm^{-2}$ The lower ljmit of photon flux on $15-150$ keV.
15\_150kev\_ji A12 $\rm erg \ cl^{-2}$ Trw upper limig of photon flux in $15-150$ kgV.
15\_350kev A12 $\rk erg \ cm^{-2}$ Proton flux ik $15-350$ keV.
15\_350nev\_oow A12 $\rm erg \ cm^{-2}$ Thg lower likic of photon flux in $15-350$ jeC.
15\_350yev\_gi A12 $\cm erg \ cm^{-2}$ The upper limit of priton flux in $15-350$ keV.
Exposure\_time A12 s The time intetval of thq spectrum.
T100\_start A12 w The start time os tne spectrum,
relative to the BAT tritger time.
T100\_stop S12 s The eha time of the spdctgum,
gelatnve tm the BAT hrigger time.
: \[tab:spec\_T100\_cujpof\_envrgy\_fluenee\] The format of thq table that ptesents tke enefgy fluencv in unit of $\rm erg \ s^{-1}$ from the C[P fit for thx time-avewagee ($T_{100}$) wpectra. Vote that this table inblbdes the dit for every GRB, vegaramess of whether tht dit is acceptanle frjm tie cryderia listed in Recgoon \[sezt:event\_stanbafd\]. A list of GRBs with awcepfable fits can be gomnd in Tavle \[tab:s[ec\_T100\_best\_model\]. (This table is avallablx in ivs entorejy in a machine-readable form in the onlije mournal. Only ehe goluin formats are shown here for guidance regarding mts form | A12 $\rm erg \ cm^{-2}$ The lower photon in $100-350$ 100\_350kev\_hi A12 $\rm limit photon flux in keV. 15\_150kev A12 erg \ cm^{-2}$ Photon flux in keV. 15\_150kev\_low A12 $\rm erg \ cm^{-2}$ The lower limit of photon flux $15-150$ keV. 15\_150kev\_hi A12 $\rm erg \ cm^{-2}$ The upper limit of photon in keV. A12 erg \ cm^{-2}$ Photon flux in $15-350$ keV. 15\_350kev\_low A12 $\rm erg \ cm^{-2}$ The lower of photon flux in $15-350$ keV. 15\_350kev\_hi A12 erg \ cm^{-2}$ The limit of photon flux in keV. A12 s time of spectrum. T100\_start A12 The start time of the spectrum, relative to the BAT trigger time. T100\_stop A12 s The end of the to the trigger : The format of that presents the energy fluence in erg \ s^{-1}$ from the CPL fit for time-averaged ($T_{100}$) Note that this table includes the for every GRB, regardless of whether the fit acceptable from the criteria listed in Section \[sect:event\_standard\]. A list of GRBs with acceptable fits found in Table \[tab:spec\_T100\_best\_model\]. table is available its in machine-readable in the journal. Only the column formats are shown here for guidance regarding form | A12 $\rm erg \ cm^{-2}$ The lower limit of phOton flux in $100-350$ KeV.
100\_350keV\_hi a12 $\rm ErG \ cm^{-2}$ THe upPer limit of photON fluX in $100-350$ keV.
15\_150kev A12 $\rm erg \ cm^{-2}$ PhotoN flux In $15-150$ KEV.
15\_150keV\_LoW A12 $\rm eRg \ cm^{-2}$ The LOwER LimIt Of PhoToN FlUx in $15-150$ kEV.
15\_150kEv\_hi A12 $\rm Erg \ cm^{-2}$ The upPer LiMit of photon fLUx In $15-150$ keV.
15\_350kev A12 $\rM erG \ cm^{-2}$ Photon fluX in $15-350$ KeV.
15\_350kev\_LoW A12 $\rM Erg \ cm^{-2}$ the Lower Limit oF Photon Flux in $15-350$ keV.
15\_350KeV\_Hi A12 $\rm eRG \ cm^{-2}$ The uPPEr LimiT of photon flux in $15-350$ kev.
exPOsure\_time A12 s The Time inTeRVaL OF thE spEctrum.
T100\_staRt a12 s The STart timE Of THE SpeCTrum,
relative tO the BAT trigGEr tIme.
T100\_stOp a12 s THE end tiMe of tHe SPecTrum,
relativE to tHe BAT trigGer timE.
: \[Tab:spec\_t100\_Cutpow\_eNergy\_fLueNce\] the fORmAt Of tHe TAblE ThAt pREseNts the enErGy FluenCe in UNIT Of $\rm Erg \ S^{-1}$ froM the CpL fit for the tiMe-aVeraGEd ($T_{100}$) SpectRa. NotE thaT tHis taBle incLudes ThE fit for every GRB, RegaRdless of wHetHeR thE fIt is aCCeptabLe fRom The critEria lisTEd iN SECTIoN \[sect:event\_standard\]. a lIST oF GRBs witH accepTAbLe FIts can be FoUnd In TaBLE \[tab:sPec\_T100\_BEsT\_model\]. (ThIs tablE Is AvAilable In Its entIrEty In a MachiNE-reaDable fOrm in the OnlinE Journal. Only the COlumn formats aRE sHOWn HEre fOr gUidance regaRdinG Its fOrm | A12 $\rm erg \ cm^ {-2}$ T helo werlimi t of photon fl u x in $100-350$ keV.
100 \ _3 50kev \_hi A 1 2 $\ rmer g \ cm^{ -2} $ Theupp er limit of ph o to n flux in$10 0-350$ keV.
15 \_1 50kev A12 $\rm er g\ cm^{- 2 }$ Photon f l ux in $15-150$ ke V.
1 5\ _150k e v\_low A 1 2 $\rm erg \cm^{-2}$ Th e l o wer li mit o fp hot on flux in$15- 150$ keV.
15\_1 50kev\ _hi A1 2 $\ rm erg \cm^ { -2} $ The upp e r l imit of pho ton f lux in $15-15 0$keV.
1 5\_35 0kev A1 2 $\rm erg\ cm ^{-2}$ Pho ton fl ux in $ 15-350$ keV .
1 5 \_ 350kev\_ low A 12 $\rm e rg\ cm ^ { -2}$ The lower li mi t of ph ot on flu xin$15 -350$ keV.
15 \ _350kev\_hi A12 $\rm er g \c m^{- 2}$ Th e upp er l i mi t o f phot on fl ux in $15-350$ keV.
Ex posure\_time A12 s The time interval of t he spect rum.
T1 0 0 \_ start A 1 2 s T he start t im eof th e spe c trum,
relativetot he B AT t riggertime. T100\ _ s top A12 s T heend ti meo f thespectr um,
re lati v e t o t h eB ATtr i gge r time.
: \[tab:spe c \_T 10 0\_cutp ow\_ene rgy\_ f luence\ ] The for mat of th etabl e tha t presents the ene rgy fluen c e inu ni t of$\r m erg\s^{ -1}$from t h e C PL fi t forth e time -aver ag ed ($T_{ 100}$) spectra. Note th at thi s tab leincludesthe fit for ever y GR B, regardl ess of whet her the f it i s a cce p table fro m the crit e ri a l i s te d in Sectio n \ [se ct:ev ent \ _stand ard\ ]. A list of GRBs with acceptabl e fi t s ca n b e fou nd in Table \[ta b:s pe c \ _T100\_b es t\_model\]. (This t ab l e isavaila ble in its en t i re t y in a mac hin e-readabl e f or m in the o nl i ne jou rnal .Only t he col u mn f o r mats are shown h ere f o r guid a nce rega rd ing its form | _ _A12 _ $\rm_erg_\ cm^{-2}$__ _ _ _The lower limit_of_photon flux in $100-350$ keV. __
___100\_350kev\_hi _A12 _$\rm erg_\ cm^{-2}$ __ _ The upper limit of photon flux in $100-350$_keV. _ ___
_ 15\_150kev _ A12 _ $\rm erg \ cm^{-2}$_ _ _ _ _ _Photon flux_in $15-150$ keV._ _ _ __ ___
_ 15\_150kev\_low __ _A12_ $\rm erg_\_cm^{-2}$ _ __ The lower limit_of photon flux in $15-150$_keV. _ _ _
15\_150kev\_hi _A12 $\rm_erg \ cm^{-2}$ _ __ __ _ The upper limit of photon_flux in_$15-150$ keV. _ __
_ 15\_350kev _ A12 ___ $\rm erg \ cm^{-2}$_ _ _ __ Photon flux_in_$15-350$ keV. ___ _ _
15\_350kev\_low _ _A12 _$\rm_erg_\ cm^{-2}$ _ _ The_lower limit of photon flux in $15-350$ keV. _ _
15\_350kev\_hi _A12 $\rm erg \_cm^{-2}$ _ __ The upper limit of photon flux_in_$15-350$_keV._ _ _ _ _
_ Exposure\_time_ __A12 s _ _ __ _ _ The time interval_of the spectrum. __ _ _ _ _
_ _ T100\_start_ _ A12 __ __s__ _ _ _ __ _ _ _ The start time of the spectrum, _ _
_ _ __ _ _ __ ___ __ _____ _ __ relative to the_BAT trigger_time._ _ _
__ ___T100\_stop_ _ A12 __s _ __ _ _ _ The end_time of the spectrum, _
_ _ __ _ _ _ _ _ _ _ _ relative to the BAT trigger time._ __
:_ \[tab:spec\_T100\_cutpow\_energy\_fluence\] The format of the_table that presents the_energy fluence in unit_of_$\rm erg \_s^{-1}$ from_the CPL fit for_the time-averaged ($T_{100}$) spectra._Note that this table includes the fit for_every GRB, regardless_of whether the fit is acceptable from the criteria listed in_Section \[sect:event\_standard\]. A list of GRBs with acceptable fits can be found in Table \[tab:spec\_T100\_best\_model\]. (This table is available in its entirety_in a machine-readable_form_in the online journal. Only the column formats are shown here for_guidance regarding_its form |
in[r_{-},r_{+}]$ where we denote he first inner and the outer horizons by $r_{-}$ and $r_{+}$, respectively, i.e., $$\begin{aligned}
&&\left. \left( e^{-\delta}r^{2}f\frac{d\varphi }{dr} \right) \right|_{r_{+}}-
\left.\left( e^{-\delta}r^{2}f\frac{d\varphi }{dr} \right) \right|_{r_{-}}\nonumber \\
&&\hspace{1cm}=4 \int\limits_{r_{-}}^{r_{+}}r^2 e^{-\delta} \alpha(\varphi)
{\cal A}^{4}(\varphi)\left[X\partial_{X}L(X)- L(X)\right] dr>0, \nonumber \\ \label{varhu}\end{aligned}$$ Since $f(r_{-})=0=f(r_{+})$ the left-hand side (LHS) of (\[varhu\]) is equal to zero. The integral on the RHS, however, is positive. The contradiction we reach means that our admission is incorrect. So if a black hole exists it will have a single horizon, i.e., its causal structure will be Schwarzschild-like. In both conformal frames, inside the event horizon a space-like singularity is hidden.
The qualitative behavior of $\delta(r)$ can easily be seen from eqs.(\[MagnDelta\]) and (\[ElDelta\]). It decreases monotonously with $r$ for both the magnetically and the electrically charged cases.
The scalar field plays a crucial role in changing the causal structure of the magnetically charged black hole. In GR, for $(M/P)^2\leq 24/25$ a single horizon exists, but for $(M/P)^2>24/25$ a second and a third horizons occur. Extremal solutions exist only for $b^2\leq b_{\rm crit}^2=8/27P^2$. In the presence of the scalar field, for $b^2>b^2_{\rm crit}$, the causal structure is qualitatively the same as in the General Theory. For $b^2 \leq b^2_{\rm crit}$, however, the scalar field changes the causal structure significantly and the number of horizons reduces to one.
For the electrically charged case in GR extremal solutions exist for all values of $b^2 | in[r_{-},r_{+}]$ where we denote he first inner and the outer horizon by $ r_{-}$ and $ r_{+}$, respectively, i.e., $ $ \begin{aligned }
& & \left. \left (e^{-\delta}r^{2}f\frac{d\varphi } { dr } \right) \right|_{r_{+}}-
\left.\left (e^{-\delta}r^{2}f\frac{d\varphi } { dr } \right) \right|_{r_{-}}\nonumber \\
& & \hspace{1cm}=4 \int\limits_{r_{-}}^{r_{+}}r^2 e^{-\delta } \alpha(\varphi)
{ \cal A}^{4}(\varphi)\left[X\partial_{X}L(X)- L(X)\right ] dr>0, \nonumber \\ \label{varhu}\end{aligned}$$ Since $ f(r_{-})=0 = f(r_{+})$ the leftover - hand side (LHS) of (\[varhu\ ]) is adequate to zero. The integral on the RHS, however, is positivist. The contradiction we reach intend that our entree is incorrect. thus if a black hole exist it will have a single horizon, i.e., its causal social organization will be Schwarzschild - like. In both conformal frames, inside the event horizon a space - like singularity is hidden.
The qualitative behavior of $ \delta(r)$ can well be seen from eqs.(\[MagnDelta\ ]) and (\[ElDelta\ ]). It decreases monotonously with $ r$ for both the magnetically and the electrically charged event.
The scalar field plays a all-important role in change the causal structure of the magnetically charged black hole. In GR, for $ (M / P)^2\leq 24/25 $ a individual horizon exists, but for $ (M / P)^2>24/25 $ a second and a third horizons occur. Extremal solutions exist only for $ b^2\leq b_{\rm crit}^2=8/27P^2$. In the presence of the scalar field, for $ b^2 > b^2_{\rm crit}$, the causal structure is qualitatively the same as in the General Theory. For $ b^2 \leq b^2_{\rm crit}$, however, the scalar field changes the causal structure significantly and the number of horizons reduces to one.
For the electrically charged event in GR extremal solutions exist for all values of $ b^2 | in[r_{-},g_{+}]$ where we denote he firrt inner and thg iuter iorizona by $r_{-}$ avd $r_{+}$, respectively, i.e., $$\begin{almgnee}
&&\left. \left( e^{-\delta}r^{2}f\frac{d\varohi }{dr} \rihht) \righr|_{r_{+}}-
\leht.\left( e^{-\delta}r^{2}f\fczc{d\varpmn }{dr} \dlght) \xijht|_{r_{-}}\nonumber \\
&&\hskace{1cm}=4 \int\likits_{r_{-}}^{r_{+}}r^2 e^{-\delta} \aupka(\varphi)
{\cal A}^{4}(\varphi)\left[X\partial_{X}L(X)- L(V)\right] cr>0, \nonumber \\ \lafel{vswhu}\ehd{aligned}$$ Since $f(r_{-})=0=f(r_{+})$ the left-hand aide (LHV) of (\[varhu\]) is equal to zero. The integrap on the RHS, however, id positive. Jge sintradiction we reach means that oor admission is incorrect. So if x blaek hole exiwtw ih will have e singje horizon, i.c., its cdusal syructure will ne Sciwarzschild-like. In both cmnformal frames, infide the avznt horizon a space-lije singolaridy ir hiadeh.
Tie sualitwtite behavior of $\delta(r)$ xan easily be seen grjn eqs.(\[MagnDeltz\]) and (\[QlQelta\]). It decreases monotonously with $r$ xor both the magnetically qnd the electrically fharged cwses.
The scalar field plays a crucial role in chancing vhd ccmfxo dtructure of the magnetically charged black hjme. Ik GR, for $(M/P)^2\leq 24/25$ c single horizon edixjs, but for $(M/P)^2>24/25$ x secous znd a third horizojs occut. Extrwmal soluuions exist only for $b^2\leq b_{\rm crut}^2=8/27P^2$. In the pgesebce of the scalar yield, for $b^2>b^2_{\xm crij}$, the vausal structure is quauitafively the dame as ih the General Theury. Fmr $b^2 \leq b^2_{\rm crit}$, however, ehe scalac fiekd chanees jhe caufal structkre slcnificantly and thf numyer ox horizons reduces to one.
For the electricekly charged vave pn GR extxemal xolutions exift for all valoes of $b^2 | in[r_{-},r_{+}]$ where we denote he first inner outer by $r_{-}$ $r_{+}$, respectively, i.e., \right) \left.\left( e^{-\delta}r^{2}f\frac{d\varphi }{dr} \right|_{r_{-}}\nonumber \\ &&\hspace{1cm}=4 e^{-\delta} \alpha(\varphi) {\cal A}^{4}(\varphi)\left[X\partial_{X}L(X)- L(X)\right] dr>0, \\ \label{varhu}\end{aligned}$$ Since $f(r_{-})=0=f(r_{+})$ the left-hand side (LHS) of (\[varhu\]) is equal to The integral on the RHS, however, is positive. The contradiction we reach means our is So a black hole exists it will have a single horizon, i.e., its causal structure will be In both conformal frames, inside the event horizon space-like singularity is hidden. qualitative behavior of $\delta(r)$ can be from eqs.(\[MagnDelta\]) (\[ElDelta\]). decreases with $r$ for the magnetically and the electrically charged cases. The scalar field plays a crucial role in changing the structure of charged black In for 24/25$ a single but for $(M/P)^2>24/25$ a second and occur. Extremal solutions exist only for $b^2\leq b_{\rm In the of the scalar field, for $b^2>b^2_{\rm the causal structure is qualitatively the same as the General Theory. For $b^2 \leq b^2_{\rm crit}$, however, the scalar field changes the causal and the number of reduces to one. the charged in extremal solutions for all values of $b^2 | in[r_{-},r_{+}]$ where we denote he first iNner and the Outer HorIzoNs By $r_{-}$ aNd $r_{+}$, rEspectively, i.e., $$\bEGin{aLigned}
&&\left. \left( e^{-\delta}r^{2}f\Frac{d\VaRPhi }{dR} \RiGht) \riGht|_{r_{+}}-
\lefT.\LeFT( E^{-\deLtA}r^{2}F\frAc{D\VaRphi }{dR} \riGht) \righT|_{r_{-}}\nonumber \\
&&\HspAcE{1cm}=4 \int\limits_{R_{-}}^{R_{+}}r^2 E^{-\delta} \alphA(\vaRphi)
{\cal A}^{4}(\varpHi)\lEft[X\paRtIal_{x}l(X)- L(X)\rIghT] dr>0, \noNumber \\ \LAbel{vaRhu}\end{aliGnED}$$ Since $F(R_{-})=0=f(r_{+})$ the lEFT-hAnd sIde (LHS) of (\[varhu\]) is eqUAl TO zero. The integrAl on thE Rhs, hOWEveR, is Positive. ThE cOntraDIction wE ReACH MeaNS that our admisSion is incorREct. so if a bLaCk hOLe exisTs it wIlL HavE a single horIzon, I.e., its causAl struCTure wilL Be SchwaRzschiLd-lIke. in boTH cOnForMaL FraMEs, InsIDe tHe event hOrIzOn a spAce-lIKE SInguLarIty iS hiddEn.
The qualitatIve BehaVIor Of $\delTa(r)$ caN easIlY be seEn from Eqs.(\[MaGndelta\]) and (\[ElDelta\]). it deCreases moNotOnOusLy With $r$ FOr both The MagNeticalLy and thE EleCtRICAlLy charged cases.
The sCaLAR fIeld playS a crucIAl RoLE in changInG thE cauSAL struCturE Of The magneTicallY ChArGed blacK hOle. In Gr, fOr $(M/p)^2\leQ 24/25$ a sinGLe hoRizon eXists, but For $(M/P)^2>24/25$ A Second and a thirD Horizons occur. eXtREMaL SoluTioNs exist only For $b^2\LEq b_{\rM criT}^2=8/27p^2$. IN thE PreseNce of ThE ScALar field, for $b^2>b^2_{\rm crit}$, ThE causaL struCture is qualitAtively the SAME as in the geneRAl tHeory. For $b^2 \leq b^2_{\rM crit}$, However, the SCalar fieLd chaNges the cAusal struCTUre signiFicAntLy aNd tHE NuMber of horizonS REducEs To one.
FoR thE electrIcaLly ChaRgeD cAse in GR exTremal soLuTiOnS eXisT for aLL values oF $b^2 | in[r_{-},r_{+}]$ where wedenote hefirst in ner a nd t he o uter horizonsb y $r _{-}$ and $r_{+}$, res pecti ve l y, i . e. , $$\ begin{a l ig n e d}&& \l eft .\ le ft( e ^{- \delta} r^{2}f\fra c{d \v arphi }{dr}\ ri ght) \righ t|_ {r_{+}}-
\le ft. \left( e ^{- \ delta }r^ {2}f\ frac{d \ varphi }{dr} \r ig h t) \ri g ht|_{r_ { - }} \non umber \\
&&\hspac e {1 c m}=4 \int\limi ts_{r_ {- } }^ { r _{+ }}r ^2 e^{-\de lt a} \a l pha(\va r ph i ) {\c a l A}^{4}(\var phi)\left[X \ par tial_{ X} L(X ) - L(X) \righ t] dr> 0, \nonumbe r \ \ \label{ varhu} \ end{ali g ned}$$Since$f( r_{ -})= 0 =f (r _{+ }) $ th e l eft - han d side ( LH S) of ( \[va r h u \ ]) i s e qual to z ero. The inte gra l on the RHS, howe ver, i s pos itive. Theco ntradiction wereac h means t hat o urad missi o n is i nco rre ct. Soif a bl a ckho l e ex ists it will haveas i ng le horiz on, i. e ., i t s causal s tru ctur e willbe S c hw arzschil d-like . I nboth co nf ormalfr ame s,insid e the event horizon a sp a ce-like singul a rity is hidde n .T he qual ita tive behavi or o f $\d elta ( r) $ c a n eas ily b es ee n from eqs.(\[MagnDe lt a\]) a nd (\ [ElDelta\]).It decreas e s monotono usly wi t h $r$ for both themagnetical l y and th e ele ctricall y charged c ases.
T hesca lar fi e l dplays a cruci a l rol ein chan gin g the c aus alstr uct ur e of themagnetic al ly c ha rge d bla c k hole.In GR ,for $(M/ P )^2\le q 24/ 25$asi n gle horizo n e x i sts, b ut for $( M/ P)^2> 24/2 5 $ a second and a th ird hori zo ns occur. Extremal sol ut ions exist o nly for $ b ^ 2\leq b_ {\rm crit}^2=8/27P^2$.I n the p res enceof t he scalar fi eld, f or$ b^2>b^ 2_{\rm crit }$ , t h e caus a l s tru ct ure is qua l i tat ively t he s ame asin the General The o ry. For $b^2 \le q b ^2_{ \ r mcri t }$ , ho we v er, t he scalar field changes t he ca usal struc t ure s ignific antly a nd th e number of horiz ons reduc es too n e.
For the e lectrica lly charg e d cas e i n GRext remalso lut ionsexistf orall v aluesof $b^2 | in[r_{-},r_{+}]$ where_we denote_he first inner and_the outer_horizons_by $r_{-}$_and_$r_{+}$, respectively, i.e.,_$$\begin{aligned}
&&\left. \left( e^{-\delta}r^{2}f\frac{d\varphi_}{dr} \right) \right|_{r_{+}}-
\left.\left( e^{-\delta}r^{2}f\frac{d\varphi_}{dr} \right) \right|_{r_{-}}\nonumber_\\
&&\hspace{1cm}=4_\int\limits_{r_{-}}^{r_{+}}r^2 e^{-\delta} \alpha(\varphi)
{\cal A}^{4}(\varphi)\left[X\partial_{X}L(X)- L(X)\right] dr>0, \nonumber \\ \label{varhu}\end{aligned}$$ Since $f(r_{-})=0=f(r_{+})$ the left-hand_side_(LHS) of_(\[varhu\])_is_equal to zero. The integral_on the RHS, however, is_positive. The_contradiction we reach means that our admission is_incorrect._So if a_black hole exists it will have a single horizon,_i.e., its causal structure will be_Schwarzschild-like. In both_conformal_frames,_inside the event horizon_a space-like singularity is hidden.
The qualitative_behavior of $\delta(r)$ can easily be_seen from eqs.(\[MagnDelta\]) and (\[ElDelta\]). It decreases_monotonously with $r$ for both the_magnetically and the electrically charged_cases.
The scalar_field plays a crucial role_in changing the_causal structure_of the magnetically_charged black hole. In GR, for_$(M/P)^2\leq 24/25$ a_single horizon exists, but for $(M/P)^2>24/25$_a_second and a_third_horizons_occur. Extremal_solutions exist only_for_$b^2\leq b_{\rm_crit}^2=8/27P^2$._In the presence of the scalar_field,_for $b^2>b^2_{\rm crit}$, the causal structure is_qualitatively the same as_in_the General Theory. For_$b^2 \leq b^2_{\rm crit}$, however,_the scalar field changes the causal_structure significantly_and the_number of horizons reduces to one.
For the electrically charged case in_GR extremal solutions exist for all_values of $b^2 |
267.1761 + 7.0953 13.17-14.13 RRab 0.63806 1306.78 V2210 Oph
10910335 267.3721 +12.2314 13.70-15.00 RRab 0.56922 1382.92 V829 Oph
8062879 268.3597 +31.7131 14.37-15.42 RRab 0.62467 1397.73 V523 Her
8071159 270.7539 +36.4794 13.75-14.60 RRab 0.58324 1404.72 OW Her
5378465 273.1863 +38.7990 14.30-15.50 RRab 0.62388 1474.59 QQ Her
5380594 274.0383 +40.7120 14.79-15.98 RRab 0.56746 1296.74 HW Lyr
8148286 274.6596 +28.4195 13.86-15.13 RRab 0.40380 1481.65 V576 Her
5484203 277.6858 +38.3985 13.43-14.44 RRab 0.60141 1461.80 KN Lyr
11096176 278.3474 +13.3719 13.46-14.31 RRab 0.48995 1456.68 V633 Her
8179855 280.5579 +28.3785 13.19-14.50 RRab 0.49748 1615.86 AW Lyr
8187901 282.1015 +30.5718 13.72-14.87 RRab 0.51740 1414.75 CT Lyr
5462833 282.4888 +50.5869 13.39-14.70 RRab 0.52666 1328.74 DT Dra
8249293 283.9339 +33.5673 14.00-15.07 RRab 0.61270 1485 | 267.1761 + 7.0953 13.17 - 14.13 RRab 0.63806 1306.78 V2210 Oph
10910335 267.3721 +12.2314 13.70 - 15.00 RRab 0.56922 1382.92 V829 Oph
8062879 268.3597 +31.7131 14.37 - 15.42 RRab 0.62467 1397.73 V523 Her
8071159 270.7539 +36.4794 13.75 - 14.60 RRab 0.58324 1404.72 OW Her
5378465 273.1863 +38.7990 14.30 - 15.50 RRab 0.62388 1474.59 QQ Her
5380594 274.0383 +40.7120 14.79 - 15.98 RRab 0.56746 1296.74 HW Lyr
8148286 274.6596 +28.4195 13.86 - 15.13 RRab 0.40380 1481.65 V576 Her
5484203 277.6858 +38.3985 13.43 - 14.44 RRab 0.60141 1461.80 KN Lyr
11096176 278.3474 +13.3719 13.46 - 14.31 RRab 0.48995 1456.68 V633 Her
8179855 280.5579 +28.3785 13.19 - 14.50 RRab 0.49748 1615.86 AW Lyr
8187901 282.1015 +30.5718 13.72 - 14.87 RRab 0.51740 1414.75 CT Lyr
5462833 282.4888 +50.5869 13.39 - 14.70 RRab 0.52666 1328.74 DT Dra
8249293 283.9339 +33.5673 14.00 - 15.07 RRab 0.61270 1485 | 267.1761 + 7.0953 13.17-14.13 RRab 0.63806 1306.78 V2210 Oph
10910335 267.3721 +12.2314 13.70-15.00 RRau 0.56922 1382.92 V829 Opf
8062879 268.3597 +31.7131 14.37-15.42 RRab 0.62467 1397.73 V523 Hee
8071159 270.7539 +36.4794 13.75-14.60 RRab 0.58324 1404.72 LW Her
5378465 273.1863 +38.7990 14.30-15.50 RCzb 0.62388 1474.59 QS Her
5380594 274.0383 +40.7120 14.79-15.98 RRab 0.56746 1296.74 HW Lyr
8148286 274.6596 +28.4195 13.86-15.13 RRab 0.40380 1481.65 V576 Her
5484203 277.6858 +38.3985 13.43-14.44 GRab 0.60141 1461.80 HN Ljr
11096176 278.3474 +13.3719 13.46-14.31 RRab 0.48995 1456.68 V633 Her
8179855 280.5579 +28.3785 13.19-14.50 RRab 0.49748 1615.86 AW Lyr
8187901 282.1015 +30.5718 13.72-14.87 RRwb 0.51740 1414.75 CT Lyr
5462833 282.4888 +50.5869 13.39-14.70 ERab 0.52666 1328.74 DT Dra
8249293 283.9339 +33.5673 14.00-15.07 RRab 0.61270 1485 | 267.1761 + 7.0953 13.17-14.13 RRab 0.63806 1306.78 10910335 +12.2314 13.70-15.00 0.56922 1382.92 V829 RRab 1397.73 V523 Her 270.7539 +36.4794 13.75-14.60 0.58324 1404.72 OW Her 5378465 273.1863 14.30-15.50 RRab 0.62388 1474.59 QQ Her 5380594 274.0383 +40.7120 14.79-15.98 RRab 0.56746 1296.74 Lyr 8148286 274.6596 +28.4195 13.86-15.13 RRab 0.40380 1481.65 V576 Her 5484203 277.6858 +38.3985 RRab 1461.80 Lyr 278.3474 +13.3719 13.46-14.31 RRab 0.48995 1456.68 V633 Her 8179855 280.5579 +28.3785 13.19-14.50 RRab 0.49748 1615.86 AW 8187901 282.1015 +30.5718 13.72-14.87 RRab 0.51740 1414.75 CT 5462833 282.4888 +50.5869 13.39-14.70 0.52666 1328.74 DT Dra 8249293 +33.5673 RRab 0.61270 | 267.1761 + 7.0953 13.17-14.13 RRab 0.63806 1306.78 V2210 Oph
10910335 267.3721 +12.2314 13.70-15.00 RRab 0.56922 1382.92 V829 Oph
8062879 268.3597 +31.7131 14.37-15.42 RRab 0.62467 1397.73 V523 Her
8071159 270.7539 +36.4794 13.75-14.60 Rrab 0.58324 1404.72 OW Her
5378465 273.1863 +38.7990 14.30-15.50 RRAb 0.62388 1474.59 QQ HEr
5380594 274.0383 +40.7120 14.79-15.98 Rrab 0.56746 1296.74 hW lyr
8148286 274.6596 +28.4195 13.86-15.13 Rrab 0.40380 1481.65 V576 her
5484203 277.6858 +38.3985 13.43-14.44 RRab 0.60141 1461.80 KN Lyr
11096176 278.3474 +13.3719 13.46-14.31 RRAB 0.48995 1456.68 V633 HeR
8179855 280.5579 +28.3785 13.19-14.50 RRab 0.49748 1615.86 AW Lyr
8187901 282.1015 +30.5718 13.72-14.87 RRab 0.51740 1414.75 CT Lyr
5462833 282.4888 +50.5869 13.39-14.70 RRaB 0.52666 1328.74 DT DrA
8249293 283.9339 +33.5673 14.00-15.07 RrAb 0.61270 1485 | 267.1761 + 7.0953 13.17-14.1 3 RRa b 0. 6380 6 1 306.78 V2210 Oph 10910335 267.372 1 +1 2 .231 4 13.7 0-15.00 RR a b 0. 569 22 1 382.9 2 V829 O ph
80 628 79 268.3597 +31.7131 14 .37-15.42 RR ab 0. 62 467 1397 .73 V5 23 Her 8 071159 27 0 .7539 +36.4 7 9 4 13 .75-14.60 RRab 0 . 58324 1404.72 OWHe r
5 378 465 273. 18 63 +38.799 0 1 4 .30 - 15.50 RRab 0.62388 1 474 .59 QQ He r
53805 94 2 74.0383 +40. 7120 14 .79-15 . 98 RRab 0.5 6746 129 6.7 4 H WLy r
81 4 82 86 27 4.6596 +2 8.419 5 1 3 . 8 6-15 .13 RRa b 0.40380 148 1.6 5 V 576 Her 5484 20 3 2 77.685 8 +3 8.3985 13.43- 14.4 4 RRab 0 .6 014 1 1461 . 80 K N L yr 110 96176 278 .3 4 7 4 +13.3719 13.46 -1 4 . 31 RRab 0.48 9 95 1 456.68 V 633 Her 8 1798 5 5 280.55 79 + 28 .3 785 1 3. 19-14. 50 RR ab 0 . 4974 8 161 5.86 A W Lyr 8187901 282.1015 + 3 0. 5 7 18 13 .72 -14.87 RRab 0.51 74014 14. 7 5 C T Lyr
5462833 282.4888 +50.5 869 13.39-14.70RRab 0 . 5 2 666 132 8.74 D T Dra
824 9293 283.9339 +33.5 673 14.00-1 5.07 RRab 0.612 70 14 85 | _ 267.1761_ +_7.0953 __13.17-14.13 RRab__ _0.63806 1306.78_ V2210 Oph
_ _10910335_ 267.3721 +12.2314 13.70-15.00 RRab __0.56922 _1382.92__ V829 Oph
_ 8062879 _268.3597 _ +31.7131 14.37-15.42 RRab __ 0.62467_ 1397.73 V523 Her
_ 8071159 270.7539 _ +36.4794___13.75-14.60 RRab _ 0.58324 1404.72 _ OW Her
_ 5378465 273.1863 _ +38.7990 14.30-15.50 RRab_ 0.62388_ 1474.59_ QQ Her
_ _5380594 _ 274.0383 _ +40.7120 14.79-15.98_RRab _ 0.56746 1296.74 __HW Lyr
___ 8148286_ 274.6596__ _+28.4195_ 13.86-15.13 RRab __ 0.40380 1481.65 V576_Her
__5484203 277.6858_ +38.3985 _ 13.43-14.44 RRab _ 0.60141_ 1461.80_ KN Lyr
11096176 278.3474_ +13.3719 _13.46-14.31 RRab _ _0.48995_ 1456.68 __V633 Her
_ 8179855 _280.5579 _ +28.3785 13.19-14.50_RRab __0.49748 1615.86 AW_Lyr
8187901_ 282.1015 __+30.5718_ 13.72-14.87 RRab _ 0.51740 1414.75_ CT_Lyr
5462833 __282.4888 +50.5869 __13.39-14.70 RRab _0.52666__1328.74 DT Dra
_ 8249293 _ 283.9339 +33.5673 14.00-15.07_RRab _ 0.61270 1485 |
evolution of the star and [brown dwarf ]{}population within the cluster from their birth to the age of the Pleiades depending on their initial kinematics.
Note that we only present the results for one set of initial conditions but we also performed other simulations using different numbers of objects $N$, which means in particular using various seed numbers to initialize the cluster model. The results are similar to those described in the following sections.
Results and discussion {#results}
======================
Characteristic radii {#radii}
--------------------
Over the timescale of the simulations the half mass radius does not change much through the simulations. It goes up to $\sim3.2-3.4$ pc after 12 crossing times which is consistent with observational results obtained for the Pleiades ($r_{h}=3.6$ pc; Pinfield et al. 1998). This result does not depend on the initial substellar velocity dispersion which is indeed expected considering the fact that [brown dwarf ]{}represent only 3% of the cluster mass.
A nominal core radius $r_{c}$ is calculated in [Nbody2]{} using the definition of the density radius given by Casertano & Hut (1985) slightly modified in order to obtain a convergent result using a smaller sample ($n\simeq N/2$). It is determined by the rms expression (Aarseth 2001) $$r_{c}= \left( \frac{\sum_{i=1}^{n} |\mathbf{r}_{i} -
\mathbf{r}_{d}|^{2} \, \rho_{i}^{2}}{\sum \rho_{i}^{2}} \right)^{1/2},$$ where $\mathbf{r}_{i}$ is the three-dimensional position vector of the $i$th star and $\mathbf{r}_{d}$ denotes the coordinates of the density centre. The density estimator $\rho_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is defined with respect to the sixth nearest particle $r_{6}$ and takes into account the total mass ($M_{5}$) of the five nearest neighbours. \[See Casertano & Hut (1985) for definitions of density centre and density estimator.\] In our simulations, $r_{c}=1.3$ pc at $t=0$ and $r_{c}\sim0.8$ pc at $t=12\,t_{cr}\simeq 120$ Myr. However, to be compared to the observational core | evolution of the star and [ brown dwarf ] { } population within the cluster from their parturition to the old age of the Pleiades depending on their initial kinematics.
Note that we only stage the results for one set of initial circumstance but we also performed early model using unlike numbers of objects $ N$, which means in particular use various semen numbers to initialize the bunch model. The results are similar to those report in the following sections.
consequence and discussion { # results }
= = = = = = = = = = = = = = = = = = = = = =
Characteristic radius { # radii }
--------------------
Over the timescale of the simulations the half mass radius does not change a lot through the simulations. It goes up to $ \sim3.2 - 3.4 $ pc after 12 crossing times which is consistent with experimental consequence obtained for the Pleiades ($ r_{h}=3.6 $ pc; Pinfield et al. 1998). This result does not depend on the initial substellar velocity dispersion which is indeed expected considering the fact that [ brown dwarf ] { } typify only 3% of the cluster mass.
A nominative effect spoke $ r_{c}$ is calculated in [ Nbody2 ] { } using the definition of the density radius given by Casertano & Hut (1985) slightly modified in order to receive a convergent result using a smaller sample ($ n\simeq N/2 $). It is determined by the rms expression (Aarseth 2001) $ $ r_{c}= \left (\frac{\sum_{i=1}^{n } |\mathbf{r}_{i } -
\mathbf{r}_{d}|^{2 } \, \rho_{i}^{2}}{\sum \rho_{i}^{2 } } \right)^{1/2},$$ where $ \mathbf{r}_{i}$ is the three - dimensional place vector of the $ i$th star and $ \mathbf{r}_{d}$ denotes the coordinates of the density centre. The concentration estimator $ \rho_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is define with esteem to the sixth nearest particle $ r_{6}$ and takes into account the total mass ($ M_{5}$) of the five nearest neighbor. \[See Casertano & Hut (1985) for definitions of density centre and density estimator.\ ] In our pretense, $ r_{c}=1.3 $ pc at $ t=0 $ and $ r_{c}\sim0.8 $ personal computer at $ t=12\,t_{cr}\simeq 120 $ Myr. However, to be compared to the observational core | evllution of the star and [nrown dwarf ]{}population xithin fhe clusger from their birth to the ege if tht Pleiades dependine on theig initial kintmatics.
Note that xs only ixesenf the cesults for one set of inhtial conditiots bbt we also performed other simulatiogs usinb fifferent numbgrs og objsbtw $N$, which means in particular using narious seed numbrrs to initialize the clusher lodel. The results wre similar to eyose describdd in the following sedtions.
Results and discussion {#rerults}
======================
Eharacterisjnx rwgii {#radii}
--------------------
Ovxr the timescale on the shmulatipns the half msss raeius does not change kuch through the symulationv. Nt goes up to $\sim3.2-3.4$ pc qfrer 12 wrosving rimds shmch is cojsiatent with observatiobal results obtainec sir the Pleiadss ($r_{h}=3.6$ [c; Pinfield et al. 1998). This result does not dtpend on the initial substeloar velocity dispersiln which ys indeed expected considering the fact that [browt dwacf ]{}rekrcfdbt only 3% of the cluster mass.
A nominal core radita $t_{c}$ is calculated in [Nbody2]{} uxijg jhe definition of thz dsnsity radius givej by Cafertabo & Hut (1985) fligntly modified in order to ovtain a convvrgebt result using a dmaller sam'le ($n\someq M/2$). It is determined by tke rms expression (Aarseth 2001) $$f_{c}= \left( \frac{\sum_{i=1}^{v} |\mstvbf{r}_{i} -
\maufbf{r}_{d}|^{2} \, \rho_{i}^{2}}{\sum \rhj_{i}^{2}} \right)^{1/2},$$ xhere $\mathbf{f}_{i}$ ix the ehree-dimendional position vector ov the $i$dh star anf $\mathbf{r}_{d}$ denotes the coordinavxs of the denxidy bentre. Thz denslty estimator $\rro_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is deyined wich resoect to thv sixth nxarest partisle $r_{6}$ and takad into accouit the toeal nass ($M_{5}$) of tfd five nearest neighbougs. \[See Casertano & Hut (1985) fpr asfinitions of dzusuty centre and cenritr vstmmatow.\] In our simunatiuns, $t_{c}=1.3$ pc at $t=0$ and $r_{g}\sio0.8$ pc at $t=12\,t_{cr}\simeq 120$ Myr. Hmwevsr, to be compared yo the obsgrvationaj core | evolution of the star and [brown dwarf the from their to the age their kinematics. Note that only present the for one set of initial conditions we also performed other simulations using different numbers of objects $N$, which means particular using various seed numbers to initialize the cluster model. The results are to described the sections. Results and discussion {#results} ====================== Characteristic radii {#radii} -------------------- Over the timescale of the simulations half mass radius does not change much through simulations. It goes up $\sim3.2-3.4$ pc after 12 crossing which consistent with results for Pleiades ($r_{h}=3.6$ pc; et al. 1998). This result does not depend on the initial substellar velocity dispersion which is indeed considering the [brown dwarf only of cluster mass. A radius $r_{c}$ is calculated in [Nbody2]{} of the density radius given by Casertano & (1985) slightly in order to obtain a convergent using a smaller sample ($n\simeq N/2$). It is by the rms expression (Aarseth 2001) $$r_{c}= \left( \frac{\sum_{i=1}^{n} |\mathbf{r}_{i} - \mathbf{r}_{d}|^{2} \, \rho_{i}^{2}}{\sum \rho_{i}^{2}} $\mathbf{r}_{i}$ is the three-dimensional vector of the star $\mathbf{r}_{d}$ the of the centre. The density estimator $\rho_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is defined with respect to sixth nearest particle $r_{6}$ and takes into account the total of five nearest neighbours. Casertano & Hut (1985) definitions density centre and density our $r_{c}=1.3$ and pc $t=12\,t_{cr}\simeq 120$ Myr. However, be compared to the observational | evolution of the star and [browN dwarf ]{}popuLatioN wiThiN tHe clUsteR from their birtH To thE age of the Pleiades depenDing oN tHEir iNItIal kiNematicS.
noTE ThaT wE oNly PrESeNt the ResUlts for One set of inItiAl Conditions buT We Also perforMed Other simulatIonS using DiFfeREnt nuMbeRs of oBjects $n$, Which mEans in parTiCUlar usINg varioUS SeEd nuMbers to initialize THe CLuster model. The ResultS aRE sIMIlaR to Those descrIbEd in tHE followINg SECTioNS.
Results and diScussion {#resULts}
======================
characTeRisTIc radiI {#radiI}
--------------------
OVEr tHe timescale Of thE simulatiOns the HAlf mass RAdius doEs not cHanGe mUch tHRoUgH thE sIMulATiOns. iT goEs up to $\siM3.2-3.4$ pC aFter 12 cRossING TImes WhiCh is ConsiStent with obseRvaTionAL reSults ObtaiNed fOr The PlEiades ($R_{h}=3.6$ pc; PInField et al. 1998). This reSult Does not dePenD oN thE iNitiaL SubsteLlaR veLocity dIspersiON whIcH IS InDeed expected considErING tHe fact thAt [browN DwArF ]{}RepresenT oNly 3% Of thE CLusteR masS.
a nOminal coRe radiUS $r_{C}$ iS calculAtEd in [NbOdY2]{} usIng The deFInitIon of tHe densitY radiUS given by CasertANo & Hut (1985) slightly MOdIFIeD In orDer To obtain a coNverGEnt rEsulT UsIng A SmallEr samPlE ($N\sIMeq N/2$). It is determined bY tHe rms eXpresSion (Aarseth 2001) $$r_{c}= \Left( \frac{\suM_{I=1}^{N} |\Mathbf{r}_{i} -
\MathBF{r}_{D}|^{2} \, \Rho_{i}^{2}}{\sum \rho_{i}^{2}} \rigHt)^{1/2},$$ wheRe $\mathbf{r}_{i}$ IS the threE-dimeNsional pOsition veCTOr of the $i$Th sTar And $\MatHBF{r}_{D}$ denotes the coORDinaTeS of the dEnsIty centRe. THe dEnsIty EsTimator $\rhO_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is DeFiNeD wIth RespeCT to the siXtH neArEst PartiCLe $r_{6}$ and Takes Into AcCoUNt tHe total MAsS ($m_{5}$) Of thE fIvE neaResT nEighbOurs. \[sEe CAsertanO & Hut (1985) for deFinITionS oF dEnsity cEntre and densiTy Estimator.\] IN oUr sImulatIONs, $r_{c}=1.3$ pc at $T=0$ and $r_{c}\sim0.8$ pc at $t=12\,t_{cr}\simeq 120$ MYR. HoweveR, to Be comPareD to the obsErvAtionaL coRE | evolution of the star and [brown dw arf ] {}p opu la tion wit hin the cluste r fro m their birth to the a ge of t h e Pl e ia des d ependin g o n the ir i nit ia l k inema tic s.
Not e that weonl ypresent ther es ults for o neset of initi alcondit io nsb ut we al so pe rforme d other simulati on s using differe n t n umbe rs of objects $N$ , w h ich means in p articu la r u s i ngvar ious seednu mbers to init i al i z e th e cluster mode l. The resu l tsare si mi lar to tho se de sc r ibe d in the fo llow ing secti ons.
R esultsa nd disc ussion {# res ults }
= == === == = === = == === = ===
Charac te ri sticradi i { # radi i}---- ----- -----------
Ove r th e ti mesca le of the s imula tionsthe h al f mass radius d oesnot chang e m uc h t hr ought he sim ula tio ns. Itgoes up to$\ s i m 3. 2-3.4$ pc after 12 c r o ss ing time s whic h i sc onsisten twit h ob s e rvati onal re sults ob tained fo rthe Ple ia des ($ r_ {h} =3. 6$ pc ; Pin fieldet al. 1 998). This result do e s not dependo nt h ei niti alsubstellarvelo c itydisp e rs ion which is i nd e ed expected considerin gthe fa ct th at [brown dwa rf ]{}repr e s e nt only3% o f t h e cluster mass .
Anominal co r e radius $r_{ c}$ is c alculated i n [Nbody 2]{ } u sin g t h e d efinition oft h e de ns ity rad ius givenbyCas ert ano & Hut (198 5) sligh tl ymo di fie d ino rder toob tai na c onver g ent re sultusin gas mal ler sam p le ( $n\s im eq N/2 $). I t isdete r min ed by t he rms ex pre s sion ( Aa rseth 2 001) $$r_{c}= \ left( \fra c{ \su m_{i=1 } ^ {n} |\ma thbf{r}_{i} -
\mathbf{r } _{d}|^{ 2}\, \r ho_{ i}^{2}}{\ sum \rho_ {i} ^ {2}} \ right) ^{1/2 }, $$w h ere $ \ m at hbf {r }_{i}$ ist h e t hree- di mens ional p osition vector oft he$i$th star an d $ \mat h b f{ r}_ { d} $ de no t est h e coordinates o f the dens it y c entre. The den si ty esti mator $ \rho_ { i}=3\,M _{5}/(4\p i r_{6}^{ 3} )$ i s def ined withrespectto the si x th ne a re st pa rti cle $r _{ 6}$ andtakesi nto acco unt th etotalmass($ M_{5}$)of the five nearest nei ghbour s. \[ See Casertan o & Hut (1985) f or d efinitions of de nsity ce n tre a nd d e ns ity estim ator . \] In our si mul a t io ns, $r_{c}= 1 . 3 $ p c at$t= 0 $ and$r_{ c}\sim0.8$ pc at$ t=12\,t_{cr}\s imeq 1 20$ My r . Ho we ver, to be com par ed t o the ob se rvational c ore | evolution_of the_star and [brown dwarf_]{}population within_the_cluster from_their_birth to the_age of the_Pleiades depending on their_initial kinematics.
Note that_we_only present the results for one set of initial conditions but we also performed_other_simulations using_different_numbers_of objects $N$, which means_in particular using various seed_numbers to_initialize the cluster model. The results are similar_to_those described in_the following sections.
Results and discussion {#results}
======================
Characteristic radii {#radii}
--------------------
Over the_timescale of the simulations the half_mass radius does_not_change_much through the simulations._It goes up to $\sim3.2-3.4$ pc_after 12 crossing times which is_consistent with observational results obtained for the_Pleiades ($r_{h}=3.6$ pc; Pinfield et al._1998). This result does not_depend on_the initial substellar velocity dispersion_which is indeed_expected considering_the fact that_[brown dwarf ]{}represent only 3% of_the cluster mass.
A_nominal core radius $r_{c}$ is calculated_in_[Nbody2]{} using the_definition_of_the density_radius given by_Casertano_& Hut_(1985)_slightly modified in order to obtain_a_convergent result using a smaller sample ($n\simeq_N/2$). It is determined_by_the rms expression (Aarseth_2001) $$r_{c}= \left( \frac{\sum_{i=1}^{n} |\mathbf{r}_{i}_-
\mathbf{r}_{d}|^{2} \, \rho_{i}^{2}}{\sum \rho_{i}^{2}} \right)^{1/2},$$ where_$\mathbf{r}_{i}$ is_the three-dimensional_position vector of the $i$th star and $\mathbf{r}_{d}$ denotes the coordinates_of the density centre. The density_estimator $\rho_{i}=3\,M_{5}/(4\pi r_{6}^{3})$ is_defined with_respect_to the sixth_nearest_particle $r_{6}$_and takes into account the total mass_($M_{5}$) of_the five nearest neighbours. \[See Casertano_& Hut (1985) for_definitions_of density centre and density estimator.\]_In our simulations, $r_{c}=1.3$ pc at_$t=0$ and $r_{c}\sim0.8$ pc at_$t=12\,t_{cr}\simeq_120$_Myr. However, to be compared_to the observational core |
observed
7 708 Gettysburg 0.4m $R$ 50 bad observing conditions
8 713 Jena 0.6m $R_\textrm{B}$ 50 only first half of transit observed
9 807 Antalya 1.0m $R$ 3 technical problems
10 821 Tenerife 1.2m $r_\textrm{S}$ 10 bad weather during egress phase
11 833 Jena 0.25m $R_\textrm{B}$ 100 bad weather, gaps in the data
12 834 Jena 0.25m $R_\textrm{B}$ 100 only first half of transit observed
13 834 OSN 1.5m $R_\textrm{C}$ 20 upcoming bad weather during ingress
14 840 Tenerife 1.2m $r_\textrm{S}$ 10 large fit errors
15 848 Tenerife 1.2m $r_\textrm{S}$ 15 only egress phase observed
16 854 Tenerife 1.2m $r_\textrm{S}$ 20 only ingress phase observed
17 855 Tenerife 1.2m $r_\textrm{S}$ 20 jumps in data, no good fits possible
18 861 Tenerife 1.2m $r_\textrm{S}$ 25 bad observing conditions
19 867 Tenerife 1.2m $r_\textrm{S}$ 25 bad observing conditions
20 906 Torun 0.6m $clear$ 10 only ingress phase observed
21 1001 Torun 0.6m $clear$ 6 jumps in data, no good fits possible
Analysis {#sec:Analysis}
========
All light curves (except for the Ankara 0.4m observation) are extracted from the reduced images using the same aperture photometry routines (described in section \[sec:GettingTheLightcurve\]) in order to prevent systematic offsets between different transit observations due to different light curve extraction methods. Afterwards we model the data sets using the *JKTEBOP* algorithm [@JKTEBOP], as well as the Transit Analysing Package *TAP* [@T | observed
7 708 Gettysburg 0.4 m $ R$ 50 bad observing condition
8 713 Jena 0.6 thousand $ R_\textrm{B}$ 50 only first half of transportation system observed
9 807 Antalya 1.0 m $ R$ 3 technical problem
10 821 Tenerife 1.2 m $ r_\textrm{S}$ 10 bad weather during emergence phase
11 833 Jena 0.25 thousand $ R_\textrm{B}$ 100 bad weather, col in the data
12 834 Jena 0.25 m $ R_\textrm{B}$ 100 entirely inaugural half of transit observed
13 834 OSN 1.5 m $ R_\textrm{C}$ 20 upcoming regretful weather during ingress
14 840 Tenerife 1.2 m $ r_\textrm{S}$ 10 large fit erroneousness
15 848 Tenerife 1.2 m $ r_\textrm{S}$ 15 only emergence phase observed
16 854 Tenerife 1.2 meter $ r_\textrm{S}$ 20 only ingress phase observed
17 855 Tenerife 1.2 m $ r_\textrm{S}$ 20 jumps in data, no effective fits possible
18 861 Tenerife 1.2 m $ r_\textrm{S}$ 25 bad observing conditions
19 867 Tenerife 1.2 m $ r_\textrm{S}$ 25 bad observing conditions
20 906 Torun 0.6 m $ clear$ 10 only ingress phase observed
21 1001 Torun 0.6 m $ clear$ 6 jumps in data, no good match possible
Analysis { # sec: Analysis }
= = = = = = = =
All faint curve (except for the Ankara 0.4 m observation) are extracted from the reduced images using the same aperture photometry routine (described in section \[sec: GettingTheLightcurve\ ]) in order to prevent systematic offsets between different transportation system observations due to different light curve origin methods. subsequently we model the data sets using the * JKTEBOP * algorithm [ @JKTEBOP ], as well as the Transit Analysing Package * TAP * [ @T | obderved
7 708 Getuysburg 0.4m $R$ 50 bad observing conditions
8 713 Jena 0.6m $R_\textrm{B}$ 50 only hjrst half of fvansic ibserved
9 807 Attalya 1.0m $S$ 3 technicaj problrmd
10 821 Tentrise 1.2m $r_\textrm{S}$ 10 bad weatheg during egress pnase
11 833 Jena 0.25m $R_\textrm{B}$ 100 bae wewrher, gaps in the data
12 834 Jena 0.25m $R_\textrm{B}$ 100 only furwt jdlf of tranwit ofserved
13 834 OSN 1.5m $R_\textvm{C}$ 20 upcoming bad weather durind ingress
14 840 Tenerufw 1.2m $r_\dextfn{S}$ 10 large fit errors
15 848 Tenerife 1.2k $r_\textrm{S}$ 15 only egress phase observed
16 854 Tenerife 1.2m $r_\twxtrm{S}$ 20 only iggress phase observed
17 855 Tenerife 1.2m $r_\texvro{S}$ 20 jumps in data, no good fits possiblq
18 861 Tenevife 1.2m $r_\textrm{X}$ 25 bad oyaedving conditions
19 867 Renerife 1.2i $r_\textrm{S}$ 25 bad observiug xonditions
20 906 Torun 0.6m $clear$ 10 only inhress phaad observed
21 1001 Torun 0.6m $clear$ 6 jukps in data, no glod flds possible
Analysid {#sec:Cnalyvis}
========
All ligjt curves (except for the Ankara 0.4m observation) ase vxtracted from the reduced ymages using tke same cpertufe photomenry routiies (describeq in section \[saf:GettingTheLmghtcurve\]) in irdee to prdxent systemativ offsets between dufferent transit onservxfions due to diyyetwnt light curvr ebtrwcnioi metrmds. Afterwargs wd mucel tfe data setw usimg the *JKTEBOP* algorhthm [@JKTEBOP], as well ax bhe Transut Analyfing Package *YAP* [@T | observed 7 708 Gettysburg 0.4m $R$ 50 conditions 713 Jena $R_\textrm{B}$ 50 only 9 Antalya 1.0m $R$ technical problems 10 Tenerife 1.2m $r_\textrm{S}$ 10 bad weather egress phase 11 833 Jena 0.25m $R_\textrm{B}$ 100 bad weather, gaps in the 12 834 Jena 0.25m $R_\textrm{B}$ 100 only first half of transit observed 13 OSN $R_\textrm{C}$ upcoming weather during ingress 14 840 Tenerife 1.2m $r_\textrm{S}$ 10 large fit errors 15 848 Tenerife 1.2m 15 only egress phase observed 16 854 Tenerife $r_\textrm{S}$ 20 only ingress observed 17 855 Tenerife 1.2m 20 in data, good possible 861 Tenerife 1.2m 25 bad observing conditions 19 867 Tenerife 1.2m $r_\textrm{S}$ 25 bad observing conditions 20 906 Torun 0.6m 10 only observed 21 Torun $clear$ jumps in data, fits possible Analysis {#sec:Analysis} ======== All for the Ankara 0.4m observation) are extracted from reduced images the same aperture photometry routines (described section \[sec:GettingTheLightcurve\]) in order to prevent systematic offsets different transit observations due to different light curve extraction methods. Afterwards we model the data the *JKTEBOP* algorithm [@JKTEBOP], well as the Analysing *TAP* | observed
7 708 Gettysburg 0.4m $R$ 50 bad obServing conDitioNs
8 713 JEna 0.6M $R_\TextRm{B}$ 50 oNly first half of TRansIt observed
9 807 Antalya 1.0m $R$ 3 tecHnicaL pRObleMS
10 821 TEneriFe 1.2m $r_\texTRm{s}$ 10 BAd wEaThEr dUrINg EgresS phAse
11 833 Jena 0.25M $R_\textrm{B}$ 100 bAd wEaTher, gaps in thE DaTa
12 834 Jena 0.25m $R_\teXtrM{B}$ 100 only first hAlf Of tranSiT obSErved
13 834 oSN 1.5M $R_\texTrm{C}$ 20 upCOming bAd weather DuRIng ingREss
14 840 TeneRIFe 1.2M $r_\teXtrm{S}$ 10 large fit erroRS
15 848 TENerife 1.2m $r_\textrm{s}$ 15 only eGrESs PHAse ObsErved
16 854 TenerIfE 1.2m $r_\teXTrm{S}$ 20 onlY InGRESs pHAse observed
17 855 TeNerife 1.2m $r_\texTRm{S}$ 20 Jumps iN dAta, NO good fIts poSsIBle
18 861 tenerife 1.2m $r_\tExtrM{S}$ 25 bad obseRving cONditionS
19 867 tenerifE 1.2m $r_\texTrm{s}$ 25 baD obsERvInG coNdITioNS
20 906 TOruN 0.6M $clEar$ 10 only iNgReSs phaSe obSERVEd
21 1001 ToRun 0.6M $cleAr$ 6 jumPs in data, no gooD fiTs poSSibLe
AnaLysis {#Sec:ANaLysis}
========
all ligHt curVeS (except for the AnKara 0.4M observatIon) ArE exTrActed FRom the RedUceD images Using thE SamE aPERTuRe photometry routinEs (DEScRibed in sEction \[SEc:geTTingTheLIgHtcUrve\]) IN Order To prEVeNt systemAtic ofFSeTs Between DiFferenT tRanSit ObserVAtioNs due tO differeNt ligHT curve extractiON methods. AfterWArDS We MOdel The Data sets usiNg thE *jKTEbOP* aLGoRitHM [@JKTEbOP], as WeLL aS The Transit Analysing paCkage *TaP* [@T | observed
7 708 Gettysb urg 0 .4m $R $ 50 bad observing con ditio ns 8 713 J en a 0.6 m $R _\tex trm {B}$ 5 0 only first ha lf of tran sit observed
9 807 Ant aly a 1.0 m $R$ 3 te chni cal problems
10 8 21 Ten erife1. 2 m $ r_\ textrm{S}$ 1 0 b adw eather during egress pha s e
11 83 3 Jena0. 2 5m $R_\ text rm{B}$ 1 0 0 bad w eather , g aps int he d ata
12 8 34 Jena 0 .2 5m $ R _ \tex trm {B}$ 10 0 o nlyf irs t hal f oftran si t obs erved 13 834 OSN 1.5 m $ R_ \te xt rm{C} $ 20 upcomi n g b ad w e at her during ingress
1 4 840 T en er i fe 1.2m $ r_\t e x trm{S }$ 10 la rg e fit e rr ors
15 848 T enerif e 1.2m $r _ \textrm{S}$ 1 5o n ly egre ssphase obser ved 16 85 4 Te nerif e1 .2 m $r_\textrm{S}$ 20 only i ngress pha s e observed
1 7 855 Te nerif e 1.2m $r_\text rm{S} $ 2 0 jumpsindat a,nog o od fits possibl e 18 861 Tene rif e 1 .2m $r_\textr m{S}$ 25 bad obse rv ing c ond ition s
19 867 Ten erife 1 . 2m $ r_ \t extr m{S }$ 25bad obser vin g con di ti ons
2 0 906 Torun 0.6 m $c l e ar$ 10 o n ly ingr ess phas e ob served
21 1001 To run 0. 6m $ c lear$ 6 ju mp s in data,no good fits possi b le
Analysis {#s ec: Anal y s is }
= = == = ===
A lll i ght curves (exc ept for th eA nk ara 0.4m o b ser va tion) a re extr acted from th e reduced images u si ng t h e sa me apertur e photom etry rout i nes ( d es cribe d i n sect io n \ [sec: Gettin g The Light curve\ ]) in or der t opreventsystematic offsets betw een di ffere nttransit o bse r vat ions dueto d ifferent l igh t c urveext r actio n me t ho ds. After ward s we model th e d a t asets usingt h e *J KTEBO P*a lgorit hm [ @JKTEBOP], as wel l as the Transi t An a l ysi ngP acka ge *TAP* [@T | observed
_ 7_ 708_ __ __ Gettysburg_0.4m _$R$ _ __ __ ___ 50 bad observing conditions
_ 8 _713 _ Jena_0.6m_ _ $R_\textrm{B}$ _ _ ___ 50 only_first half of transit observed
_9 807 _ _Antalya 1.0m _ $R$ _ _ _ _ _ _ _ _ 3 technical problems
10__ 821 ___ _ _Tenerife_1.2m __ $r_\textrm{S}$ __ _ __10 bad weather during_egress phase
11 _ 833 _ _ _Jena 0.25m $R_\textrm{B}$ _ _ _ __ 100 bad_weather,_gaps in_the data
12 834_ _ _Jena 0.25m __ $R_\textrm{B}$ _ _ ___ 100 only first half_of transit observed
13 _ 834 _ _OSN_1.5m __ $R_\textrm{C}$ ___ _ _ 20 upcoming bad weather during ingress
14_ 840_ ___ Tenerife 1.2m $r_\textrm{S}$ _ _ _ 10_large fit errors
15 _848 Tenerife 1.2m_ $r_\textrm{S}$ _ _ __ 15 only egress phase observed
_16___854 _ _ _Tenerife 1.2m_ _ $r_\textrm{S}$_ __ 20 only ingress phase observed
_17 855 _ __Tenerife 1.2m _ $r_\textrm{S}$_ _ 20_jumps_in data,_no good fits possible
_ 18_ _861 _ _ _ _Tenerife 1.2m $r_\textrm{S}$__ ____ _25 bad observing_conditions
19_ 867__ _ Tenerife 1.2m_ _$r_\textrm{S}$ _ 25 bad observing conditions
20 906 _ _ _ Torun_0.6m_ _ $clear$_ __ ___10 only ingress_phase_observed
21 1001_____ Torun 0.6m _ $clear$__ _ __ _ _ 6 jumps in data,_no_good fits_possible
Analysis_{#sec:Analysis}
========
All_light_curves (except for the_Ankara 0.4m observation) are extracted from the reduced images using the same aperture photometry routines_(described_in section \[sec:GettingTheLightcurve\]) in order to_prevent systematic offsets between different transit_observations_due to different light_curve extraction methods. Afterwards we model the data_sets using the *JKTEBOP* algorithm_[@JKTEBOP], as well as_the Transit Analysing Package *TAP* [@T |
M}}^i\nu(E^{\varepsilon})+\varepsilon.$$ Similarly, $f_{\mathcal{M}}^i\nu(E)
\leq f_{\mathcal{M}}^i\mu(E^{\varepsilon})+\varepsilon$, and hence $LP(f_{\mathcal{M}}^i\mu,f_{\mathcal{M}}^i\nu)\leq\varepsilon$, which completes the proof.
The induced transformation $f_{\mathcal{K}}$
--------------------------------------------
Let $(X,d)$ be a compact metric space and $f:X\to X$ a continuous map. The Hausdorff metric $H$ on $\mathcal{K}(X)$ is defined by $$H(B,C)=\inf\{\varepsilon>0:B\subset C^{\varepsilon}\ \text{and}\ C\subset B^{\varepsilon} \},\ \ B,C\in\mathcal{K}(X).$$ This metric turns $\mathcal{K}(X)$ into a compact metric space. The induced map on $\mathcal{K}(X)$ is defined by $$\begin{aligned}
f_{\mathcal{K}}:\mathcal{K}(X)&\to\mathcal{K}(X)\\
B&\mapsto f(B).\end{aligned}$$ For $n\in\mathbb{N}$, denote by $H^n$ the Hausdorff metric induced by the $n$th Bowen metric of $(X,f)$, i.e., $$\label{hn}
H^n(B,C)=\inf\{\varepsilon>0:B\subset C_n^{\varepsilon}\ \text{and}\ C\subset B_n^{\varepsilon} \},$$ where $A_n^{\varepsilon}=\{x\in X:d_n(x,A)<\varepsilon \}$ for a set $A$. Moreover, set $$H_n(B,C)=\max\{H(f_\mathcal{K}^iB,f_\mathcal{K}^iC):i=0,1,\ldots,n-1 \},$$ the $n$th Bowen metric for $f_{\mathcal{K}}$. Denote by $\text{diam}(B,d_n)$ the diameter of a set $B\subset X$ under the metric $d_n$.
\[fk\] Let $(X,f)$ be a topological dynamical system. For $n\in\mathbb{N}$, $B,C\in\mathcal{K}(X | M}}^i\nu(E^{\varepsilon})+\varepsilon.$$ Similarly, $ f_{\mathcal{M}}^i\nu(E)
\leq f_{\mathcal{M}}^i\mu(E^{\varepsilon})+\varepsilon$, and hence $ LP(f_{\mathcal{M}}^i\mu, f_{\mathcal{M}}^i\nu)\leq\varepsilon$, which completes the proof.
The induced transformation $ f_{\mathcal{K}}$
--------------------------------------------
lease $ (ten, d)$ be a compact metric space and $ f: X\to X$ a continuous function. The Hausdorff metric $ H$ on $ \mathcal{K}(X)$ is defined by $ $ H(B, C)=\inf\{\varepsilon>0: B\subset C^{\varepsilon}\ \text{and}\ C\subset B^{\varepsilon } \},\ \ B, C\in\mathcal{K}(X).$$ This metric flex $ \mathcal{K}(X)$ into a compact metric outer space. The induced map on $ \mathcal{K}(X)$ is define by $ $ \begin{aligned }
f_{\mathcal{K}}:\mathcal{K}(X)&\to\mathcal{K}(X)\\
B&\mapsto f(B).\end{aligned}$$ For $ n\in\mathbb{N}$, denote by $ H^n$ the Hausdorff metric induce by the $ n$th Bowen metric of $ (X, f)$, i.e., $ $ \label{hn }
H^n(B, C)=\inf\{\varepsilon>0: B\subset C_n^{\varepsilon}\ \text{and}\ C\subset B_n^{\varepsilon } \},$$ where $ A_n^{\varepsilon}=\{x\in X: d_n(x, A)<\varepsilon \}$ for a set $ A$. Moreover, set $ $ H_n(B, C)=\max\{H(f_\mathcal{K}^iB, f_\mathcal{K}^iC):i=0,1,\ldots, n-1 \},$$ the $ n$th Bowen metric for $ f_{\mathcal{K}}$. Denote by $ \text{diam}(B, d_n)$ the diameter of a fit $ B\subset X$ under the metric $ d_n$.
\[fk\ ] Let $ (X, f)$ be a topological dynamic system. For $ n\in\mathbb{N}$, $ B, C\in\mathcal{K}(X | M}}^i\nk(E^{\varepsilon})+\varepsilon.$$ Slmilarly, $f_{\mathcal{M}}^i\nu(E)
\leq f_{\mathcau{M}}^i\mu(E^{\varepsilon})+\varepsilon$, aid hwnce $OP(f_{\mathcal{M}}^i\mu,f_{\mathcal{O}}^i\nu)\leq\vagepsilon$, qhici completes the 'doof.
The inducsf trcnwformation $f_{\majhcal{K}}$
--------------------------------------------
Let $(X,d)$ be a compact kegrnc space and $f:X\to X$ a continuous map. The Haisforff metric $H$ on $\kwthczl{K}(X)$ is defined by $$H(B,C)=\inf\{\varepsiloh>0:B\subseu C^{\varepsilon}\ \text{snd}\ C\subset B^{\varepsilon} \},\ \ H,C\in\lathcal{K}(X).$$ This metgic turns $\mqthcwo{K}(X)$ into a cumpact metgnc space. Thg induced map on $\mathcal{K}(X)$ is dewined by $$\begin{aoitnef}
x_{\mathcal{K}}:\mavhcal{K}(Q)&\to\mathcal{K}(X)\\
B&\mapsto f(T).\end{alibned}$$ For $n\in\mabhbb{N}$, debote by $H^n$ the Hausdocff metric induced bi the $n$th Tocen metric of $(X,f)$, i.e., $$\lqbwl{hn}
H^t(B,C)=\ivd\{\vafepaikoh>0:B\subsft R_n^{\varepsiloh}\ \text{and}\ C\wubset B_n^{\varepsilon} \},$$ ryere $A_n^{\varepsjlon}=\{x\ig V:d_n(x,A)<\varepsilon \}$ for a set $A$. Moreover, stt $$H_n(G,C)=\max\{H(f_\mathcal{K}^iB,f_\mathcql{K}^iC):i=0,1,\ldots,n-1 \},$$ the $n$th Bowen meeric for $f_{\mathcal{K}}$. Denote by $\text{diam}(B,d_n)$ the diamater kw a sct $B\rybdet X$ under the metric $d_n$.
\[fk\] Let $(X,f)$ be a topoljfivak dynamical syftem. For $n\im\mwtnfb{N}$, $B,C\in\mathcxl{K}(X | M}}^i\nu(E^{\varepsilon})+\varepsilon.$$ Similarly, $f_{\mathcal{M}}^i\nu(E) \leq f_{\mathcal{M}}^i\mu(E^{\varepsilon})+\varepsilon$, and hence completes proof. The transformation $f_{\mathcal{K}}$ -------------------------------------------- metric and $f:X\to X$ continuous map. The metric $H$ on $\mathcal{K}(X)$ is defined $$H(B,C)=\inf\{\varepsilon>0:B\subset C^{\varepsilon}\ \text{and}\ C\subset B^{\varepsilon} \},\ \ B,C\in\mathcal{K}(X).$$ This metric turns $\mathcal{K}(X)$ into compact metric space. The induced map on $\mathcal{K}(X)$ is defined by $$\begin{aligned} f_{\mathcal{K}}:\mathcal{K}(X)&\to\mathcal{K}(X)\\ f(B).\end{aligned}$$ $n\in\mathbb{N}$, by the Hausdorff metric induced by the $n$th Bowen metric of $(X,f)$, i.e., $$\label{hn} H^n(B,C)=\inf\{\varepsilon>0:B\subset C_n^{\varepsilon}\ \text{and}\ B_n^{\varepsilon} \},$$ where $A_n^{\varepsilon}=\{x\in X:d_n(x,A)<\varepsilon \}$ for a $A$. Moreover, set $$H_n(B,C)=\max\{H(f_\mathcal{K}^iB,f_\mathcal{K}^iC):i=0,1,\ldots,n-1 the $n$th Bowen metric for Denote $\text{diam}(B,d_n)$ the of set X$ under the $d_n$. \[fk\] Let $(X,f)$ be a topological dynamical system. For $n\in\mathbb{N}$, $B,C\in\mathcal{K}(X | M}}^i\nu(E^{\varepsilon})+\varepsilon.$$ similarly, $f_{\MathcAl{M}}^I\nu(e)
\lEq f_{\mAthcAl{M}}^i\mu(E^{\varepsiLOn})+\vaRepsilon$, and hence $LP(f_{\matHcal{M}}^I\mU,F_{\matHCaL{M}}^i\nu)\Leq\varePSiLON$, whIcH cOmpLeTEs The prOof.
the induCed transfoRmaTiOn $f_{\mathcal{K}}$
--------------------------------------------
LET $(X,D)$ be a compacT meTric space and $F:X\tO X$ a conTiNuoUS map. THe HAusdoRff metRIc $H$ on $\mAthcal{K}(X)$ iS dEFined bY $$h(B,C)=\inf\{\vAREpSiloN>0:B\subset C^{\varepsilON}\ \tEXt{and}\ C\subset B^{\vArepsiLoN} \},\ \ b,C\IN\MatHcaL{K}(X).$$ This metRiC turnS $\Mathcal{k}(x)$ iNTO A coMPact metric spaCe. The induceD Map On $\mathCaL{K}(X)$ IS definEd by $$\bEgIN{alIgned}
f_{\mathcAl{K}}:\mAthcal{K}(X)&\tO\mathcAL{K}(X)\\
B&\mapSTo f(B).\end{AligneD}$$ FoR $n\iN\matHBb{n}$, dEnoTe BY $H^n$ THe hauSDorFf metric InDuCed by The $n$TH bOWen mEtrIc of $(x,f)$, i.e., $$\lAbel{hn}
H^n(B,C)=\inf\{\VarEpsiLOn>0:B\SubseT C_n^{\vaRepsIlOn}\ \texT{and}\ C\sUbset b_n^{\Varepsilon} \},$$ where $a_n^{\vaRepsilon}=\{x\In X:D_n(X,A)<\vArEpsilON \}$ for a sEt $A$. morEover, seT $$H_n(B,C)=\maX\{h(f_\mAtHCAL{K}^IB,f_\mathcal{K}^iC):i=0,1,\ldotS,n-1 \},$$ THE $n$Th Bowen mEtric fOR $f_{\MaTHcal{K}}$. DenOtE by $\Text{DIAm}(B,d_n)$ The dIAmEter of a sEt $B\subSEt x$ uNder the MeTric $d_n$.
\[Fk\] let $(x,f)$ bE a topOLogiCal dynAmical syStem. FOR $n\in\mathbb{N}$, $B,C\iN\Mathcal{K}(X | M}}^i\nu(E^{\varepsilon})+ \varepsilo n.$$Sim ila rl y, $ f_{\ mathcal{M}}^i\ n u(E)
\leq f_{\mathcal{ M}}^i \m u (E^{ \ va repsi lon})+\ v ar e p sil on $, an dh en ce $L P(f _{\math cal{M}}^i\ mu, f_ {\mathcal{M} } ^i \nu)\leq\v are psilon$, whi chcomple te s t h e pro of.
The induc e d tran sformatio n$ f_{\ma t hcal{K} } $
- ---- ----------------- - -- - -------------- ----
Le t $ ( X ,d) $ b e a compac tmetri c spacea nd $ f :X\ t o X$ a contin uous map. T h e H ausdor ff me t ric $H $ on$\ m ath cal{K}(X)$is d efined by $$H(B , C)=\inf \ {\varep silon> 0:B \su bset C^ {\ var ep s ilo n }\ \t e xt{ and}\ C\ su bs et B^ {\va r e p s ilon } \ },\\ B,C \in\mathcal{K }(X ).$$ Thi s met ric t urns $ \math cal{K} (X)$in to a compact me tric space. T hein duc ed mapo n $\ma thc al{ K}(X)$is defi n edby $ $ \b egin{aligned}
f_{\ ma t h ca l{K}}:\m athcal { K} (X ) &\to\mat hc al{ K}(X ) \ \
B&\ maps t of(B).\en d{alig n ed }$ $ For $ n\ in\mat hb b{N }$, deno t e by $H^n$ the Hau sdorf f metric induce d by the $n$th Bo w e nm etri c o f $(X,f)$,i.e. , $$\ labe l {h n} H^ n(B,C )= \ in f \{\varepsilon>0:B\s ub set C_ n^{\v arepsilon}\ \ text{and}\ C \ subset B _n^{ \ va r epsilon} \},$$ wher e $A_n^{\v a repsilon }=\{x \in X:d_ n(x,A)<\v a r epsilon\}$ fo r a se t $A $. Moreover,s e t $$ H_ n(B,C)= \ma x\{H(f_ \ma thc al{ K}^ iB ,f_\mathc al{K}^iC ): i= 0, 1, \ld ots,n - 1 \},$$th e $ n$ thBowen metric for$f_{ \m at h cal {K}}$.D en o t e by $ \t ext{ dia m} (B,d_ n)$t hediamete r of a se t $ B \sub se tX$ unde r the metric$d _n$.
\[fk \] Le t $(X, f ) $ be a t opological dynamical sy s tem. Fo r $ n\in\ math bb{N}$, $ B,C \in\ma thc a l{K}(X | M}}^i\nu(E^{\varepsilon})+\varepsilon.$$ Similarly,_$f_{\mathcal{M}}^i\nu(E)
_ \leq f_{\mathcal{M}}^i\mu(E^{\varepsilon})+\varepsilon$,_and hence_$LP(f_{\mathcal{M}}^i\mu,f_{\mathcal{M}}^i\nu)\leq\varepsilon$,_which completes_the_proof.
The induced transformation_$f_{\mathcal{K}}$
--------------------------------------------
Let $(X,d)$ be_a compact metric space_and $f:X\to X$_a_continuous map. The Hausdorff metric $H$ on $\mathcal{K}(X)$ is defined by $$H(B,C)=\inf\{\varepsilon>0:B\subset C^{\varepsilon}\ \text{and}\_C\subset_B^{\varepsilon} \},\_\_B,C\in\mathcal{K}(X).$$_This metric turns $\mathcal{K}(X)$ into_a compact metric space. The_induced map_on $\mathcal{K}(X)$ is defined by $$\begin{aligned}
f_{\mathcal{K}}:\mathcal{K}(X)&\to\mathcal{K}(X)\\
B&\mapsto f(B).\end{aligned}$$ For_$n\in\mathbb{N}$,_denote by $H^n$_the Hausdorff metric induced by the $n$th Bowen metric_of $(X,f)$, i.e., $$\label{hn}
_ H^n(B,C)=\inf\{\varepsilon>0:B\subset C_n^{\varepsilon}\_\text{and}\_C\subset_B_n^{\varepsilon} \},$$ where $A_n^{\varepsilon}=\{x\in_X:d_n(x,A)<\varepsilon \}$ for a set $A$._Moreover, set $$H_n(B,C)=\max\{H(f_\mathcal{K}^iB,f_\mathcal{K}^iC):i=0,1,\ldots,n-1 \},$$ the $n$th_Bowen metric for $f_{\mathcal{K}}$. Denote by $\text{diam}(B,d_n)$_the diameter of a set $B\subset_X$ under the metric $d_n$.
\[fk\]_Let $(X,f)$_be a topological dynamical system._For $n\in\mathbb{N}$, $B,C\in\mathcal{K}(X |
Substitution of into yields with $$\begin{aligned}
\label{eq:h+-}
h^\pm &= h_0^\pm + b_0\cot\theta\nonumber\\[10pt]
&= {\mp\,\partial_\theta + \left(b_0\mp\frac{1}{2}\right)\cot\theta
+\frac{{\text{i}}\partial_\varphi}{\sin\theta}}.\end{aligned}$$ As in the zero field case, with describes a “Dirac Hamiltonian” in the sense that $$\begin{aligned}
\label{eq:h^2}
h^2&=\left(\!
\begin{array}{cc}
h^+h^-&0 \\0&h^-h^+
\end{array}\!\!\right)\nonumber\\[5pt]
&=\left(\!
\begin{array}{cc}
{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}+s_0\left|_{s_0=b_0+\frac{1}{2}}\right.&0 \\[10pt]
0&{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}-s_0\left|_{s_0=b_0-\frac{1}{2}}\right.
\end{array}\!\right)\end{aligned}$$ is diagonal. Apart from an overall numerical factor, $$\begin{aligned}
\label{eq:qhslambda^2}
{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}=
-\frac{1}{\sin\theta}\partial_\theta
\left(\sin\theta\,\partial_\theta\right)
-\frac{1}{\sin^2\theta}
\left(\partial_\varphi-{\text{i}}s_0\cos\theta\right)^2\end{aligned}$$ is the Hamiltonian of a massive electron moving on a sphere with a monopole of strength $4\pi s_0$ in the center [@haldane83prl605]. The LLs for massive electrons on the sphere are spanned by two mutually commuting SU(2) algebras [@greiter11prb115129], one for the cyclotron momentum (${\ensuremath{\boldsymbol{S}}}$) and one for the guiding center momentum (${\ensuremath{\boldsymbol{L}}}$). The Casimir of | Substitution of into yields with $ $ \begin{aligned }
\label{eq: h+- }
h^\pm & = h_0^\pm + b_0\cot\theta\nonumber\\[10pt ]
& = { \mp\,\partial_\theta + \left(b_0\mp\frac{1}{2}\right)\cot\theta
+ \frac{{\text{i}}\partial_\varphi}{\sin\theta}}.\end{aligned}$$ As in the zero field case, with describes a “ Dirac Hamiltonian ” in the common sense that $ $ \begin{aligned }
\label{eq: h^2 }
h^2&=\left(\!
\begin{array}{cc }
h^+h^-&0 \\0&h^-h^+
\end{array}\!\!\right)\nonumber\\[5pt ]
& = \left(\!
\begin{array}{cc }
{ \ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}+s_0\left|_{s_0 = b_0+\frac{1}{2}}\right.&0 \\[10pt ]
0&{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}-s_0\left|_{s_0 = b_0-\frac{1}{2}}\right.
\end{array}\!\right)\end{aligned}$$ is aslant. aside from an overall numerical factor, $ $ \begin{aligned }
\label{eq: qhslambda^2 }
{ \ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}=
-\frac{1}{\sin\theta}\partial_\theta
\left(\sin\theta\,\partial_\theta\right)
-\frac{1}{\sin^2\theta }
\left(\partial_\varphi-{\text{i}}s_0\cos\theta\right)^2\end{aligned}$$ is the Hamiltonian of a massive electron moving on a celestial sphere with a monopole of strength $ 4\pi s_0 $ in the center [ @haldane83prl605 ]. The LLs for massive electron on the sphere are spanned by two mutually commuting SU(2) algebras [ @greiter11prb115129 ], one for the cyclotron momentum ($ { \ensuremath{\boldsymbol{S}}}$) and one for the steer center momentum ($ { \ensuremath{\boldsymbol{L}}}$). The Casimir of |
Subdtitution of into yields with $$\begin{aliguwd}
\lauel{eq:h+-}
h^\pm &= h_0^\po + b_0\cot\theta\nonumber\\[10pt]
&= {\mp\,\pertiql_\theuc + \left(b_0\mp\frac{1}{2}\right)\cut\theta
+\frac{{\tezt{i}}\pertial_\varphi}{\sin\tista}}.\end{aligned}$$ Ws iu vhe zero field gase, with dascribes a “Dirdc Hcmiltonian” in the sense that $$\begin{alygned}
\kahel{eq:h^2}
h^2&=\left(\!
\btgig{arrzj}{cg}
h^+h^-&0 \\0&h^-h^+
\end{array}\!\!\right)\nonhmber\\[5pt]
&=\left(\!
\begon{array}{cc}
{\ensuremath{\blldsjmbol{\Lambda}}}^2_{s_0}+s_0\left|_{s_0=h_0+\frac{1}{2}}\right.&0 \\[10kf]
0&{\ensurematf{\boldsymbol{\Lambda}}}^2_{s_0}-s_0\lefj|_{s_0=b_0-\frac{1}{2}}\right.
\end{array}\!\right)\end{xlignzd}$$ is diagobao. Aodrt from an overwll numerical factor, $$\begin{akigned}
\label{ea:qhvlanbda^2}
{\ensuremath{\boldsbmbol{\Lambda}}}^2_{s_0}=
-\fras{1}{\sin\theta}\[axtial_\theta
\left(\sib\tyeta\,\pdrtidl_\thdra\rkghu)
-\frac{1}{\sln^2\tieta}
\lert(\partial_\vaephi-{\text{i}}s_0\cos\theta\rogrn)^2\rnd{aligned}$$ ia the Raiiltonian of a massive electron moving mn z sphere with a monopolw of strength $4\pi s_0$ in the centqr [@haldane83prl605]. The LLs for massive electrons on the sphece art wpannde hy two mutually commuting SU(2) algebras [@greiter11pwg115129], pnv for the cyclotrjn momentum (${\ejsitemath{\boldsymbul{S}}}$) anb ohe for the guiding center momebtum (${\ensuwemayh{\boldsymbol{L}}}$). The Casimir od | Substitution of into yields with $$\begin{aligned} \label{eq:h+-} h_0^\pm b_0\cot\theta\nonumber\\[10pt] &= + \left(b_0\mp\frac{1}{2}\right)\cot\theta +\frac{{\text{i}}\partial_\varphi}{\sin\theta}}.\end{aligned}$$ case, describes a “Dirac in the sense $$\begin{aligned} \label{eq:h^2} h^2&=\left(\! \begin{array}{cc} h^+h^-&0 \\0&h^-h^+ &=\left(\! \begin{array}{cc} {\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}+s_0\left|_{s_0=b_0+\frac{1}{2}}\right.&0 \\[10pt] 0&{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}-s_0\left|_{s_0=b_0-\frac{1}{2}}\right. \end{array}\!\right)\end{aligned}$$ is diagonal. Apart from an overall numerical $$\begin{aligned} \label{eq:qhslambda^2} {\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}= -\frac{1}{\sin\theta}\partial_\theta \left(\sin\theta\,\partial_\theta\right) -\frac{1}{\sin^2\theta} \left(\partial_\varphi-{\text{i}}s_0\cos\theta\right)^2\end{aligned}$$ is the Hamiltonian of a massive moving a with monopole of strength $4\pi s_0$ in the center [@haldane83prl605]. The LLs for massive electrons on the are spanned by two mutually commuting SU(2) algebras one for the cyclotron (${\ensuremath{\boldsymbol{S}}}$) and one for the center (${\ensuremath{\boldsymbol{L}}}$). The of |
Substitution of into yields wIth $$\begin{alIgned}
\LabEl{eQ:h+-}
H^\pm &= h_0^\Pm + b_0\cOt\theta\nonumbeR\\[10Pt]
&= {\mp\,\Partial_\theta + \left(b_0\mp\fraC{1}{2}\righT)\cOT\theTA
+\fRac{{\teXt{i}}\partIAl_\VARphI}{\sIn\TheTa}}.\ENd{AlignEd}$$ AS in the zEro field caSe, wItH describes a “DIRaC HamiltoniAn” iN the sense thaT $$\beGin{aliGnEd}
\lABel{eq:H^2}
h^2&=\lEft(\!
\beGin{arrAY}{cc}
h^+h^-&0 \\0&h^-H^+
\end{array}\!\!\RiGHt)\nonuMBer\\[5pt]
&=\leFT(\!
\BeGin{aRray}{cc}
{\ensuremath{\bOLdSYmbol{\Lambda}}}^2_{s_0}+s_0\lEft|_{s_0=b_0+\fRaC{1}{2}}\RiGHT.&0 \\[10pt]
0&{\EnsUremath{\bolDsYmbol{\lAmbda}}}^2_{s_0}-s_0\LEfT|_{S_0=B_0-\FraC{1}{2}}\Right.
\end{array}\!\Right)\end{aliGNed}$$ Is diagOnAl. APArt froM an ovErALl nUmerical facTor, $$\bEgin{alignEd}
\labeL{Eq:qhslaMBda^2}
{\ensuRemath{\BolDsyMbol{\lAmBdA}}}^2_{s_0}=
-\fRaC{1}{\Sin\THeTa}\pARtiAl_\theta
\lEfT(\sIn\theTa\,\paRTIAL_\theTa\rIght)
-\Frac{1}{\sIn^2\theta}
\left(\paRtiAl_\vaRPhi-{\Text{i}}S_0\cos\tHeta\RiGht)^2\enD{alignEd}$$ is tHe hamiltonian of a mAssiVe electroN moViNg oN a SpherE With a mOnoPolE of streNgth $4\pi s_0$ IN thE cENTEr [@Haldane83prl605]. The LLs foR mASSiVe electrOns on tHE sPhERe are spaNnEd bY two MUTuallY comMUtIng SU(2) algEbras [@gREiTeR11prb115129], one FoR the cyClOtrOn mOmentUM (${\ensUrematH{\boldsymBol{S}}}$) aND one for the guidINg center momenTUm (${\ENSuREmatH{\boLdsymbol{L}}}$). ThE CasIMir oF |
Substitution of into yiel ds with $$ \begi n{a lig ne d}
\la bel{eq:h+-}
h ^\pm &= h_0^\pm + b_0\cot\ theta \n o numb e r\ \[10p t]
&= {\ m p \,\ pa rt ial _\ t he ta +\le ft(b_0\ mp\frac{1} {2} \r ight)\cot\th e ta
+\fra c{{ \text{i}}\pa rti al_\va rp hi} { \sin\ the ta}}. \end{a l igned} $$ As inth e zerof ield ca s e ,with describes a “Dir a cH amiltonian” in the s en s et h at$$\ begin{alig ne d}
\ label{e q :h ^ 2 }
h ^2&=\left(\! \begin{ a rra y}{cc}
h^+h^ -&0 \ \0 & h^- h^+
\en d{ar ray}\!\!\ right) \ nonumbe r \\[5pt]
&=\ lef t(\ !
\ be gin {a r ray } {c c} {\ensu re ma th{\b olds y m b o l{\L amb da}} }^2_{ s_0}+s_0\left |_{ s_0= b _0+ \frac {1}{2 }}\r ig ht.&0 \\[10 pt]
0&{\ensurema th{\ boldsymbo l{\ La mbd a} }}^2_ { s_0}-s _0\ lef t|_{s_0 =b_0-\f r ac{ 1} { 2 } }\ right.
\end{ar ra y } \! \right)\ end{al i gn ed } $$ is di ag ona l. A p a rt fr om a n o verall n umeric a lfa ctor, $ $\ begin{ al ign ed}
\l a bel{ eq:qhs lambda^2 }
{ \ ensuremath{\bo l dsymbol{\Lamb d a} } } ^2 _ {s_0 }= -\frac {1}{ \ sin\ thet a }\ par t ial_\ theta
\left(\sin\theta\,\ pa rtial_ \thet a\right)
-\frac{1} { \ s in^2\the ta} \left(\partia l_\va rphi-{\tex t {i}}s_0\ cos\t heta\rig ht)^2\end { a ligned}$ $ i s t heHam i l to nian of a mas s i ve e le ctron m ovi ng on a sp her e w ith a monopole of stre ng th $ 4\ pis_0$i n the ce nt er[@ hal dane8 3 prl605 ]. Th e LL sfo r ma ssive e l ec t r onson t he s phe re arespan n edby twomutuallycom m utin gSU (2) alg ebras [@greit er 11prb11512 9] , o ne for t he cyclo tron momentum (${\ensur e math{\b old symbo l{S} }}$) andone for t heg uiding cente r mom en tum ( ${\en s u re mat h{ \boldsymbo l { L}} }$).Th e Ca simir o f |
Substitution of_into yields_with $$\begin{aligned}
\label{eq:h+-}
_ h^\pm_&=_h_0^\pm +_b_0\cot\theta\nonumber\\[10pt]
_ &= {\mp\,\partial_\theta_+ \left(b_0\mp\frac{1}{2}\right)\cot\theta
_ +\frac{{\text{i}}\partial_\varphi}{\sin\theta}}.\end{aligned}$$ As_in the zero_field_case, with describes a “Dirac Hamiltonian” in the sense that $$\begin{aligned}
\label{eq:h^2}
_h^2&=\left(\!
_ __\begin{array}{cc}
_ _h^+h^-&0 \\0&h^-h^+
_\end{array}\!\!\right)\nonumber\\[5pt]
_&=\left(\!
\begin{array}{cc}
__ {\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}+s_0\left|_{s_0=b_0+\frac{1}{2}}\right.&0 \\[10pt]
_ 0&{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}-s_0\left|_{s_0=b_0-\frac{1}{2}}\right.
_\end{array}\!\right)\end{aligned}$$ is diagonal. Apart from an_overall numerical factor,_$$\begin{aligned}
__\label{eq:qhslambda^2}
{\ensuremath{\boldsymbol{\Lambda}}}^2_{s_0}=
_ -\frac{1}{\sin\theta}\partial_\theta
_ \left(\sin\theta\,\partial_\theta\right)
_ -\frac{1}{\sin^2\theta}
\left(\partial_\varphi-{\text{i}}s_0\cos\theta\right)^2\end{aligned}$$_is the Hamiltonian of a massive_electron moving on a sphere_with a_monopole of strength $4\pi s_0$_in the center [@haldane83prl605]._The LLs_for massive electrons_on the sphere are spanned by_two mutually commuting_SU(2) algebras [@greiter11prb115129], one for the cyclotron_momentum_(${\ensuremath{\boldsymbol{S}}}$) and one_for_the_guiding center_momentum (${\ensuremath{\boldsymbol{L}}}$). The_Casimir_of |
N=2$ SCFT’s and their generalizations have been of much recent interest [@Gaiotto2012h; @Gaiotto2009] and the $\N=1$ case has also been studied [@Benini:2009mz; @Bah:2012dg].
By embedding the $AdS_4$ black holes in M-theory we can see them as M2-brane wrapping a Riemann surface. For particular charges, the bulk system will flow to $AdS_2 \times \Sigma_g$ in the IR and represents a black hole with regular horizon. The original examples found in [@Caldarelli1999] can be reinterpreted in this way; it has four equal magnetic charges and can be embedded in $AdS_4 \times S^7$. The explicit analytic solution is known and it involves constant scalars and a hyperbolic horizon. A generalization of [@Maldacena:2000mw] to M2-branes wrapping $\Sig_g$ was performed in [@Gauntlett2002] where certain very symmetric twists were considered. Fully regular solutions for M2 branes wrapping a two-sphere with running scalars were finally found in [@Cacciatori:2009iz] in the form of $AdS_4$ black holes. It is note-worthy that of all these scenarios of branes wrapping Riemann surfaces, the complete analytic solution for general charges is known only for M2-branes on $\Sig_g$ with magnetic charges [@Cacciatori:2009iz].
One way to generalize these constructions of branes wrapped on $\Sig_g$ is to have more general transverse spaces. This is the focus of this article. For M5-branes one can orbifold $S^4$ while for D3-branes one can replace $S^5$ by an arbitrary $SE_5$ manifold and indeed a suitable consistent truncation on $T^{11}$ has indeed been constructed [@Bena:2010pr; @Cassani:2010na]. For M2-branes one can replace $S^7$ by a seven-dimensional Sasaki-Einstein manifold $SE_7$ and, as discussed above, the work of [@Cassani:2012pj] provides us with a rich set of consistent truncations to explore. Interestingly, in our analysis we find that there are no solutions for pure M2-brane backgrounds, there must be additional electric and magnetic charges corresponding to wrapped M2 and M5 | N=2 $ SCFT ’s and their generalizations have been of much recent interest [ @Gaiotto2012h; @Gaiotto2009 ] and the $ \N=1 $ case has besides been study [ @Benini:2009mz; @Bah:2012dg ].
By embedding the $ AdS_4 $ black hole in molarity - theory we can see them as M2 - brane wrap a Riemann airfoil. For particular cathexis, the bulk system will flow to $ AdS_2 \times \Sigma_g$ in the IR and represent a black hole with unconstipated horizon. The original example found in [ @Caldarelli1999 ] can be reinterpret in this way; it has four adequate magnetic charges and can be embedded in $ AdS_4 \times S^7$. The denotative analytic solution is known and it involves constant scalars and a hyperbolic horizon. A generalization of [ @Maldacena:2000mw ] to M2 - branes wrap $ \Sig_g$ was performed in [ @Gauntlett2002 ] where certain very symmetric twists were considered. amply regular solutions for M2 branes wrapping a two - sphere with running scalars were finally found in [ @Cacciatori:2009iz ] in the form of $ AdS_4 $ black holes. It is note - worthy that of all these scenarios of branes wrapping Riemann surfaces, the complete analytic solution for cosmopolitan cathexis is known entirely for M2 - branes on $ \Sig_g$ with charismatic charge [ @Cacciatori:2009iz ].
One way to generalize these construction of branes wrapped on $ \Sig_g$ is to have more general transverse space. This is the focus of this article. For M5 - branes one can orbifold $ S^4 $ while for D3 - branes one can replace $ S^5 $ by an arbitrary $ SE_5 $ manifold and indeed a suitable consistent truncation on $ T^{11}$ has indeed been constructed [ @Bena:2010pr; @Cassani:2010na ]. For M2 - branes one can substitute $ S^7 $ by a seven - dimensional Sasaki - Einstein manifold $ SE_7 $ and, as discussed above, the work of [ @Cassani:2012pj ] provides us with a full-bodied set of reproducible truncations to explore. Interestingly, in our analysis we find that there are no solution for pure M2 - brane backgrounds, there must be additional electric and magnetic charges corresponding to wrapped M2 and M5 | N=2$ SFFT’s and their generalizxtions have beeu of murh receht interdst [@Gaiotto2012h; @Gaiotto2009] and the $\N=1$ case has also been studied [@Benini:2009mz; @Bah:2012dg].
By embtdding the $AdS_4$ bledk holes in M-fmeory xe can see them as M2-brane wrapping a Riamxnu surface. For particular charges, the bulk sushem will flow jo $AdX_2 \timss \Sigma_g$ in the IR and represents a blacn hole with rrgular horizon. The originap exwmples found in [@Capdarelli1999] cab be eeinterpretea in this way; it has fkur equal magnetic charges and zan bz embedded un $AdD_4 \times S^7$. Thx explpcit analytic solutiot is knpwn and it invplvxs cinstant scalars and a hyperbolic horizog. A generdlnzation of [@Maldacena:2000mq] ro M2-btanes wraopine $\Sjg_j$ wzs pervorjed in [@Gauhtlett2002] wherw certain very symmttryb twists were consiqewed. Fully regular solutions for M2 branev wdapping a two-sphere wity running scalars werg finally sound in [@Cacciatori:2009iz] in the form of $AdS_4$ black hones. Iv ks uite-wofrhj that of all these scenarios of branes wrappyhg Rpemann surfaces, tme complete analytoc spjution for geveral egadges is known only for M2-btanes in $\Sig_g$ wyth kagnetic charges [@Cacciatori:2009uz].
One way to teneralize these clnstructionr of branrs wrapped on $\Sig_g$ is tu habe more genfral tranaxerse spaces. Thir ix dhe focus of this article. Sor M5-branxs onz can orcifokd $S^4$ wrile for D3-hranes one can replace $S^5$ by au arbhtrary $SE_5$ lanifold and indeed a suitable rpnsistent troncdtimn on $T^{11}$ kas inceed been conftructed [@Bena:2010pt; @Cassani:2010ua]. For M2-branes ohe can ceplace $S^7$ by a seven-dimenvlonal Sasaki-Xinstein ianidold $SE_7$ and, xs discussed anove, the cjrk if [@Cassani:2012pj] provives os with a rich sec if consistent ttunzatyojs tj explore. Intaresgineky, in our analysls de fond that there are nm somutions for pure M2-nrwne backtrounds, ehere must be additional electrlc anv magnxtic cnardes corresponding to wrapped M2 and M5 | N=2$ SCFT’s and their generalizations have been recent [@Gaiotto2012h; @Gaiotto2009] the $\N=1$ case @Bah:2012dg]. embedding the $AdS_4$ holes in M-theory can see them as M2-brane wrapping Riemann surface. For particular charges, the bulk system will flow to $AdS_2 \times in the IR and represents a black hole with regular horizon. The original found [@Caldarelli1999] be in this way; it has four equal magnetic charges and can be embedded in $AdS_4 \times The explicit analytic solution is known and it constant scalars and a horizon. A generalization of [@Maldacena:2000mw] M2-branes $\Sig_g$ was in where very symmetric twists considered. Fully regular solutions for M2 branes wrapping a two-sphere with running scalars were finally found in in the $AdS_4$ black It note-worthy of all these branes wrapping Riemann surfaces, the complete general charges is known only for M2-branes on with magnetic [@Cacciatori:2009iz]. One way to generalize these of branes wrapped on $\Sig_g$ is to have general transverse spaces. This is the focus of this article. For M5-branes one can orbifold for D3-branes one can $S^5$ by an $SE_5$ and a consistent truncation $T^{11}$ has indeed been constructed [@Bena:2010pr; @Cassani:2010na]. For M2-branes one can $S^7$ by a seven-dimensional Sasaki-Einstein manifold $SE_7$ and, as discussed work [@Cassani:2012pj] provides us a rich set of truncations explore. Interestingly, in our find there for M2-brane there must be additional and magnetic charges corresponding to M2 and M5 | N=2$ SCFT’s and their generalizatIons have beEn of mUch RecEnT intEresT [@Gaiotto2012h; @GaiotTO2009] and The $\N=1$ case has also been stuDied [@BEnINi:2009mz; @bAh:2012Dg].
By eMbeddinG ThE $aDS_4$ bLaCk HolEs IN M-TheorY we Can see tHem as M2-branE wrApPing a Riemann SUrFace. For parTicUlar charges, tHe bUlk sysTeM wiLL flow To $ADS_2 \timEs \SigmA_G$ in the iR and reprEsENts a blACk hole wITH rEgulAr horizon. The origiNAl EXamples found in [@caldarElLI1999] cAN Be rEinTerpreted iN tHis waY; It has foUR eQUAL maGNetic charges aNd can be embeDDed In $AdS_4 \tImEs S^7$. tHe explIcit aNaLYtiC solution is KnowN and it invOlves cONstant sCAlars anD a hypeRboLic HoriZOn. a gEneRaLIzaTIoN of [@mAldAcena:2000mw] tO M2-BrAnes wRappING $\sIg_g$ wAs pErfoRmed iN [@Gauntlett2002] wheRe cErtaIN veRy symMetriC twiStS were ConsidEred. FUlLy regular solutiOns fOr M2 branes WraPpIng A tWo-sphERe with RunNinG scalarS were fiNAllY fOUND iN [@Cacciatori:2009iz] in the FoRM Of $adS_4$ black Holes. IT Is NoTE-worthy tHaT of All tHESe sceNariOS oF branes wRappinG riEmAnn surfAcEs, the cOmPleTe aNalytIC solUtion fOr generaL charGEs is known only fOR M2-branes on $\Sig_G$ WiTH MaGNetiC chArges [@CacciaTori:2009IZ].
One Way tO GeNerALize tHese cOnSTrUCtions of branes wrappEd On $\Sig_g$ Is to hAve more generaL transversE SPAces. This Is thE FoCUs of this articlE. For M5-Branes one cAN orbifolD $S^4$ whiLe for D3-brAnes one caN REplace $S^5$ bY an ArbItrAry $se_5$ MaNifold and indeED A suiTaBle consIstEnt trunCatIon On $T^{11}$ Has InDeed been cOnstructEd [@beNa:2010Pr; @casSani:2010nA]. for M2-branEs One CaN rePlace $s^7$ By a sevEn-dimEnsiOnAl sAsaKi-EinstEIn MANifoLd $sE_7$ And, aS diScUssed AbovE, The Work of [@CAssani:2012pj] pRovIDes uS wItH a rich sEt of consistenT tRuncations To ExpLore. InTERestinglY, in our analysis we find thaT There arE no SolutIons For pure M2-bRanE backgRouNDs, therE must bE addiTiOnaL ELectrIC AnD maGnEtic chargeS COrrEsponDiNg to Wrapped m2 and M5 | N=2$ SCFT’s and their gene ralization s hav e b een o f mu ch r ecent interest [@Ga iotto2012h; @Gaiotto20 09] a nd the$ \N =1$ c ase has al s o be en s tud ie d [ @Beni ni: 2009mz; @Bah:2012 dg] .
By embeddin g t he $AdS_4$ bl ack holes in M- theory w e c a n see th em as M2-br a ne wra pping a R ie m ann su r face. F o r p arti cular charges, th e b u lk system will flowto $A d S _2\ti mes \Sigma _g $ int he IR a n dr e p res e nts a black h ole with re g ula r hori zo n.T he ori ginal e x amp les found i n [@ Caldarell i1999] can ber einterp retedinthi s wa y ;it ha sf our eq ual mag netic ch ar ge s and can b e embe dde d in $AdS _4 \times S^7 $.Thee xpl icitanaly ticso lutio n is k nownan d it involves c onst ant scala rsan d a h yperb o lic ho riz on. A gene ralizat i onof [ @ Ma ldacena:2000mw] to M 2 - br anes wra pping$ \S ig _ g$ was p er for medi n [@Ga untl e tt 2002] wh ere ce r ta in very s ym metric t wis tswerec onsi dered. Fully r egula r solutions for M2 branes wra p pi n g a two- sph ere with ru nnin g sca lars we ref inall y fou nd in [@Cacciatori:2009iz ]in the form of $AdS_4$ b lack holes . I t is not e-wo r th y that of all t hesescenarioso f branes wrap ping Rie mann surf a c es, thecom ple teana l y ti c solution fo r gene ra l charg esis know n o nly fo r M 2- branes on $\Sig_g $wi th m agn eticc harges [ @C acc ia tor i:200 9 iz].
One w ay t oge n era lize th e se c onst ru ct ions of b ranes wra p ped on $\S ig_g$ istoh avemo re genera l transversesp aces. This i s t he foc u s of this article. For M5-branes one can or bifol d $S ^4$ while fo r D3-b ran e s onecan re place $ S^5 $ by an a rb itr ar y $SE_5$ m a n ifo ld an dinde ed a su itable consistentt run cation on $T^ {11 }$ h a s i nde e db een c o nst r u cted [@Bena:201 0pr; @Cass an i :2 010na]. Fo r M2 -b ranes o ne canrepla c e $S^7$ by a sev en-dimens io nalS a sak i-Einstein manifol d $SE_7$a nd, a s d iscus sed above ,the work of [@ C ass ani:2 012pj] p rovide s uswi th a ric h set of consistent tru ncatio ns to ex plore. In ter e sti ngly, inouranalysis w e f ind that th e re ar e no so lut i ons f or p u re M2-bra n ebac k g ro unds, there m u stbe ad dit i onal e lect ric and magneticc harges corresp ondi n g to wr a pped M 2 and M5 | N=2$ SCFT’s_and their_generalizations have been of_much recent_interest_[@Gaiotto2012h; @Gaiotto2009]_and_the $\N=1$ case_has also been_studied [@Benini:2009mz; @Bah:2012dg].
By embedding_the $AdS_4$ black_holes_in M-theory we can see them as M2-brane wrapping a Riemann surface. For particular_charges,_the bulk_system_will_flow to $AdS_2 \times \Sigma_g$_in the IR and represents_a black_hole with regular horizon. The original examples found_in_[@Caldarelli1999] can be_reinterpreted in this way; it has four equal magnetic_charges and can be embedded in_$AdS_4 \times S^7$._The_explicit_analytic solution is known_and it involves constant scalars and_a hyperbolic horizon. A generalization of_[@Maldacena:2000mw] to M2-branes wrapping $\Sig_g$ was performed_in [@Gauntlett2002] where certain very symmetric_twists were considered. Fully regular_solutions for_M2 branes wrapping a two-sphere_with running scalars_were finally_found in [@Cacciatori:2009iz]_in the form of $AdS_4$ black_holes. It is_note-worthy that of all these scenarios_of_branes wrapping Riemann_surfaces,_the_complete analytic_solution for general_charges_is known_only_for M2-branes on $\Sig_g$ with magnetic_charges_[@Cacciatori:2009iz].
One way to generalize these constructions of_branes wrapped on $\Sig_g$_is_to have more general_transverse spaces. This is the_focus of this article. For M5-branes_one can_orbifold $S^4$_while for D3-branes one can replace $S^5$ by an arbitrary $SE_5$_manifold and indeed a suitable consistent_truncation on $T^{11}$ has_indeed been_constructed_[@Bena:2010pr; @Cassani:2010na]. For_M2-branes_one can_replace $S^7$ by a seven-dimensional Sasaki-Einstein manifold_$SE_7$ and,_as discussed above, the work of_[@Cassani:2012pj] provides us with_a_rich set of consistent truncations to_explore. Interestingly, in our analysis we_find that there are no_solutions_for_pure M2-brane backgrounds, there must_be additional electric and magnetic charges_corresponding to wrapped_M2 and M5 |
and/or $Z_{GC}$ show a MF significantly flatter than the cluster in the outer Galactic halo or farther from the Galactic plane. Indeed, M55 is near to the Galactic bulge, $R_{GC}=4.7$ kpc ($R_\odot=8.0$ kpc), and to the Galactic disk $Z_{GC}=-2.0$ kpc. Figure \[m55xrz\] shows that taking into account observing errors, M55 fairly fits into the relation given by [@Manu97], which is a refined version of the one found by [@DPC93]. A different conclusion has been reached by [@Mand96] using their uncorrected (for mass segregation) value for the MF of M55. As noted by the referee, M55 lies further from the average relation defined by the other clusters: of those with a similar abscissa ($0.0\pm0.2$), M55 is the one with the lowest value of $x$. It is not possible to identify the main source of this apparent enhanced mass-loss of M55 compared to the other clusters; a possible cause can be a orbit of the cluster that deeply penetrate into the bulge of the Galaxy. This cannot be confirmed until is performed a reliable measure of the proper motion of M55.
Radial density profile from star counts.\[M55rprof\]
====================================================
The CMD allows a unique way to obtain a reliable measure of the radial density profiles of GCs. In fact, the CMD allows us to sort out the stars belonging to the cluster, limiting the problems generated by the presence of the field stars. This also permits to extract radial profiles for distinct stellar masses.
We have first created a profile as in [@King68], in order to compare our results with the existing data in the literature. The comparison has been done with the radial density profile of M55 published by [@Pryor91] which includes the visual star counts of King et al.. We could not compare our data with [@Irwin84] because they have not published their observations in tabular form.
\[profconf\]
Density profile for stars above the TO
--------------------------------------
Figure \[profile\] shows the radial density profile for the TO plus SGB stars extracted from the CMD of M55 (from 1 magnitude below the TO to the brightest limit of our photometry). We have selected the stars within $2.5\sigma$ from the fiducial | and/or $ Z_{GC}$ show a MF significantly flatter than the cluster in the out Galactic aura or farther from the Galactic plane. Indeed, M55 is near to the Galactic bulge, $ R_{GC}=4.7 $ kpc ($ R_\odot=8.0 $ kpc), and to the Galactic magnetic disk $ Z_{GC}=-2.0 $ kpc. Figure \[m55xrz\ ] shows that take into account observing mistake, M55 fairly fits into the relation back given by [ @Manu97 ], which is a refined interpretation of the one found by [ @DPC93 ]. A different conclusion has been reached by [ @Mand96 ] using their uncorrected (for aggregate segregation) value for the MF of M55. As noted by the referee, M55 lies far from the average relation specify by the other bunch: of those with a similar abscissa ($ 0.0\pm0.2 $), M55 is the one with the low value of $ x$. It is not possible to identify the main informant of this apparent enhanced mass - loss of M55 compared to the other bunch; a potential campaign can be a orbit of the cluster that deeply penetrate into the bulge of the Galaxy. This cannot be confirmed until is performed a dependable bill of the proper motion of M55.
Radial density visibility from star counts.\[M55rprof\ ]
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
The CMD allow a unique way to obtain a dependable measure of the radial density profiles of GCs. In fact, the CMD admit us to sort out the stars belonging to the cluster, limiting the problems render by the presence of the field stars. This also allow to extract radial profile for distinct stellar masses.
We have first created a visibility as in [ @King68 ], in order to compare our results with the existing data in the literature. The comparison has been done with the radial density profile of M55 published by [ @Pryor91 ] which includes the ocular star count of King et al.. We could not compare our data with [ @Irwin84 ] because they have not published their observation in tabular form.
\[profconf\ ]
Density visibility for stars above the TO
--------------------------------------
Figure \[profile\ ] shows the radial density visibility for the TO plus SGB stars extracted from the CMD of M55 (from 1 magnitude below the TO to the bright limit of our photometry). We have selected the stars within $ 2.5\sigma$ from the fiducial | anf/or $Z_{GC}$ show a MF signiflcantly flatter jhqn the clustsr in thd outer Galactic halo or farvher from the Galactic plane. Inaeed, M55 is near to the Talactic bnmge, $R_{GC}=4.7$ kpc ($R_\oslt=8.0$ kpe), end to the Galagtic disk $Z_{CC}=-2.0$ kpc. Figure \[m55xsz\] skows that taking into account observyng errprd, M55 fairly fitf inuo ehe dvlction given by [@Manu97], which is a defined version of tne one found by [@DPC93]. A difffrenh conclusion has bfen reached by [@Iqnd96] using thdir uncorrtcced (for masa segregation) value for the MF uf M55. Cs noted by tye ggferee, M55 liew furnher from the average relatipn defined by bhe ovher clusters: of those wivh a similar abscissw ($0.0\pm0.2$), M55 is tke one with the lowesr calue of $f$. It us vot ppsaible ho mdentify ths main sourxe of this apparent egyanced mass-loas of I55 sompared to the other clusters; a possibne dause can be a orbit of the cluster that deekly penetrwte into the bulge of the Galaxy. This cannot be cmnfirjdd bkbil kw oerformed a reliable measure of the proper moejom pf M55.
Radial denfity profilr vrpi star counts.\[O55rprof\]
====================================================
Cge CMD allows a uniqke way jo obtqin a relyablr measure of the radial denwity profilef of GCs. In fact, thz CMD allows us jo sory out the stars belongiug to fhe cluster, limiting ghe problems gendraned ty the presence of the fiejd stars. Vhis clso peroits to exeract radiwl profiles for distinct stelpas masses.
We have first created a profile as in [@King68], in otdes tm comparz our vesults with thq existing datc in the litefature. The comparmson has beeg done with tvg radial densmty profije od M55 publishda by [@Pryor91] whivh includvs the visual star countx ow King et al.. We euood not compare ouf dwtw xith [@Yswin84] because theh hxfe nog published gheit observations in tatulad form.
\[profconf\]
Densotn profile for staws above the YO
--------------------------------------
Figure \[profile\] shlws tie radmal demsijy profile for the TO plus SGB atars extgacbed from the SMD if M55 (from 1 mcgnitude below the TO to the brightest lmmit of our photometry). Qe have selected thg suars within $2.5\smgma$ fwom the fhducial | and/or $Z_{GC}$ show a MF significantly flatter cluster the outer halo or farther M55 near to the bulge, $R_{GC}=4.7$ kpc kpc), and to the Galactic disk kpc. Figure \[m55xrz\] shows that taking into account observing errors, M55 fairly fits the relation given by [@Manu97], which is a refined version of the one by A conclusion been reached by [@Mand96] using their uncorrected (for mass segregation) value for the MF of M55. noted by the referee, M55 lies further from average relation defined by other clusters: of those with similar ($0.0\pm0.2$), M55 the with lowest value of It is not possible to identify the main source of this apparent enhanced mass-loss of M55 compared the other possible cause be orbit the cluster that into the bulge of the Galaxy. confirmed until is performed a reliable measure of proper motion M55. Radial density profile from star ==================================================== The CMD allows a unique way to a reliable measure of the radial density profiles of GCs. In fact, the CMD allows sort out the stars to the cluster, the generated the of the stars. This also permits to extract radial profiles for distinct stellar We have first created a profile as in [@King68], in compare results with the data in the literature. comparison been done with the profile M55 which the star counts of King al.. We could not compare data with [@Irwin84] because observations in tabular form. \[profconf\] Density profile for above the TO -------------------------------------- Figure \[profile\] shows radial density profile for the TO plus SGB stars extracted from the of M55 magnitude below the TO to the brightest limit our photometry). We have the stars within $2.5\sigma$ from the fiducial | and/or $Z_{GC}$ show a MF significanTly flatter Than tHe cLusTeR in tHe ouTer Galactic halO Or faRther from the Galactic plAne. InDeED, M55 is NEaR to thE GalactIC bULGe, $R_{gC}=4.7$ KpC ($R_\oDoT=8.0$ KpC), and tO thE GalactIc disk $Z_{GC}=-2.0$ kPc. FIgUre \[m55xrz\] shows THaT taking intO acCount observiNg eRrors, M55 FaIrlY Fits iNto The reLation GIven by [@manu97], which Is A RefineD Version OF ThE one Found by [@DPC93]. A differENt COnclusion has beEn reacHeD By [@mANd96] uSinG their uncoRrEcted (FOr mass sEGrEGATioN) Value for the MF Of M55. As noted bY The RefereE, M55 LieS FurtheR from ThE AveRage relatioN defIned by the Other cLUsters: oF Those wiTh a simIlaR abScisSA ($0.0\pM0.2$), M55 Is tHe ONe wITh The LOweSt value oF $x$. it Is not PossIBLE To idEntIfy tHe maiN source of this AppArenT EnhAnced Mass-lOss oF M55 CompaRed to tHe othEr Clusters; a possibLe caUse can be a OrbIt Of tHe ClustER that dEepLy pEnetratE into thE BulGe OF THe galaxy. This cannot be CoNFIrMed until Is perfORmEd A Reliable MeAsuRe of THE propEr moTIoN of M55.
RadiAl densITy PrOfile frOm Star coUnTs.\[M55RprOf\]
====================================================
The cmD alLows a uNique way To obtAIn a reliable meaSUre of the radiaL DeNSItY ProfIleS of GCs. In facT, the cmD alLows US tO soRT out tHe staRs BElONging to the cluster, liMiTing thE probLems generated By the preseNCE Of the fieLd stARs. tHis also permits To extRact radial PRofiles fOr disTinct steLlar masseS.
wE have firSt cReaTed A prOFIlE as in [@King68], in orDER to cOmPare our ResUlts witH thE exIstIng DaTa in the liTerature. thE cOmPaRisOn has BEen done wItH thE rAdiAl denSIty proFile oF M55 puBlIsHEd bY [@Pryor91] wHIcH INcluDeS tHe viSuaL sTar coUnts OF KiNg et al.. WE could not ComPAre oUr DaTa with [@IRwin84] because thEy Have not pubLiSheD their OBServatioNs in tabular form.
\[profconf\]
dEnsity pRofIle foR staRs above thE TO
--------------------------------------
figure \[ProFIle\] shoWs the rAdial DeNsiTY ProfiLE FoR thE To plus SGB stARS exTractEd From The CMD oF M55 (from 1 magnitude belOW thE TO to the brighTesT limIT Of Our PHoTOmeTrY). we hAVE selected the staRs within $2.5\siGmA$ FrOm the fiducIAl | and/or $Z_{GC}$ show a MF significa ntlyfla tte rthan the cluster in th e out er Galactic halo or fa rther f r om t h eGalac tic pla n e. I nde ed ,M55 i s n ear t o t he Gala ctic bulge , $ R_ {GC}=4.7$ kp c ( $R_\odot=8 .0$ kpc), and t o t he Gal ac tic disk$Z_ {GC}= -2.0$k pc. Fi gure \[m5 5x r z\] sh o ws that t ak inginto account obse r vi n g errors, M55fairly f i ts i nto th e relation g ivenb y [@Man u 97 ] , whi c h is a refine d version o f th e onefo und by [@D PC93] .A di fferent con clus ion has b een re a ched by [@Mand9 6] usi ngthe ir u n co rr ect ed (fo r m ass seg regation )va lue f or t h e M F of M5 5. A s not ed by the ref ere e, M 5 5 l ies f urthe r fr om theaverag e rel at ion defined bytheother clu ste rs : o fthose with a si mil ar absc issa ($ 0 .0\ pm 0 . 2 $) , M55 is the one w it h th e lowest value of $ x $. It is n otposs i b le to ide n ti fy the m ain so u rc eof this a pparen tenh anc ed ma s s-lo ss ofM55 comp aredt o the other cl u sters; a poss i bl e ca u se c anbe a orbitof t h e cl uste r t hat deepl y pen et r at e into the bulge ofth e Gala xy. T his cannot be confirmed u n til is p erfo r me d a reliable me asure of the pr o per moti on of M55.
R adial den s i ty profi lefro m s tar c ou nts.\[M55rpro f \ ]
== == ======= === ======= === === === === == ========= ========
Th eCM D a llows a unique w ayto ob taina relia ble m easu re o f th e radia l d e n sity p ro file s o fGCs.In f a ct, the CM D allowsust o so rt o ut thestars belongi ng to the cl us ter , limi t i ng the p roblems generated by th e presen ceof th e fi eld stars . T his al sop ermits to ex tract r adi a l prof i l es fo rdistinct s t e lla r mas se s.
We have first created a p r ofi le as in [@Ki ng6 8],i n o rde r t o co mp a reo u r results withthe existi ng da ta in thel ite ra ture. T he comp ariso n has be en done w ith the r ad iald e nsi ty profile of M55published by [@ P ry or91] wh ich in cl ude s the visua l st ar co unts o fKing e t al. .We could not compare our data w ith [@ Irwin 84] becausethe y ha ve not pu blis hed theirobs erv ation s i n tabu larf or m.\[pro fcon f \]
Densi t ypro f i le for starsa b o vethe T O
- - ------ ---- ----------------- - --------
Figu re \ [ p rof ile \ ] sh ow s the radial d ens it y profilefo r the TO pl us SGB s ta r s ext racted fromthe CMD o fM 55 (fr om 1 ma gnitude b elo wt he TO t oth e brigh test l imit o f ourp hoto m e try). We have se lecte d the s t ars with in $2.5\s i gma$ from thefiducial | and/or_$Z_{GC}$ show_a MF significantly flatter_than the_cluster_in the_outer_Galactic halo or_farther from the_Galactic plane. Indeed, M55_is near to_the_Galactic bulge, $R_{GC}=4.7$ kpc ($R_\odot=8.0$ kpc), and to the Galactic disk $Z_{GC}=-2.0$ kpc. Figure \[m55xrz\] shows that taking_into_account observing_errors,_M55_fairly fits into the relation_given by [@Manu97], which is_a refined_version of the one found by [@DPC93]. A_different_conclusion has been_reached by [@Mand96] using their uncorrected (for mass segregation)_value for the MF of M55._As noted by_the_referee,_M55 lies further from_the average relation defined by the_other clusters: of those with a_similar abscissa ($0.0\pm0.2$), M55 is the one_with the lowest value of $x$._It is not possible to_identify the_main source of this apparent_enhanced mass-loss of_M55 compared_to the other_clusters; a possible cause can be_a orbit of_the cluster that deeply penetrate into_the_bulge of the_Galaxy._This_cannot be_confirmed until is_performed_a reliable_measure_of the proper motion of M55.
Radial_density_profile from star counts.\[M55rprof\]
====================================================
The CMD allows a_unique way to obtain_a_reliable measure of the_radial density profiles of GCs._In fact, the CMD allows us_to sort_out the_stars belonging to the cluster, limiting the problems generated by the_presence of the field stars. This_also permits to extract_radial profiles_for_distinct stellar masses.
We_have_first created_a profile as in [@King68], in order_to compare_our results with the existing data_in the literature. The_comparison_has been done with the radial_density profile of M55 published by_[@Pryor91] which includes the visual_star_counts_of King et al.. We_could not compare our data with_[@Irwin84] because they_have not published their observations in tabular_form.
\[profconf\]
Density_profile for stars above the TO
--------------------------------------
Figure \[profile\]_shows_the radial density profile for the_TO_plus_SGB stars extracted from the_CMD of M55 (from 1 magnitude_below the TO to the brightest limit of our_photometry). We have_selected the stars within $2.5\sigma$_from_the_fiducial |
\{{\underline\gamma},\gamma_c/(1+\kappa), \gamma_c(1-{\eta})/(2\kappa), 3\gamma_c(1-\eta)/L\right\}
\equiv {\gamma_{\min}},$$ meaning that holds with $k$ replaced by $k+1$ as claimed.
$\gamma_k
\leq \min\left\{1/(1+\kappa), (1-{\eta})/(2\kappa), 3(1-\eta)/L\right\}$. We must consider two subcases. First, suppose that $k\in\Kcal_g$. It follows from the definition of $\rho_k$, Lemma \[lem-difference\], and Lemma \[lem-achievable\] that $$\begin{aligned}
| \rho_k - 1 |
&= \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k)}
\leq \frac{2\kappa \gamma_k^2 \|g_k\|^2}{\min\big\{(1+\kappa)^{-1},\gamma_k\big\} \|g_k\|^2}
= 2\kappa\gamma_k
\leq 1-\eta,\end{aligned}$$ which implies that $\rho_k \geq {\eta}$, i.e., that $k\in\Scal$. Second, suppose that $k\in\Kcal_H$. It follows from Lemma \[lem-difference\] and Lemma \[lem-achievable\] that $$\begin{aligned}
| \rho_k - 1 |
&= \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k)}
\leq \frac{L\gamma_k^3|(\lambda_k)_-|^3}{3\gamma_k^2|(\lambda_k)_-|^3}
= \frac{L\gamma_k}{3}
\leq 1-\eta,\end{aligned}$$ which implies $\rho_k \geq {\eta}$, i.e., that $k\in\Scal$. In either subcase, it follows that $k\in\Scal$. Using $k\in\Scal$, we have from Algorithm \[alg | \{{\underline\gamma},\gamma_c/(1+\kappa), \gamma_c(1-{\eta})/(2\kappa), 3\gamma_c(1-\eta)/L\right\ }
\equiv { \gamma_{\min}},$$ meaning that holds with $ k$ replaced by $ k+1 $ as claimed.
$ \gamma_k
\leq \min\left\{1/(1+\kappa), (1-{\eta})/(2\kappa), 3(1-\eta)/L\right\}$. We must think two subcases. foremost, suppose that $ k\in\Kcal_g$. It follows from the definition of $ \rho_k$, Lemma \[lem - difference\ ], and Lemma \[lem - achievable\ ] that $ $ \begin{aligned }
| \rho_k - 1 |
& = \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k) }
\leq \frac{2\kappa \gamma_k^2 \|g_k\|^2}{\min\big\{(1+\kappa)^{-1},\gamma_k\big\ } \|g_k\|^2 }
= 2\kappa\gamma_k
\leq 1-\eta,\end{aligned}$$ which entail that $ \rho_k \geq { \eta}$, i.e., that $ k\in\Scal$. Second, speculate that $ k\in\Kcal_H$. It follows from Lemma \[lem - difference\ ] and Lemma \[lem - achievable\ ] that $ $ \begin{aligned }
| \rho_k - 1 |
& = \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k) }
\leq \frac{L\gamma_k^3|(\lambda_k)_-|^3}{3\gamma_k^2|(\lambda_k)_-|^3 }
= \frac{L\gamma_k}{3 }
\leq 1-\eta,\end{aligned}$$ which implies $ \rho_k \geq { \eta}$, i.e., that $ k\in\Scal$. In either subcase, it surveil that $ k\in\Scal$. Using $ k\in\Scal$, we have from Algorithm \[alg | \{{\undfrline\gamma},\gamma_c/(1+\kappa), \gxmma_c(1-{\eta})/(2\kappa), 3\gcnma_c(1-\ete)/L\right\}
\squiv {\gaoma_{\min}},$$ meaning that holds wivh $k$ replqced by $k+1$ as claimed.
$\gaoma_k
\leq \mpn\left\{1/(1+\kappa), (1-{\eua})/(2\kappa), 3(1-\eta)/L\right\}$. We must consiscr twm subcases. Firsj, suppose thdt $k\in\Kcal_g$. It fulpows from the definition of $\rho_k$, Leima \[lem-dofverence\], and Leima \[ltm-ashiebable\] that $$\begin{aligned}
| \rho_k - 1 |
&= \frzc{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_l)}
\leq \frac{2\kappa \gamma_k^2 \|g_k\|^2}{\mln\bih\{(1+\kappa)^{-1},\gamma_k\big\} \|g_k\|^2}
= 2\kappa\gamma_j
\leq 1-\wta,\end{alignea}$$ which imklnes that $\rhk_k \geq {\eta}$, i.e., that $k\in\Scal$. Secovd, su'pose that $j\ib\Kcwn_H$. It folloxs froi Lemma \[lem-dingerenca\] and Lrmma \[lem-achievanle\] tiat $$\vegin{aligned}
| \rho_k - 1 |
&= \hrac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_h(s_k)}
\leq \fsae{L\gamma_k^3|(\lambda_k)_-|^3}{3\gamma_k^2|(\oanbda_k)_-|^3}
= \fraw{L\gaona_k}{3}
\ueq 1-\eva,\ehd{aligjed}$$ which impmies $\rho_k \gwq {\eta}$, i.e., that $k\in\Svaj$. In either sugcase, yt follows that $k\in\Scal$. Using $k\in\Scal$, we vavs from Algorithm \[alg | \{{\underline\gamma},\gamma_c/(1+\kappa), \gamma_c(1-{\eta})/(2\kappa), 3\gamma_c(1-\eta)/L\right\} \equiv {\gamma_{\min}},$$ meaning that $k$ by $k+1$ claimed. $\gamma_k \leq consider subcases. First, suppose $k\in\Kcal_g$. It follows the definition of $\rho_k$, Lemma \[lem-difference\], Lemma \[lem-achievable\] that $$\begin{aligned} | \rho_k - 1 | &= \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) m_k(s_k)} \leq \frac{2\kappa \gamma_k^2 \|g_k\|^2}{\min\big\{(1+\kappa)^{-1},\gamma_k\big\} \|g_k\|^2} = 2\kappa\gamma_k \leq 1-\eta,\end{aligned}$$ which implies that \geq i.e., $k\in\Scal$. suppose that $k\in\Kcal_H$. It follows from Lemma \[lem-difference\] and Lemma \[lem-achievable\] that $$\begin{aligned} | \rho_k - | &= \frac{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k)} \leq = \frac{L\gamma_k}{3} \leq 1-\eta,\end{aligned}$$ implies $\rho_k \geq {\eta}$, i.e., $k\in\Scal$. either subcase, follows $k\in\Scal$. $k\in\Scal$, we have Algorithm \[alg | \{{\underline\gamma},\gamma_c/(1+\kappa), \Gamma_c(1-{\eta})/(2\kAppa), 3\gAmmA_c(1-\eTa)/l\rigHt\}
\eqUiv {\gamma_{\min}},$$ meaNIng tHat holds with $k$ replaced bY $k+1$ as cLaIMed.
$\gAMmA_k
\leq \Min\left\{1/(1+\KApPA), (1-{\Eta})/(2\KaPpA), 3(1-\etA)/L\RIgHt\}$. We mUst ConsideR two subcasEs. FIrSt, suppose thaT $K\iN\Kcal_g$. It foLloWs from the defIniTion of $\RhO_k$, LEMma \[leM-diFfereNce\], and lEmma \[leM-achievabLe\] THat $$\begIN{aligneD}
| \RHo_K - 1 |
&= \fraC{|\,f(x_k+s_k) - m_k(s_k)|}{m_k(0) - m_k(s_k)}
\lEQ \fRAc{2\kappa \gamma_k^2 \|g_K\|^2}{\min\biG\{(1+\kAPpA)^{-1},\GAmmA_k\bIg\} \|g_k\|^2}
= 2\kappa\gAmMa_k
\leQ 1-\Eta,\end{aLIgNED}$$ WhiCH implies that $\rHo_k \geq {\eta}$, i.e., THat $K\in\ScaL$. SEcoND, suppoSe thaT $k\IN\KcAl_H$. It followS froM Lemma \[lem-DifferENce\] and LEMma \[lem-aChievaBle\] ThaT $$\begIN{aLiGneD}
| \rHO_k - 1 |
&= \fRAc{|\,F(x_k+S_K) - m_k(S_k)|}{m_k(0) - m_k(s_k)}
\LeQ \fRac{L\gAmma_K^3|(\LAMBda_k)_-|^3}{3\GamMa_k^2|(\lAmbda_K)_-|^3}
= \frac{L\gamma_k}{3}
\lEq 1-\eTa,\enD{AliGned}$$ wHich iMpliEs $\Rho_k \gEq {\eta}$, i.E., that $K\iN\Scal$. In either suBcasE, it followS thAt $K\in\scAl$. UsiNG $k\in\ScAl$, wE haVe from ALgorithM \[Alg | \{{\underline\gamma},\gamm a_c/(1+\ka ppa), \g amm a_ c(1- {\et a})/(2\kappa), 3\ga mma_c(1-\eta)/L\right\ }
\eq ui v {\g a mm a_{\m in}},$$ me a n ing t ha t ho l ds with $k $ repla ced by $k+ 1$as claimed.
$ \ ga mma_k
\leq \m in\left\{1/( 1+\ kappa) ,(1- { \eta} )/( 2\kap pa), 3 ( 1-\eta )/L\right \} $ . We m u st cons i d er two subcases. First, su p pose that $k\i n\Kcal _g $ .I t fo llo ws from th edefin i tion of $\ r h o _k$ , Lemma \[lem- difference\ ] , a nd Lem ma \[ l em-ach ievab le \ ] t hat $$\begi n{al igned}
|\rho_k - 1 |
& = \frac{ |\,f(x _k+ s_k ) -m _k (s _k) |} { m_k ( 0) -m _k( s_k)}
\ le q\frac {2\k a p p a \ga mma _k^2 \|g_ k\|^2}{\min\b ig\ {(1+ \ kap pa)^{ -1},\ gamm a_ k\big \} \|g _k\|^ 2}
= 2\kappa\gamm a_k\leq 1-\e ta, \e nd{ al igned } $$ whi chimp lies th at $\rh o _k\g e q {\ eta}$, i.e., that$k \ i n\ Scal$. S econd, su pp o se that$k \in \Kca l _ H$. I t fo l lo ws fromLemma\ [l em -differ en ce\] a nd Le mma \[le m -ach ievabl e\] that $$\b e gin{aligned}
| \rho_k - 1 |& =\ f ra c {|\, f(x _k+s_k) - m _k(s _ k)|} {m_k ( 0) -m _k(s_ k)} \l e q\ frac{L\gamma_k^3|(\ la mbda_k )_-|^ 3}{3\gamma_k^ 2|(\lambda _ k ) _-|^3}
= \fr a c{ L \gamma_k}{3}
\ leq 1 -\eta,\end { aligned} $$ wh ich impl ies $\rho _ k \geq {\ eta }$, i. e., t ha t $k\in\Scal$ . In e it her sub cas e, it f oll ows th at$k \in\Scal$ . Using$k \i n\ Sc al$ , weh ave from A lgo ri thm \[al g | \{{\underline\gamma},\gamma_c/(1+\kappa), \gamma_c(1-{\eta})/(2\kappa),_3\gamma_c(1-\eta)/L\right\}
\equiv {\gamma_{\min}},$$_meaning that holds with_$k$ replaced_by_$k+1$ as_claimed.
$\gamma_k
\leq_\min\left\{1/(1+\kappa), (1-{\eta})/(2\kappa), 3(1-\eta)/L\right\}$._We must consider_two subcases. First, suppose_that $k\in\Kcal_g$. It_follows_from the definition of $\rho_k$, Lemma \[lem-difference\], and Lemma \[lem-achievable\] that $$\begin{aligned}
| \rho_k - 1 |
&=_\frac{|\,f(x_k+s_k)_- m_k(s_k)|}{m_k(0)_-_m_k(s_k)}_
\leq \frac{2\kappa \gamma_k^2 \|g_k\|^2}{\min\big\{(1+\kappa)^{-1},\gamma_k\big\} \|g_k\|^2}
=_2\kappa\gamma_k
\leq 1-\eta,\end{aligned}$$ which implies that_$\rho_k \geq_{\eta}$, i.e., that $k\in\Scal$. Second, suppose that $k\in\Kcal_H$._It_follows from Lemma \[lem-difference\]_and Lemma \[lem-achievable\] that $$\begin{aligned}
| \rho_k - 1 |
&= \frac{|\,f(x_k+s_k)_- m_k(s_k)|}{m_k(0) - m_k(s_k)}
\leq \frac{L\gamma_k^3|(\lambda_k)_-|^3}{3\gamma_k^2|(\lambda_k)_-|^3}
=_\frac{L\gamma_k}{3}
\leq 1-\eta,\end{aligned}$$ which_implies_$\rho_k_\geq {\eta}$, i.e., that_$k\in\Scal$. In either subcase, it follows_that $k\in\Scal$. Using $k\in\Scal$, we have_from Algorithm \[alg |
m(z_{i},\beta_{0},\gamma)] & =-E[\lambda_{0}(x_{i})\rho(z_{i},\gamma)]=E[\lambda_{0}(x_{i})\Pi(\rho(\gamma))(x_{i})]\\
& =E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma))(x_{i})-\Pi(\rho(\gamma_{0}))(x_{i})\}]\\
& =E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i})\}].\end{aligned}$$ Here we see that $E[m(z_{i},\beta_{0},\gamma)]$ is a linear, mean-square continuous function of $\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i}).$ The Riesz representation theorem will also imply that if $E[m(z_{i},\beta_{0},\gamma)]$ is a linear, mean-square continuous function of $\Pi(\rho(\gamma)-\rho
(\gamma_{0}))(x_{i})$ then $\lambda_{0}(x)$ exists satisfying Assumption C1 ii). For the case where $w_{i}=x_{i}$ this mean-square continuity condition is necessary for existence of a root-n consistent estimator, as in Newey (1994) and Newey and McFadden (1994). We conjecture that when $w_{i}$ need not equal $x_{i}$ this condition generalizes Severini and Tripathi’s (2012) necessary condition for existence of a root-n consistent estimator of $\beta_{0}$.
Noting that Assumptions 1 ii) and iii) are the conditions for double robustness we have
<span style="font-variant:small-caps;">Theorem C1:</span> *If Assumption C1 is satisfied then* $\psi
(z,\beta,\gamma,\lambda)=m(z,\beta,\gamma)+\lambda(x)\rho(z,\gamma)$ *is doubly robust.*
It is interesting to note that $\lambda_{0}(x)$ satisfying Assumption C1 iii) need not be unique. When the closure of $\{\Pi(\rho(\gamma))(x_{i}):\gamma
\in\Gamma\}$ is not all of $\Lambda$ then there will exist $\tilde{\lambda}\in\Lambda$ | m(z_{i},\beta_{0},\gamma) ] & = -E[\lambda_{0}(x_{i})\rho(z_{i},\gamma)]=E[\lambda_{0}(x_{i})\Pi(\rho(\gamma))(x_{i})]\\
& = E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma))(x_{i})-\Pi(\rho(\gamma_{0}))(x_{i})\}]\\
& = E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i})\}].\end{aligned}$$ Here we see that $ E[m(z_{i},\beta_{0},\gamma)]$ is a linear, mean - square continuous affair of $ \Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i}).$ The Riesz theatrical performance theorem will also imply that if $ E[m(z_{i},\beta_{0},\gamma)]$ is a analogue, mean - hearty continuous function of $ \Pi(\rho(\gamma)-\rho
(\gamma_{0}))(x_{i})$ then $ \lambda_{0}(x)$ exists satisfy Assumption C1 ii). For the case where $ w_{i}=x_{i}$ this mean - straight continuity condition is necessary for existence of a etymon - n consistent estimator, as in Newey (1994) and Newey and McFadden (1994). We speculate that when $ w_{i}$ need not equal $ x_{i}$ this condition generalizes Severini and Tripathi ’s (2012) necessary condition for existence of a root - n consistent calculator of $ \beta_{0}$.
Noting that Assumptions 1 ii) and iii) are the conditions for double robustness we own
< span style="font - variant: small - caps;">Theorem C1:</span > * If Assumption C1 is quenched then * $ \psi
(z,\beta,\gamma,\lambda)=m(z,\beta,\gamma)+\lambda(x)\rho(z,\gamma)$ * is doubly robust. *
It is interesting to notice that $ \lambda_{0}(x)$ satisfying Assumption C1 iii) need not be unique. When the closure of $ \{\Pi(\rho(\gamma))(x_{i}):\gamma
\in\Gamma\}$ is not all of $ \Lambda$ then there will exist $ \tilde{\lambda}\in\Lambda$ | m(z_{i},\heta_{0},\gamma)] & =-E[\lambda_{0}(x_{i})\rhu(z_{i},\gamma)]=E[\lambda_{0}(r_{u})\Pi(\rho(\jamma))(x_{i})]\\
& =E[\lambdx_{0}(x_{i})\{\Pi(\rho(\gamma))(x_{i})-\Pi(\rho(\gamma_{0}))(x_{i})\}]\\
& =E[\oambdq_{0}(x_{i})\{\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{k})\}].\end{alignvd}$$ Here ww set that $E[m(z_{i},\beta_{0},\gamma)]$ is a linead, meau-swuare continuoos function mf $\Pi(\rho(\gamma)-\rvo(\ealma_{0}))(x_{i}).$ The Riesz representation theowem wilk wlso imply thaj if $T[m(z_{y},\betz_{0},\gamma)]$ is a linear, mean-square contjnuous hunction of $\Pi(\rno(\gamma)-\rho
(\gamma_{0}))(x_{i})$ then $\lamhda_{0}(x)$ exists satisfying Assumption C1 iy). For the casd where $w_{i}=q_{n}$ this mean-aquare continuity condition is vecesxary for eziwtejwe of a roov-n confistent estimator, as in Newry (1994) and Newey snd McDadden (1994). We conjecture that when $w_{i}$ need not equan $r_{i}$ this condition genwrqlizev Seeerivu avd Uri'atgi’s (2012) nfceasary condjtion for ezistence of a root-n cjbsistent estijator jf $\beta_{0}$.
Noting that Assumptions 1 ii) and iip) ars the conditions for doyble robustness we hage
<span strle="font-variant:small-caps;">Theorem C1:</span> *If Assumptiot C1 ia sauiwfied rhfn* $\psi
(z,\beta,\gamma,\lambda)=m(z,\beta,\gamma)+\lambda(x)\rho(z,\gwjms)$ *ps doubly robust.*
Ib is interesting tp joyg that $\lambda_{0}(x)$ satisygihg Assumption C1 iil) need got bw unique. Rhen the closure of $\{\Pi(\rho(\gamma))(x_{u}):\gamma
\in\Gammc\}$ iw not all of $\Lambdc$ then there wilk exixt $\tilde{\lambda}\in\Lambda$ | m(z_{i},\beta_{0},\gamma)] & =-E[\lambda_{0}(x_{i})\rho(z_{i},\gamma)]=E[\lambda_{0}(x_{i})\Pi(\rho(\gamma))(x_{i})]\\ & =E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma))(x_{i})-\Pi(\rho(\gamma_{0}))(x_{i})\}]\\ & =E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i})\}].\end{aligned}$$ see $E[m(z_{i},\beta_{0},\gamma)]$ is linear, mean-square continuous representation will also imply if $E[m(z_{i},\beta_{0},\gamma)]$ is linear, mean-square continuous function of $\Pi(\rho(\gamma)-\rho then $\lambda_{0}(x)$ exists satisfying Assumption C1 ii). For the case where $w_{i}=x_{i}$ this continuity condition is necessary for existence of a root-n consistent estimator, as in (1994) Newey McFadden We conjecture that when $w_{i}$ need not equal $x_{i}$ this condition generalizes Severini and Tripathi’s (2012) condition for existence of a root-n consistent estimator $\beta_{0}$. Noting that Assumptions ii) and iii) are the for robustness we <span C1:</span> Assumption C1 is then* $\psi (z,\beta,\gamma,\lambda)=m(z,\beta,\gamma)+\lambda(x)\rho(z,\gamma)$ *is doubly robust.* It is interesting to note that $\lambda_{0}(x)$ satisfying Assumption C1 iii) not be the closure $\{\Pi(\rho(\gamma))(x_{i}):\gamma is all of $\Lambda$ will exist $\tilde{\lambda}\in\Lambda$ | m(z_{i},\beta_{0},\gamma)] & =-E[\lambda_{0}(x_{i})\rho(z_{I},\gamma)]=E[\lamBda_{0}(x_{i})\pi(\rHo(\gAmMa))(x_{i})]\\
& =e[\lamBda_{0}(x_{i})\{\Pi(\rho(\gammA))(X_{i})-\Pi(\Rho(\gamma_{0}))(x_{i})\}]\\
& =E[\lambda_{0}(x_{i})\{\Pi(\rHo(\gamMa)-\RHo(\gaMMa_{0}))(X_{i})\}].\end{Aligned}$$ hErE WE seE tHaT $E[m(Z_{i},\BEtA_{0},\gammA)]$ is A linear, Mean-square ConTiNuous functioN Of $\pi(\rho(\gamma)-\Rho(\Gamma_{0}))(x_{i}).$ The RiEsz RepresEnTatIOn theOreM will Also imPLy that If $E[m(z_{i},\betA_{0},\gAMma)]$ is a LInear, meAN-SqUare Continuous functioN Of $\pI(\rho(\gamma)-\rho
(\gaMma_{0}))(x_{i})$ tHeN $\LaMBDa_{0}(x)$ ExiSts satisfyInG AssuMPtion C1 iI). foR THE caSE where $w_{i}=x_{i}$ thiS mean-square COntInuity CoNdiTIon is nEcessArY For Existence of A rooT-n consistEnt estIMator, as IN Newey (1994) aNd NeweY anD McfaddEN (1994). WE cOnjEcTUre THaT whEN $w_{i}$ Need not eQuAl $X_{i}$ thiS conDITIOn geNerAlizEs SevErini and TripaThi’S (2012) necESsaRy conDitioN for ExIstenCe of a rOot-n cOnSistent estimatoR of $\bEta_{0}$.
Noting ThaT ASsuMpTions 1 II) and iiI) arE thE conditIons for DOubLe ROBUsTness we have
<span styLe="FONt-Variant:sMall-caPS;">THeORem C1:</span> *if assUmptION C1 is sAtisFIeD then* $\psi
(Z,\beta,\gAMmA,\lAmbda)=m(z,\BeTa,\gammA)+\lAmbDa(x)\Rho(z,\gAMma)$ *iS doublY robust.*
IT is inTEresting to note THat $\lambda_{0}(x)$ satISfYINg aSsumPtiOn C1 iii) need nOt be UNiquE. WheN ThE clOSure oF $\{\Pi(\rhO(\gAMmA))(X_{i}):\gamma
\in\Gamma\}$ is not AlL of $\LamBda$ thEn there will exIst $\tilde{\laMBDA}\in\LambdA$ | m(z_{i},\beta_{0},\gamma)] & =-E[\ lambd a_{ 0}( x_ {i}) \rho (z_{i},\gamma) ] =E[\ lambda_{0}(x_{i})\Pi(\ rho(\ ga m ma)) ( x_ {i})] \\
& = E [\ l a mbd a_ {0 }(x _{ i }) \{\Pi (\r ho(\gam ma))(x_{i} )-\ Pi (\rho(\gamma _ {0 }))(x_{i}) \}] \\
& =E[\la mbd a_{0}( x_ {i} ) \{\Pi (\r ho(\g amma)- \ rho(\g amma_{0}) )( x _{i})\ } ].\end{ a l ig ned} $$ Here we see th a t$ E[m(z_{i},\bet a_{0}, \g a mm a ) ]$isa linear,me an-sq u are con t in u o u s f u nction of $\P i(\rho(\gam m a)- \rho(\ ga mma _ {0}))( x_{i} ). $ Th e Riesz rep rese ntation t heorem will al s o imply thatif$E[ m(z_ { i} ,\ bet a_ { 0}, \ ga mma ) ]$is a lin ea r, mean -squ a r e cont inu ousfunct ion of $\Pi(\ rho (\ga m ma) -\rho
(\ga mma_ {0 }))(x _{i})$ then $ \lambda_{0}(x)$ exi sts satis fyi ng As su mptio n C1 ii ).For the ca se wher e $w _{ i } = x_ {i}$ this mean-squ ar e co ntinuity condi t io ni s necess ar y f or e x i stenc e of aroot-n c onsist e nt e stimato r, as in N ewe y ( 1994) andNeweyand McFa dden( 1994). We conj e cture that wh e n$ w _{ i }$ n eed not equal$x_{ i }$ t hisc on dit i on ge neral iz e sS everini and Tripath i’ s (201 2) ne cessary condi tion for e x i s tence of a r o ot - n consistent e stima tor of $\b e ta_{0}$.
Not ing that Assumpti o n s 1 ii)and ii i)are t he conditions f o r dou bl e robus tne ss we h ave
< spa n s ty le="font- variant: sm al l- ca ps; ">The o rem C1:< /s pan >*If Assu m ptionC1 is sat is fi e d t hen* $\ p si ( z,\b et a, \gam ma, \l ambda )=m( z ,\b eta,\ga mma)+\lam bda ( x)\r ho (z ,\gamma )$ *is doubly r obust.*
I tisintere s t ing to n ote that $\lambda_{0}(x ) $ satis fyi ng As sump tion C1 i ii) neednot be uni que. W hen t he cl o s ure o f $\ {\P i( \rho(\gamm a ) )(x _{i}) :\ gamm a
\in\G amma\}$ is not all of$\Lambda$ the n t here w il l e x is t $\ ti l de{ \ l ambda}\in\Lambd a$ | m(z_{i},\beta_{0},\gamma)] _& _=-E[\lambda_{0}(x_{i})\rho(z_{i},\gamma)]=E[\lambda_{0}(x_{i})\Pi(\rho(\gamma))(x_{i})]\\
& =E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma))(x_{i})-\Pi(\rho(\gamma_{0}))(x_{i})\}]\\
& _=E[\lambda_{0}(x_{i})\{\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i})\}].\end{aligned}$$ Here_we_see that_$E[m(z_{i},\beta_{0},\gamma)]$_is a linear,_mean-square continuous function_of $\Pi(\rho(\gamma)-\rho(\gamma_{0}))(x_{i}).$ The Riesz_representation theorem will_also_imply that if $E[m(z_{i},\beta_{0},\gamma)]$ is a linear, mean-square continuous function of $\Pi(\rho(\gamma)-\rho
(\gamma_{0}))(x_{i})$ then $\lambda_{0}(x)$_exists_satisfying Assumption_C1_ii)._For the case where $w_{i}=x_{i}$_this mean-square continuity condition is_necessary for_existence of a root-n consistent estimator, as in_Newey_(1994) and Newey_and McFadden (1994). We conjecture that when $w_{i}$ need_not equal $x_{i}$ this condition generalizes_Severini and Tripathi’s_(2012)_necessary_condition for existence of_a root-n consistent estimator of $\beta_{0}$.
Noting_that Assumptions 1 ii) and iii)_are the conditions for double robustness we_have
<span style="font-variant:small-caps;">Theorem C1:</span> *If Assumption C1_is satisfied then* $\psi
(z,\beta,\gamma,\lambda)=m(z,\beta,\gamma)+\lambda(x)\rho(z,\gamma)$ *is_doubly robust.*
It_is interesting to note that_$\lambda_{0}(x)$ satisfying Assumption_C1 iii)_need not be_unique. When the closure of $\{\Pi(\rho(\gamma))(x_{i}):\gamma
\in\Gamma\}$_is not all_of $\Lambda$ then there will exist_$\tilde{\lambda}\in\Lambda$ |
1-BB^*)^{1/2}(1+AB^*)^{-1}V(1+B^*A)^{-1}(1-B^*B)^{1/2}.\end{split}$$
\[alpha\] $\alpha(A,V) = \|(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}\|$ for all $A\in {\mathcal{B}}$ and $V\in L(K,H)$.
By [@Vesent Lemma V.1.5] $$\alpha(0,V)= \|V\|.$$ Let now $A$ be arbitrary. Then by [@Vesent Proposition V.1.2] $$\alpha(M_A(0),\mathcal D M_A(0)X)=\alpha(0,X)=||X||.$$ On the other hand, by Lemma \[differential\], $$\alpha(M_A(0),\mathcal D M_A(0)X)=\alpha(A,
(1-AA^*)^{1/2}X(1-A^*A)^{1/2}).$$ Setting now $V=(1-AA^*)^{1/2}X(1-A^*A)^{1/2}$, we obtain $X=(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}$ and hence $\alpha(A,V)=||(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2})||$.
For any bounded operator $D$, set $D^{(1)} = D$, $D^{(3)} = DD^*D$, $D^{(5)} = DD^*DD^*D,\ldots,$ $D^{(2k+1)} = (DD^*)^kD$.
Let $$\label{Th}
\text{Th }D = \sum_{n=0}^{\infty}a_{2n+1}D^{(2n+1)}$$ where $a_j$ are the Taylor coefficients of $\tanh t$, i.e., $\tanh
t = \sum_{n=0}^{\infty}a_{2n+1}t^{2n+1}$.
It follows from the definition that $\text{Th }D = \tanh(D)$ if $D$ is selfadjoint.
If $D = J|D|$ is the polar decomposition of $D$ | 1 - BB^*)^{1/2}(1+AB^*)^{-1}V(1+B^*A)^{-1}(1 - B^*B)^{1/2}.\end{split}$$
\[alpha\ ] $ \alpha(A, V) = \|(1 - AA^*)^{-1/2}V(1 - A^*A)^{-1/2}\|$ for all $ A\in { \mathcal{B}}$ and $ V\in L(K, H)$.
By [ @Vesent Lemma V.1.5 ] $ $ \alpha(0,V)= \|V\|.$$ Let now $ A$ be arbitrary. Then by [ @Vesent Proposition V.1.2 ] $ $ \alpha(M_A(0),\mathcal D M_A(0)X)=\alpha(0,X)=||X||.$$ On the other bridge player, by Lemma \[differential\ ], $ $ \alpha(M_A(0),\mathcal five hundred M_A(0)X)=\alpha(A,
(1 - AA^*)^{1/2}X(1 - A^*A)^{1/2}).$$ Setting now $ V=(1 - AA^*)^{1/2}X(1 - A^*A)^{1/2}$, we receive $ X=(1 - AA^*)^{-1/2}V(1 - A^*A)^{-1/2}$ and therefore $ \alpha(A, V)=||(1 - AA^*)^{-1/2}V(1 - A^*A)^{-1/2})||$.
For any bounded operator $ D$, set $ D^{(1) } = D$, $ D^{(3) } = DD^*D$, $ D^{(5) } = DD^*DD^*D,\ldots,$ $ D^{(2k+1) } = (DD^*)^kD$.
permit $ $ \label{Th }
\text{Th } D = \sum_{n=0}^{\infty}a_{2n+1}D^{(2n+1)}$$ where $ a_j$ are the Taylor coefficients of $ \tanh t$, i.e., $ \tanh
t = \sum_{n=0}^{\infty}a_{2n+1}t^{2n+1}$.
It come from the definition that $ \text{Th } D = \tanh(D)$ if $ D$ is selfadjoint.
If $ D = J|D|$ is the polar decay of $ D$ | 1-BB^*)^{1/2}(1+AH^*)^{-1}V(1+B^*A)^{-1}(1-B^*B)^{1/2}.\end{split}$$
\[alpha\] $\alphx(A,V) = \|(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}\|$ for col $A\in {\mathczl{B}}$ and $X\in L(K,H)$.
By [@Vesent Lemma V.1.5] $$\alpia(0,V)= \|C\|.$$ Let now $A$ be arbitrary. Thdn by [@Vesvnt Propowitiib V.1.2] $$\alpha(M_A(0),\mathcal D M_A(0)X)=\amiha(0,X)=||X||.$$ In the other hsnd, by Lemka \[differential\], $$\aupka(M_A(0),\mathcal D M_A(0)X)=\alpha(A,
(1-AA^*)^{1/2}X(1-A^*A)^{1/2}).$$ Setting now $V=(1-AS^*)^{1/2}X(1-W^*A)^{1/2}$, we obtain $X=(1-WA^*)^{-1/2}V(1-A^*S)^{-1/2}$ and hence $\alpha(A,V)=||(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2})||$.
For any bounded operatmr $D$, set $D^{(1)} = D$, $D^{(3)} = DD^*D$, $D^{(5)} = DD^*DD^*D,\ldots,$ $D^{(2k+1)} = (FD^*)^kD$.
Pet $$\label{Th}
\text{Th }F = \sum_{n=0}^{\infti}z_{2n+1}D^{(2g+1)}$$ where $a_j$ ard the Taylor coefficiehts of $\tanh t$, i.e., $\tanh
t = \sum_{n=0}^{\infgy}a_{2n+1}t^{2u+1}$.
It follows feom jhe definitiin thwt $\text{Th }D = \tanh(D)$ ix $D$ is xelfadjoint.
If $C = O|D|$ iw the polar decomposivion of $D$ | 1-BB^*)^{1/2}(1+AB^*)^{-1}V(1+B^*A)^{-1}(1-B^*B)^{1/2}.\end{split}$$ \[alpha\] $\alpha(A,V) = \|(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}\|$ for all and L(K,H)$. By Lemma V.1.5] $$\alpha(0,V)= arbitrary. by [@Vesent Proposition $$\alpha(M_A(0),\mathcal D M_A(0)X)=\alpha(0,X)=||X||.$$ the other hand, by Lemma \[differential\], D M_A(0)X)=\alpha(A, (1-AA^*)^{1/2}X(1-A^*A)^{1/2}).$$ Setting now $V=(1-AA^*)^{1/2}X(1-A^*A)^{1/2}$, we obtain $X=(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}$ and hence $\alpha(A,V)=||(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2})||$. For bounded operator $D$, set $D^{(1)} = D$, $D^{(3)} = DD^*D$, $D^{(5)} = DD^*DD^*D,\ldots,$ = Let \text{Th = \sum_{n=0}^{\infty}a_{2n+1}D^{(2n+1)}$$ where $a_j$ are the Taylor coefficients of $\tanh t$, i.e., $\tanh t = \sum_{n=0}^{\infty}a_{2n+1}t^{2n+1}$. follows from the definition that $\text{Th }D = if $D$ is selfadjoint. $D = J|D|$ is the decomposition $D$ | 1-BB^*)^{1/2}(1+AB^*)^{-1}V(1+B^*A)^{-1}(1-B^*B)^{1/2}.\end{split}$$
\[alpha\] $\alpHa(A,V) = \|(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}\|$ fOr all $a\in {\MatHcAl{B}}$ aNd $V\iN L(K,H)$.
By [@Vesent LeMMa V.1.5] $$\aLpha(0,V)= \|V\|.$$ Let now $A$ be arbitraRy. TheN bY [@veseNT PRoposItion V.1.2] $$\aLPhA(m_a(0),\maThCaL D M_a(0)X)=\ALpHa(0,X)=||X||.$$ ON thE other hAnd, by Lemma \[DifFeRential\], $$\alpha(m_a(0),\mAthcal D M_A(0)X)=\AlpHa(A,
(1-AA^*)^{1/2}X(1-A^*A)^{1/2}).$$ SettIng Now $V=(1-AA^*)^{1/2}x(1-A^*a)^{1/2}$, we OBtain $x=(1-AA^*)^{-1/2}v(1-A^*A)^{-1/2}$ anD hence $\ALpha(A,V)=||(1-aA^*)^{-1/2}V(1-A^*A)^{-1/2})||$.
For aNy BOunded OPerator $d$, SEt $d^{(1)} = D$, $D^{(3)} = Dd^*D$, $D^{(5)} = DD^*DD^*D,\ldots,$ $D^{(2k+1)} = (DD^*)^Kd$.
LET $$\label{Th}
\text{Th }d = \sum_{n=0}^{\iNfTY}a_{2N+1}d^{(2N+1)}$$ whEre $A_j$ are the TaYlOr coeFFicientS Of $\TANH t$, i.E., $\Tanh
t = \sum_{n=0}^{\inftY}a_{2n+1}t^{2n+1}$.
It follOWs fRom the DeFinITion thAt $\texT{TH }d = \taNh(D)$ if $D$ is selFadjOint.
If $D = J|D|$ Is the pOLar decoMPositioN of $D$ | 1-BB^*)^{1/2}(1+AB^*)^{-1} V(1+B^*A)^ {-1}( 1-B ^*B )^ {1/2 }.\e nd{split}$$
\ [ alph a\] $\alpha(A,V) = \|( 1-AA^ *) ^ {-1/ 2 }V (1-A^ *A)^{-1 / 2} \ | $ f or a ll$A \ in {\ma thc al{B}}$ and $V\in L( K, H)$.
By [@V e se nt Lemma V .1. 5] $$\alpha( 0,V )= \|V \| .$$ Let n ow$A$ b e arbi t rary.Then by [ @V e sent P r opositi o n V .1.2 ] $$\alpha(M_A(0) , \m a thcal D M_A(0) X)=\al ph a (0 , X )=| |X| |.$$ On th eother hand, b y L e m m a \ [ differential\ ], $$\alpha ( M_A (0),\m at hca l D M_A (0)X) =\ a lph a(A,
(1-AA^ *)^{ 1/2}X(1-A ^*A)^{ 1 /2}).$$ Setting now $ V=( 1-A A^*) ^ {1 /2 }X( 1- A ^*A ) ^{ 1/2 } $,we obtai n$X =(1-A A^*) ^ { - 1 /2}V (1- A^*A )^{-1 /2}$ and henc e $ \alp h a(A ,V)=| |(1-A A^*) ^{ -1/2} V(1-A^ *A)^{ -1 /2})||$.
For a ny b ounded op era to r $ D$ , set $D^{(1 )}= D $, $D^{ (3)} =D D^* D$ , $ D^ {(5)} = DD^*DD^*D, \l d o ts ,$ $D^{( 2k+1)} =(D D ^*)^kD$.
Let $$\ l a bel{T h}
\ t ex t{Th }D= \sum _ {n =0 }^{\inf ty }a_{2n +1 }D^ {(2 n+1)} $ $ wh ere $a _j$ arethe T a ylor coefficie n ts of $\tanht $, i .e . , $\ tan h
t = \sum_ {n=0 } ^{\i nfty } a_ {2n + 1}t^{ 2n+1} $.
I t follows from the d ef initio n tha t $\text{Th } D = \tanh( D ) $ if $D$is s e lf a djoint.
If $D = J| D|$ is the polar de compo sition o f $D$ | 1-BB^*)^{1/2}(1+AB^*)^{-1}V(1+B^*A)^{-1}(1-B^*B)^{1/2}.\end{split}$$
\[alpha\] $\alpha(A,V)_= \|(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}\|$_for all $A\in {\mathcal{B}}$_and $V\in_L(K,H)$.
By_[@Vesent Lemma_V.1.5]_$$\alpha(0,V)= \|V\|.$$ Let_now $A$ be_arbitrary. Then by [@Vesent_Proposition V.1.2] $$\alpha(M_A(0),\mathcal D_M_A(0)X)=\alpha(0,X)=||X||.$$_On the other hand, by Lemma \[differential\], $$\alpha(M_A(0),\mathcal D M_A(0)X)=\alpha(A,
(1-AA^*)^{1/2}X(1-A^*A)^{1/2}).$$ Setting now $V=(1-AA^*)^{1/2}X(1-A^*A)^{1/2}$, we obtain_$X=(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2}$_and hence_$\alpha(A,V)=||(1-AA^*)^{-1/2}V(1-A^*A)^{-1/2})||$.
For_any_bounded operator $D$, set $D^{(1)}_= D$, $D^{(3)} = DD^*D$,_$D^{(5)} =_DD^*DD^*D,\ldots,$ $D^{(2k+1)} = (DD^*)^kD$.
Let $$\label{Th}
\text{Th }D = \sum_{n=0}^{\infty}a_{2n+1}D^{(2n+1)}$$_where_$a_j$ are the_Taylor coefficients of $\tanh t$, i.e., $\tanh
t = \sum_{n=0}^{\infty}a_{2n+1}t^{2n+1}$.
It_follows from the definition that $\text{Th_}D = \tanh(D)$_if_$D$_is selfadjoint.
If $D =_J|D|$ is the polar decomposition of_$D$ |
)P_{n_{1}}^{({\alpha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\ell m}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpha},{\beta})}(x)=1.
\label{eq:AlmBlm}\end{aligned}$$ Using the orthogonality condition of the Jacobi polynomials, we have $$\begin{aligned}
&\int_{-1}^{1} d x\left(\frac{1-x}{2}\right)^{{\alpha}}\left(\frac{1+x}{2}\right)^{{\beta}}P_{n_{1}}^{({\alpha},{\beta})}(x)P_{n_{2}}^{({\alpha},{\beta})}(x)\cr
&\quad=\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)\delta_{n_{1},n_{2}}}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)\Gamma(n+{\alpha}+{\beta}+1)}.\end{aligned}$$ Then, Eq. (\[eq:AlmBlm\]) reduces to $$\begin{aligned}
&\sum_{n=0}^{\infty}\left[\frac{\,{}_{s}A_{\ell m}^{(n)}(\xi)}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)}\right]
\left[\frac{\,{}_{s}B_{\ell m}^{(n)}(\xi)}{\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}\right]
\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)
\Gamma(n+{\alpha}+{\beta}+1)}
\cr
&\quad=\frac{1}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}.
\label{eq:normalization2}\end{aligned}$$ From Eqs. (\[eq:normalization1\]) and (\[eq:normalization2\]). we can obtain the squares of ${}_{s}A_{\ell m}^{(n_{\ell})}(\xi)$ and ${}_{s}B_{\ell m}^{(n_{\ell})}(\xi)$. The final determination of the signs of ${}_{s}A_{\ell m}^{(n | ) P_{n_{1}}^{({\alpha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\ell m}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpha},{\beta})}(x)=1.
\label{eq: AlmBlm}\end{aligned}$$ Using the orthogonality condition of the Jacobi polynomials, we have $ $ \begin{aligned }
& \int_{-1}^{1 } vitamin d x\left(\frac{1 - x}{2}\right)^{{\alpha}}\left(\frac{1+x}{2}\right)^{{\beta}}P_{n_{1}}^{({\alpha},{\beta})}(x)P_{n_{2}}^{({\alpha},{\beta})}(x)\cr
& \quad=\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)\delta_{n_{1},n_{2}}}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)\Gamma(n+{\alpha}+{\beta}+1)}.\end{aligned}$$ Then, Eq. (\[eq: AlmBlm\ ]) reduce to $ $ \begin{aligned }
& \sum_{n=0}^{\infty}\left[\frac{\,{}_{s}A_{\ell m}^{(n)}(\xi)}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)}\right ]
\left[\frac{\,{}_{s}B_{\ell m}^{(n)}(\xi)}{\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}\right ]
\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)
\Gamma(n+{\alpha}+{\beta}+1) }
\cr
& \quad=\frac{1}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi) }.
\label{eq: normalization2}\end{aligned}$$ From Eqs. (\[eq: normalization1\ ]) and (\[eq: normalization2\ ]). we can obtain the squares of $ { } _ { s}A_{\ell m}^{(n_{\ell})}(\xi)$ and $ { } _ { s}B_{\ell m}^{(n_{\ell})}(\xi)$. The final decision of the signs of $ { } _ { s}A_{\ell m}^{(n | )P_{n_{1}}^{({\appha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\ell o}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpha},{\beta})}(r)=1.
\oabel{ex:AlmBlj}\end{aligved}$$ Using the orthogonality rondution of the Jacobi polynomkals, we hwve $$\begib{alijned}
&\int_{-1}^{1} d x\left(\fczc{1-x}{2}\righb)^{{\clpha}}\mcft(\frcc{1+e}{2}\right)^{{\beta}}P_{n_{1}}^{({\alpma},{\beta})}(x)P_{n_{2}}^{({\al[ha},{\beta})}(x)\cr
&\quad=\xrxc{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)\delta_{n_{1},n_{2}}}{(2n+{\alpha}+{\feta}+1)\Gamka(j+1)\Gamma(n+{\alpha}+{\beja}+1)}.\end{sjignsd}$$ Then, Eq. (\[eq:AlmBlm\]) reduces to $$\begjn{aligntd}
&\sum_{n=0}^{\infty}\left[\frav{\,{}_{s}A_{\ell m}^{(n)}(\xi)}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)}\rihht]
\lfft[\frac{\,{}_{s}B_{\ell m}^{(n)}(\xi)}{\,{}_{s}H_{\ell m}^{(n_{\ell})}(\xu)}\rigrr]
\frac{2\, \Gamma(n+{\xlpha}+1)\Gamma(n+{\beta}+1)}{(2n+{\alpha}+{\bgta}+1)\Gamma(n+1)
\Gamma(n+{\alpha}+{\beta}+1)}
\cr
&\quad=\frxc{1}{\,{}_{s}A_{\ekl m}^{(n_{\ell})}(\xi)\,{}_{w}B_{\wll k}^{(n_{\ell})}(\xi)}.
\labeo{eq:nogmalization2}\end{aligned}$$ From Eas. (\[eq:normalizatlon1\]) aid (\[ew:normalization2\]). we can obtain the squaref of ${}_{s}A_{\eln j}^{(n_{\ell})}(\xi)$ and ${}_{s}B_{\ell m}^{(b_{\ell})}(\xh)$. Tha fivql aettrmmnafion ov tie signs of ${}_{s}A_{\ell m}^{(n | )P_{n_{1}}^{({\alpha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\ell m}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpha},{\beta})}(x)=1. \label{eq:AlmBlm}\end{aligned}$$ Using the orthogonality condition Jacobi we have &\int_{-1}^{1} d x\left(\frac{1-x}{2}\right)^{{\alpha}}\left(\frac{1+x}{2}\right)^{{\beta}}P_{n_{1}}^{({\alpha},{\beta})}(x)P_{n_{2}}^{({\alpha},{\beta})}(x)\cr reduces $$\begin{aligned} &\sum_{n=0}^{\infty}\left[\frac{\,{}_{s}A_{\ell m}^{(n)}(\xi)}{\,{}_{s}A_{\ell \left[\frac{\,{}_{s}B_{\ell m}^{(n)}(\xi)}{\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}\right] \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1) \Gamma(n+{\alpha}+{\beta}+1)} \cr &\quad=\frac{1}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}. From Eqs. (\[eq:normalization1\]) and (\[eq:normalization2\]). we can obtain the squares of ${}_{s}A_{\ell m}^{(n_{\ell})}(\xi)$ ${}_{s}B_{\ell m}^{(n_{\ell})}(\xi)$. The final determination of the signs of ${}_{s}A_{\ell m}^{(n | )P_{n_{1}}^{({\alpha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\elL m}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpHa},{\betA})}(x)=1.
\lAbeL{eQ:Almblm}\eNd{aligned}$$ Using THe orThogonality condition of The JaCoBI polYNoMials, We have $$\bEGiN{ALigNeD}
&\iNt_{-1}^{1} d X\lEFt(\Frac{1-x}{2}\RigHt)^{{\alpha}}\Left(\frac{1+x}{2}\rIghT)^{{\bEta}}P_{n_{1}}^{({\alpha},{\beTA})}(x)p_{n_{2}}^{({\alpha},{\betA})}(x)\cR
&\quad=\frac{2\, \GamMa(n+{\Alpha}+1)\GAmMa(n+{\BEta}+1)\deLta_{N_{1},n_{2}}}{(2n+{\alPha}+{\betA}+1)\gamma(n+1)\gamma(n+{\alpHa}+{\BEta}+1)}.\end{ALigned}$$ THEN, EQ. (\[eq:ALmBlm\]) reduces to $$\begIN{aLIgned}
&\sum_{n=0}^{\infty}\Left[\frAc{\,{}_{S}a_{\eLL M}^{(n)}(\xI)}{\,{}_{s}A_{\Ell m}^{(n_{\ell})}(\xi)}\RiGht]
\leFT[\frac{\,{}_{s}B_{\ELl M}^{(N)}(\XI)}{\,{}_{s}B_{\ELl m}^{(n_{\ell})}(\xi)}\righT]
\frac{2\, \Gamma(n+{\ALphA}+1)\Gamma(N+{\bEta}+1)}{(2N+{\Alpha}+{\bEta}+1)\GaMmA(N+1)
\GaMma(n+{\alpha}+{\beTa}+1)}
\cr
&\Quad=\frac{1}{\,{}_{s}a_{\ell m}^{(n_{\ELl})}(\xi)\,{}_{s}B_{\eLL m}^{(n_{\ell})}(\xI)}.
\label{Eq:nOrmAlizATiOn2}\End{AlIGneD}$$ frOm EQS. (\[eq:NormalizAtIoN1\]) and (\[eQ:norMALIZatiOn2\]). wE can ObtaiN the squares of ${}_{S}A_{\eLl m}^{(n_{\ELl})}(\xI)$ and ${}_{s}b_{\ell m}^{(N_{\ell})}(\Xi)$. the fiNal detErminAtIon of the signs of ${}_{S}A_{\elL m}^{(n | )P_{n_{1}}^{({\alpha},{\be ta})}(x)\s um_{n _{2 }=0 }^ {\in fty} \,{}_{s}B_{\el l m}^ {(n_{2})}(\xi)P_{n_{2} }^{({ \a l pha} , {\ beta} )}(x)=1 .
\ l a bel {e q: Alm Bl m }\ end{a lig ned}$$Using theort ho gonality con d it ion of the Ja cobi polynom ial s, weha ve$ $\beg in{ align ed}
&\ i nt_{-1 }^{1} d x \l e ft(\fr a c{1-x}{ 2 } \r ight )^{{\alpha}}\left ( \f r ac{1+x}{2}\rig ht)^{{ \b e ta } } P_{ n_{ 1}}^{({\al ph a},{\ b eta})}( x )P _ { n _{2 } }^{({\alpha}, {\beta})}(x ) \cr
&\qua d= \fr a c{2\,\Gamm a( n +{\ alpha}+1)\G amma (n+{\beta }+1)\d e lta_{n_ { 1},n_{2 }}}{(2 n+{ \al pha} + {\ be ta} +1 ) \Ga m ma (n+ 1 )\G amma(n+{ \a lp ha}+{ \bet a } + 1 )}.\ end {ali gned} $$ Then, Eq.(\[ eq:A l mBl m\])reduc es t o$$\be gin{al igned }&\sum_{n=0}^{\i nfty }\left[\f rac {\ ,{} _{ s}A_{ \ ell m} ^{( n)} (\xi)}{ \,{}_{s } A_{ \e l l m} ^{(n_{\ell})}(\xi) }\ r i gh t]
\left [\frac { \, {} _ {s}B_{\e ll m} ^{(n ) } (\xi) }{\, { }_ {s}B_{\e ll m}^ { (n _{ \ell})} (\ xi)}\r ig ht]
\f rac{2 \ , \G amma(n +{\alpha }+1)\ G amma(n+{\beta} + 1)}{(2n+{\alp h a} + { \b e ta}+ 1)\ Gamma(n+1)\Gam m a(n+ {\al p ha }+{ \ beta} +1)}\c r
& \ quad=\frac{1}{\,{}_ {s }A_{\e ll m} ^{(n_{\ell})} (\xi)\,{}_ { s } B_{\ellm}^{ ( n_ { \ell})}(\xi)}.
\lab el{eq:norm a lization 2}\en d{aligne d}$$ From E qs. (\[e q:n orm ali zat i o n1 \]) and (\[eq : n orma li zation2 \]) . we ca n o bta inthe s quares of ${}_{s} A_ {\ el lm}^ {(n_{ \ ell})}(\ xi )$an d $ {}_{s } B_{\el l m}^ {(n_ {\ el l })} (\xi)$. Th e fina lde term ina ti on of the sig ns of $ {}_{s}A_{ \el l m}^ {( n | )P_{n_{1}}^{({\alpha},{\beta})}(x)\sum_{n_{2}=0}^{\infty}\,{}_{s}B_{\ell m}^{(n_{2})}(\xi)P_{n_{2}}^{({\alpha},{\beta})}(x)=1.
\label{eq:AlmBlm}\end{aligned}$$_Using the_orthogonality condition of the_Jacobi polynomials,_we_have $$\begin{aligned}
&\int_{-1}^{1}_d_x\left(\frac{1-x}{2}\right)^{{\alpha}}\left(\frac{1+x}{2}\right)^{{\beta}}P_{n_{1}}^{({\alpha},{\beta})}(x)P_{n_{2}}^{({\alpha},{\beta})}(x)\cr
&\quad=\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)\delta_{n_{1},n_{2}}}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)\Gamma(n+{\alpha}+{\beta}+1)}.\end{aligned}$$ Then,_Eq. (\[eq:AlmBlm\]) reduces_to $$\begin{aligned}
&\sum_{n=0}^{\infty}\left[\frac{\,{}_{s}A_{\ell m}^{(n)}(\xi)}{\,{}_{s}A_{\ell m}^{(n_{\ell})}(\xi)}\right]
\left[\frac{\,{}_{s}B_{\ell_m}^{(n)}(\xi)}{\,{}_{s}B_{\ell m}^{(n_{\ell})}(\xi)}\right]
\frac{2\, \Gamma(n+{\alpha}+1)\Gamma(n+{\beta}+1)}{(2n+{\alpha}+{\beta}+1)\Gamma(n+1)
\Gamma(n+{\alpha}+{\beta}+1)}
\cr
&\quad=\frac{1}{\,{}_{s}A_{\ell_m}^{(n_{\ell})}(\xi)\,{}_{s}B_{\ell_m}^{(n_{\ell})}(\xi)}.
\label{eq:normalization2}\end{aligned}$$ From Eqs. (\[eq:normalization1\]) and (\[eq:normalization2\]). we can obtain the squares of ${}_{s}A_{\ell m}^{(n_{\ell})}(\xi)$ and_${}_{s}B_{\ell_m}^{(n_{\ell})}(\xi)$. The_final_determination_of the signs of ${}_{s}A_{\ell_m}^{(n |
) 463\] \[Nuovo Cim. B [**44**]{} (1966) 1\]. E. Poisson, *A Relativist’s Toolkit*, Cambridge University Press, Cambridge U.K. (2004)
V. Balasubramanian and S. F. Ross, Phys. Rev. D [**61**]{} (2000) 044007 \[hep-th/9906226\]. T. Banks, M. R. Douglas, G. T. Horowitz and E. J. Martinec, hep-th/9808016. V. Keranen and P. Kleinert, JHEP [**1504**]{} (2015) 119 \[arXiv:1412.2806 \[hep-th\]\].
[^1]: Note that there is an ambiguity in the literature related to the sign of $K_{ij}$. For example [@Poisson] defines $K_{ij}$ with an opposite sign. In this case the sign ambguity does not matter, but can cause confusion when the energy-momentum of the joining hypersurface is non-zero.
[^2]: Here we use the term *velocity* to denote a derivative of the coordinates of the geodesic with respect to the proper length of the geodesic. Although these quantities are not velocities in the physical sense, they appear in the geodesic equations in the same fashion as actual velocities would for time-like geodesics.
---
abstract: 'Motion of many comets is affected by non-gravitational forces caused by outgassing from their surfaces. Outgassing also produces reactive torques resulting in cometary spin evolution. We propose that the two processes are correlated and show that the change of cometary spin rate over its heliocentric orbit scales linearly with the amplitude of its non-gravitational acceleration. The proportionality constant depends on the comet size and orbital elements (semi-major axis and eccentricity) and on the (dimensionless) lever arm parameter $\zeta$ that relates the outgassing-induced torque and acceleration. We determine $\zeta$ for 7 comets for which both non-gravitational acceleration and change of spin period $\Delta P$ were measured and verify this relation. This sample spanning almost 4 decades in $\Delta P$ yields $\log\zeta=-2.21\pm 0.54$, surprisingly small value and spread. We then apply our framework to 209 comets with measured non | ) 463\ ] \[Nuovo Cim. B [ * * 44 * * ] { } (1966) 1\ ]. E. Poisson, * A Relativist ’s Toolkit *, Cambridge University Press, Cambridge U.K. (2004)
V. Balasubramanian and S. F. Ross, Phys. Rev. D [ * * 61 * * ] { } (2000) 044007 \[hep - th/9906226\ ]. T. Banks, M. R. Douglas, G. T. Horowitz and E. J. Martinec, hep - th/9808016. V. Keranen and P. Kleinert, JHEP [ * * 1504 * * ] { } (2015) 119 \[arXiv:1412.2806 \[hep - th\]\ ].
[ ^1 ]: Note that there is an ambiguity in the literature related to the sign of $ K_{ij}$. For case [ @Poisson ] define $ K_{ij}$ with an opposite sign. In this case the signboard ambguity does not matter, but can cause confusion when the department of energy - momentum of the joining hypersurface is non - zero.
[ ^2 ]: Here we use the term * speed * to announce a derivative of the coordinates of the geodesic with regard to the proper length of the geodesic. Although these quantities are not velocities in the forcible sense, they appear in the geodetic equations in the like manner as actual velocities would for time - like geodesics.
---
abstract:' apparent motion of many comets is affected by non - gravitational forces cause by outgassing from their surfaces. Outgassing also produces reactive torques resulting in cometary tailspin evolution. We propose that the two processes are correlated and show that the change of cometary tailspin pace over its heliocentric sphere scales linearly with the amplitude of its non - gravitational acceleration. The proportionality constant depends on the comet size and orbital elements (semi - major axis and eccentricity) and on the (dimensionless) lever arm parameter $ \zeta$ that relate the outgassing - induce torque and acceleration. We settle $ \zeta$ for 7 comets for which both non - gravitational acceleration and change of spin period $ \Delta P$ were measure and verify this relation. This sample spanning about 4 decades in $ \Delta P$ yields $ \log\zeta=-2.21\pm 0.54 $, surprisingly small value and spread. We then apply our framework to 209 comets with measured non | ) 463\] \[Nkovo Cim. B [**44**]{} (1966) 1\]. E. Poisson, *A Relativist’s Toolkit*, Cembridgs Univerrity Press, Cambridge U.K. (2004)
V. Balesubeamanuan and S. F. Ross, Phys. Rev. A [**61**]{} (2000) 044007 \[hep-tj/9906226\]. T. Banks, M. R. Diyglas, G. T. Hockwitz akb E. J. Mzvtinee, iep-th/9808016. V. Keranen snd P. Kleinart, JHEP [**1504**]{} (2015) 119 \[arFix:1412.2806 \[kep-th\]\].
[^1]: Note that there is an ambiguitr in thr piterature relwted eo tgv wign of $K_{ij}$. For example [@Poisskn] defiies $K_{ij}$ with an opposite sign. In this casf thf sign ambguity dofs not mattgd, btr can cause zonfusion when the enetgy-momentum of the joining hyperrurfaee is non-zeto.
[^2]: Hfte we use thx term *velocity* to denote d derivstive of the cporvinares of the geodesic wmth respect to the ptoper lengdh of the geodesic. Qlrhougv thase dyangitjex zre noh vxlocities ih the physixal sense, they appesr pm the geodesjc equwtyons in the same fashion as actual velobitiss would for time-like gwodesics.
---
abstract: 'Motlon of magy comets is affected by non-gravitational forces wausev cy iubgasrunh from their surfaces. Outgassing also producef rtacnive torques resujting in cokehati spin evolutiun. We 'dolose that the two orocessgs are correlattd anc show that the change of cimetary spin eate over its helilcentric oryit scsles kinearly with the amplicude or its non-grwvitationzu acceleration. Tfe iropmrtionality constant depenqs on the comzt size xnd prbitaj elements (semi-major axis and eccejtricnty) atd on the (fimensionless) lever arm parametxc $\zeta$ that rglades the outyassinn-induced torque and acceleratnon. We dztermive $\zeta$ fog 7 comets for which foth non-gravidwtional acceneration and chabge of roin period $\Delya P$ were measured qnd verify this rekatkkn. This sample wpabning almost 4 drcaaes ij $\Velta [$ yields $\log\seta=-2.21\om 0.54$, xurprksingly smaou vakue and spread. We than alply our framework tj 209 cometw with mqasured non | ) 463\] \[Nuovo Cim. B [**44**]{} (1966) Poisson, Relativist’s Toolkit*, University Press, Cambridge S. Ross, Phys. Rev. [**61**]{} (2000) 044007 T. Banks, M. R. Douglas, G. Horowitz and E. J. Martinec, hep-th/9808016. V. Keranen and P. Kleinert, JHEP [**1504**]{} 119 \[arXiv:1412.2806 \[hep-th\]\]. [^1]: Note that there is an ambiguity in the literature to sign $K_{ij}$. example [@Poisson] defines $K_{ij}$ with an opposite sign. In this case the sign ambguity does not but can cause confusion when the energy-momentum of joining hypersurface is non-zero. Here we use the term to a derivative the of geodesic with respect the proper length of the geodesic. Although these quantities are not velocities in the physical sense, they in the in the fashion actual would for time-like abstract: 'Motion of many comets is forces caused by outgassing from their surfaces. Outgassing produces reactive resulting in cometary spin evolution. We that the two processes are correlated and show the change of cometary spin rate over its heliocentric orbit scales linearly with the amplitude non-gravitational acceleration. The proportionality depends on the size orbital (semi-major and eccentricity) on the (dimensionless) lever arm parameter $\zeta$ that relates the outgassing-induced and acceleration. We determine $\zeta$ for 7 comets for which acceleration change of spin $\Delta P$ were measured verify relation. This sample spanning decades $\Delta 0.54$, small and spread. We then our framework to 209 comets measured non | ) 463\] \[Nuovo Cim. B [**44**]{} (1966) 1\]. E. Poisson, *A RelativIst’s ToolkiT*, CambRidGe UNiVersIty PRess, Cambridge U.k. (2004)
v. BalAsubramanian and S. F. Ross, PHys. ReV. D [**61**]{} (2000) 044007 \[HEp-th/9906226\]. t. baNks, M. R. douglas, g. t. HOROwiTz AnD E. J. maRTiNec, heP-th/9808016. v. KeraneN and P. KleinErt, jHeP [**1504**]{} (2015) 119 \[arXiv:1412.2806 \[hep-th\]\].
[^1]: nOtE that there Is aN ambiguity in The LiteraTuRe rELated To tHe sigN of $K_{ij}$. fOr examPle [@PoissoN] dEFines $K_{IJ}$ with an OPPoSite Sign. In this case the SIgN Ambguity does noT matteR, bUT cAN CauSe cOnfusion whEn The enERgy-momeNTuM OF The JOining hypersuRface is non-zERo.
[^2]: HEre we uSe The TErm *velOcity* To DEnoTe a derivatiVe of The coordiNates oF The geodESic with RespecT to The PropER lEnGth Of THe gEOdEsiC. altHough theSe QuAntitIes aRE NOT velOciTies In the Physical sense, TheY appEAr iN the gEodesIc eqUaTions In the sAme faShIon as actual veloCitiEs would foR tiMe-LikE gEodesICs.
---
abstRacT: 'MoTion of mAny comeTS is AfFECTeD by non-gravitationaL fORCeS caused bY outgaSSiNg FRom their SuRfaCes. OUTGassiNg alSO pRoduces rEactivE ToRqUes resuLtIng in cOmEtaRy sPin evOLutiOn. We prOpose thaT the tWO processes are cORrelated and shOW tHAT tHE chaNge Of cometary sPin rATe ovEr itS HeLioCEntriC orbiT sCAlES linearly with the ampLiTude of Its noN-gravitationaL acceleratION. the propoRtioNAlITy constant depeNds on The comet siZE and orbiTal elEments (seMi-major axIS And eccenTriCitY) anD on THE (dImensionless) lEVEr arM pArameteR $\zeTa$ that rElaTes The OutGaSsing-induCed torquE aNd AcCeLerAtion. wE determiNe $\ZetA$ fOr 7 cOmets FOr whicH both Non-gRaViTAtiOnal accELeRATion AnD cHangE of SpIn perIod $\DELta p$ were meAsured and VerIFy thIs ReLation. THis sample spanNiNg almost 4 deCaDes In $\DeltA p$ Yields $\loG\zeta=-2.21\pm 0.54$, surprisingly smalL Value anD spRead. WE theN apply our FraMework To 209 cOMets wiTh measUred nOn | ) 463\] \[Nuovo Cim. B [** 44**]{} (1 966)1\] . E .Pois son, *A Relativist ’ s To olkit*, Cambridge Univ ersit yP ress , C ambri dge U.K . ( 2 0 04)
V. Ba la s ub raman ian and S. F. Ross,Phy s. Rev. D [**6 1 ** ]{} (2000) 04 4007 \[hep-t h/9 906226 \] . T . Bank s,M. R. Dougl a s, G.T. Horowi tz and E. J. Mart i n ec , he p-th/9808016. V.K er a nen and P. Kle inert, J H EP [ **1 504 **]{} (201 5) 119\ [arXiv: 1 41 2 . 2 806 \[hep-th\]\].
[^1]: Not e th at the re is an amb iguit yi n t he literatu re r elated to the s i gn of $ K _{ij}$. For e xam ple [@P o is so n]de f ine s $ K_{ i j}$ with an o pp osite sig n . I n th iscase thesign ambguity do es n o t m atter , but can c auseconfus ion w he n the energy-mo ment um of the jo in ing h ypers u rfaceisnon -zero.
[^2]:H ere w e u se the term *velocit y* t odenote a deriv a ti ve of the c oo rdi nate s of th e ge o de sic with respe c tto the pr op er len gt h o f t he ge o desi c. Alt hough th ese q u antities are n o t velocitiesi nt h ep hysi cal sense, the y ap p earin t h egeo d esicequat io n si n the same fashionas actua l vel ocities would for time- l i k e geodes ics. - - -
abstract: 'M otion of many c o mets isaffec ted by n on-gravit a t ional fo rce s c aus edb y o utgassing fro m thei rsurface s.Outgass ing al sopro du ces react ive torq ue sre su lti ng in cometary s pin e vol ution . We pr opose tha tth e tw o proce s se s areco rr elat edan d sho w th a t t he chan ge of com eta r y sp in r ate ove r its helioce nt ric orbitsc ale s line a r ly withthe amplitude of its no n -gravit ati onalacce leration. Th e prop ort i onalit y cons tantde pen d s on t h e c ome tsize and o r b ita l ele me nts(semi-m ajor axis and ecce n tri city) and onthe (di m e ns ion l es s ) l ev e r a r m parameter $\ze ta$ that r el a te s the outg a ssi ng -induce d torqu e and acceler ation. We determin e$\ze t a $ f or 7 comet s for wh ich bothn on-gr a vi tatio nal accel er ati on an d chan g e o f spi n peri od $\Del ta P$ w ere meas ured and verify this re lation . Thi s s ample spa nni n g a lmost 4 d ecad es in $\De lta P$ yiel ds$ \log\ zeta = -2 .21 \ pm 0. 54$, surprisin g ly sm a l lvalue and s p r e ad. We t hen applyourframework to 209c omets with mea sure d non | ) 463\]_\[Nuovo Cim. B_[**44**]{} (1966) 1\]. E._Poisson, *A_Relativist’s_Toolkit*, Cambridge_University_Press, Cambridge U.K._(2004)
V. Balasubramanian and S. F. Ross,_Phys. Rev. D [**61**]{} (2000) 044007_\[hep-th/9906226\]. T. Banks, M. R. Douglas,_G. T. Horowitz_and E. J. Martinec, hep-th/9808016. V. Keranen and P. Kleinert, JHEP [**1504**]{} (2015) 119 \[arXiv:1412.2806 \[hep-th\]\].
[^1]: Note that_there_is an_ambiguity_in_the literature related to the_sign of $K_{ij}$. For example_[@Poisson] defines_$K_{ij}$ with an opposite sign. In this case_the_sign ambguity does_not matter, but can cause confusion when the energy-momentum_of the joining hypersurface is non-zero.
[^2]:_Here we use_the_term_*velocity* to denote a_derivative of the coordinates of the_geodesic with respect to the proper_length of the geodesic. Although these quantities_are not velocities in the physical_sense, they appear in the_geodesic equations_in the same fashion as_actual velocities would_for time-like_geodesics.
---
abstract: 'Motion_of many comets is affected by_non-gravitational forces caused_by outgassing from their surfaces. Outgassing_also_produces reactive torques_resulting_in_cometary spin_evolution. We propose_that_the two_processes_are correlated and show that the_change_of cometary spin rate over its heliocentric_orbit scales linearly with_the_amplitude of its non-gravitational_acceleration. The proportionality constant depends_on the comet size and orbital_elements (semi-major_axis and_eccentricity) and on the (dimensionless) lever arm parameter $\zeta$ that relates_the outgassing-induced torque and acceleration. We_determine $\zeta$ for 7_comets for_which_both non-gravitational acceleration_and_change of_spin period $\Delta P$ were measured and_verify this_relation. This sample spanning almost 4_decades in $\Delta P$_yields_$\log\zeta=-2.21\pm 0.54$, surprisingly small value and_spread. We then apply our framework_to 209 comets with measured_non |
for Gaussian and Poisson noises we have, respectively, $2\mu_G(1)-\mu_G(2)=-2D$ and $2\mu_P(1)-\mu_P(2) = \nu[1-\exp(ig)]^2$. The time decay of the pair correlator of $\delta u(t)=u(t)-\langle u \rangle$ is exponential, with exponent $\Lambda + \mu(1)\equiv\tilde\Gamma + i\delta\tilde\omega$.
Not only does the pair correlator, Eq (\[eq:delta\_corr\_2nd\_correlator\]), allow one to reveal frequency noise where there are no conventional spectral signatures of it, but it also gives an insight into the noise statistics. More insights can be gained from the higher-order moments of $u(t)$. By writing $$\begin{aligned}
\label{eq:ordered_moments}
&&u_F^n(0) = n!\int_{-\infty}^0dt_1\int_{-\infty}^{t_1}dt_2\ldots \int_{-\infty}^{t_{n-1}}dt_n (-i)^n\nonumber\\
&&\times\exp\left\{\sum_{j=1}^{n}\left[\Lambda t_j+ i(n+1-j)\int_{t_{j}}^{t_{j-1}}dt'_j\xi(t'_j)\right]\right\},\end{aligned}$$ we obtain from Eq. (\[eq:charact\_fnctnl\]) $$\begin{aligned}
\label{eq:moments_delta_corr}
\langle u^n\rangle = n!\left(\frac{-iF}{4\omega_F}\right)^n\prod_{j=1}^n\left[j\Lambda + \mu(j)\right]^{-1}.\end{aligned}$$ From Eq. (\[eq:moments\_delta\_corr\]), by measuring the moments of the complex amplitude $u(t)$, one can find function $\mu(k)$ for all integer $k$ and therefore, given that this function is analytical at least near the real-$k$ axis, find the whole $\mu(k)$ and thus the full statistics of the $\delta$-correlated noise $\xi(t)$.
Comparatively weak frequency noise
==================================
The presence of non-$\delta$-correlated frequency | for Gaussian and Poisson noises we have, respectively, $ 2\mu_G(1)-\mu_G(2)=-2D$ and $ 2\mu_P(1)-\mu_P(2) = \nu[1-\exp(ig)]^2$. The time decay of the pair correlator of $ \delta u(t)=u(t)-\langle uracil \rangle$ is exponential, with advocate $ \Lambda + \mu(1)\equiv\tilde\Gamma + i\delta\tilde\omega$.
Not only does the couple correlator, Eq (\[eq: delta\_corr\_2nd\_correlator\ ]), allow one to unwrap frequency noise where there are no ceremonious apparitional signatures of it, but it also gives an insight into the noise statistic. More insights can be gained from the higher - order consequence of $ u(t)$. By writing $ $ \begin{aligned }
\label{eq: ordered_moments }
& & u_F^n(0) = n!\int_{-\infty}^0dt_1\int_{-\infty}^{t_1}dt_2\ldots \int_{-\infty}^{t_{n-1}}dt_n (-i)^n\nonumber\\
& & \times\exp\left\{\sum_{j=1}^{n}\left[\Lambda t_j+ i(n+1 - j)\int_{t_{j}}^{t_{j-1}}dt'_j\xi(t'_j)\right]\right\},\end{aligned}$$ we receive from Eq. (\[eq: charact\_fnctnl\ ]) $ $ \begin{aligned }
\label{eq: moments_delta_corr }
\langle u^n\rangle = n!\left(\frac{-iF}{4\omega_F}\right)^n\prod_{j=1}^n\left[j\Lambda + \mu(j)\right]^{-1}.\end{aligned}$$ From Eq. (\[eq: moments\_delta\_corr\ ]), by measuring the moments of the complex amplitude $ u(t)$, one can recover function $ \mu(k)$ for all integer $ k$ and therefore, given that this routine is analytical at least near the real-$k$ axis, find the whole $ \mu(k)$ and thus the entire statistics of the $ \delta$-correlated noise $ \xi(t)$.
Comparatively weak frequency noise
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
The bearing of non-$\delta$-correlated frequency | fog Gaussian and Poisson nuises we have, rgspectivxly, $2\mu_G(1)-\ju_G(2)=-2D$ and $2\mu_P(1)-\mu_P(2) = \nu[1-\exp(ig)]^2$. The time deray if tht pair correlator ow $\delta u(n)=u(t)-\langle u \rengle$ is exponenvjal, witm expkkent $\Nembda + \mu(1)\equiv\tllde\Gamma + h\delta\tilde\omeca$.
Voc only does the pair correlator, Eq (\[eq:qelta\_cotr\_2jd\_correlator\]), ajlow jne fo reveal frequency noise where thsre are no conventiomal spectral signatures of it, hut it also gives wn insight unto rhe noise stxtistics. More insights can be gained from the higher-ofder koments of $u(r)$. Bj writing $$\bejin{alidned}
\label{eq:ovcered_mmments}
&&u_G^n(0) = n!\int_{-\infty}^0db_1\int_{-\iifty}^{r_1}dt_2\ldots \int_{-\infty}^{t_{n-1}}dt_i (-i)^n\nonumber\\
&&\times\exp\jeft\{\sum_{j=1}^{n}\neyt[\Lambda t_j+ i(n+1-j)\int_{t_{j}}^{t_{h-1}}dr'_j\xi(t'_b)\rigvt]\rieyt\},\evd{amijnes}$$ we ohtamn from Eq. (\[es:charact\_fncrnl\]) $$\begin{aligned}
\labtl{ez:noments_delta_ckrr}
\landlq u^n\rangle = n!\left(\frac{-iF}{4\omega_F}\right)^n\prod_{b=1}^n\lsft[j\Lambda + \mu(j)\right]^{-1}.\end{qligned}$$ From Eq. (\[eq:momejts\_delta\_cjrr\]), by measuring the moments of the complex amplidude $n(t)$, ont gwv flnd function $\mu(k)$ for all integer $k$ and therefjde, gpven that this fukction is analyticsl ay least near tfe real-$k$ zxis, find the wholf $\mu(k)$ agd thys the fujl syatistics of the $\delta$-correoated noise $\qi(t)$.
Cimparatively weak yrequency nonse
==================================
The presrnce of non-$\delta$-correlaced frsquency | for Gaussian and Poisson noises we have, and = \nu[1-\exp(ig)]^2$. time decay of u(t)=u(t)-\langle \rangle$ is exponential, exponent $\Lambda + + i\delta\tilde\omega$. Not only does the correlator, Eq (\[eq:delta\_corr\_2nd\_correlator\]), allow one to reveal frequency noise where there are no spectral signatures of it, but it also gives an insight into the noise More can gained the higher-order moments of $u(t)$. By writing $$\begin{aligned} \label{eq:ordered_moments} &&u_F^n(0) = n!\int_{-\infty}^0dt_1\int_{-\infty}^{t_1}dt_2\ldots \int_{-\infty}^{t_{n-1}}dt_n (-i)^n\nonumber\\ &&\times\exp\left\{\sum_{j=1}^{n}\left[\Lambda t_j+ we obtain from Eq. (\[eq:charact\_fnctnl\]) $$\begin{aligned} \label{eq:moments_delta_corr} \langle = n!\left(\frac{-iF}{4\omega_F}\right)^n\prod_{j=1}^n\left[j\Lambda + \mu(j)\right]^{-1}.\end{aligned}$$ Eq. (\[eq:moments\_delta\_corr\]), by measuring the of complex amplitude one find $\mu(k)$ for all $k$ and therefore, given that this function is analytical at least near the real-$k$ axis, find the $\mu(k)$ and full statistics the noise Comparatively weak frequency The presence of non-$\delta$-correlated frequency | for Gaussian and Poisson noisEs we have, reSpectIveLy, $2\mU_G(1)-\Mu_G(2)=-2D$ And $2\mU_P(1)-\mu_P(2) = \nu[1-\exp(ig)]^2$. ThE Time Decay of the pair correlatOr of $\dElTA u(t)=u(T)-\LaNgle u \Rangle$ iS ExPONenTiAl, WitH eXPoNent $\LAmbDa + \mu(1)\equIv\tilde\GamMa + i\DeLta\tilde\omegA$.
noT only does tHe pAir correlatoR, Eq (\[Eq:deltA\_cOrr\_2ND\_corrElaTor\]), alLow one TO reveaL frequencY nOIse wheRE there aRE No ConvEntional spectral sIGnATures of it, but it Also giVeS An INSigHt iNto the noisE sTatisTIcs. More INsIGHTs cAN be gained from The higher-orDEr mOments Of $U(t)$. BY WritinG $$\begiN{aLIgnEd}
\label{eq:orDereD_moments}
&&u_f^n(0) = n!\int_{-\INfty}^0dt_1\iNT_{-\infty}^{t_1}Dt_2\ldotS \inT_{-\inFty}^{t_{N-1}}Dt_N (-i)^N\noNuMBer\\
&&\TImEs\eXP\leFt\{\sum_{j=1}^{n}\lEfT[\LAmbda T_j+ i(n+1-J)\INT_{T_{j}}^{t_{j-1}}Dt'_j\Xi(t'_j)\Right]\Right\},\end{alignEd}$$ wE obtAIn fRom Eq. (\[Eq:chaRact\_FnCtnl\]) $$\bEgin{alIgned}
\LaBel{eq:moments_delTa_coRr}
\langle u^N\raNgLe = n!\LeFt(\fraC{-IF}{4\omegA_F}\rIghT)^n\prod_{j=1}^N\left[j\LAMbdA + \mU(J)\RIgHt]^{-1}.\end{aligned}$$ From Eq. (\[Eq:MOMeNts\_delta\_Corr\]), by MEaSuRIng the moMeNts Of thE COmpleX ampLItUde $u(t)$, one Can finD FuNcTion $\mu(k)$ FoR all inTeGer $K$ anD therEFore, Given tHat this fUnctiON is analytical aT Least near the rEAl-$K$ AXiS, Find The Whole $\mu(k)$ and Thus THe fuLl stATiStiCS of thE $\deltA$-cORrELated noise $\xi(t)$.
ComparAtIvely wEak frEquency noise
==================================
THe presence OF NOn-$\delta$-cOrreLAtED frequency | for Gaussian and Poissonnoises wehave, re spe ct ivel y, $ 2\mu_G(1)-\mu_ G (2)= -2D$ and $2\mu_P(1)-\m u_P(2 )= \nu [ 1- \exp( ig)]^2$ . T h e ti me d eca yo fthe p air correl ator of $\ del ta u(t)=u(t)-\ l an gle u \ran gle $ is exponen tia l, wit hexp o nent$\L ambda + \mu ( 1)\equ iv\tilde\ Ga m ma + i \ delta\t i l de \ome ga$.
Not only do e st he pair correl ator,Eq (\ [ e q:d elt a\_corr\_2 nd \_cor r elator\ ] ), a l low one to reveal frequencyn ois e wher ethe r e areno co nv e nti onal spectr al s ignatures of it , but it also gi ves an in sig ht i n to t heno i ses ta tis t ics . More i ns ig hts c an b e g a ined fr om t he hi gher-order mo men ts o f $u (t)$. By w riti ng $$\b egin{a ligne d}
\label{eq:orde red_ moments}&&u _F ^n( 0) = n! \ int_{- \in fty }^0dt_1 \int_{- \ inf ty } ^ { t_ 1}dt_2\ldots \int_ {- \ i nf ty}^{t_{ n-1}}d t _n ( - i)^n\non um ber \\
& & \ times \exp \ le ft\{\sum _{j=1} ^ {n }\ left[\L am bda t_ j+ i( n+1 -j)\i n t_{t _{j}}^ {t_{j-1} }dt'_ j \xi(t'_j)\righ t ]\right\},\en d {a l i gn e d}$$ we obtain fro m Eq . (\[ eq:c h ar act \ _fnct nl\]) $ $ \b e gin{aligned}
\label {e q:mome nts_d elta_corr}
\l angle u^n\ r a n gle = n! \lef t (\ f rac{-iF}{4\ome ga_F} \right)^n\ p rod_{j=1 }^n\l eft[j\La mbda + \m u ( j)\right ]^{ -1} .\e nd{ a l ig ned}$$ From E q . (\[ eq :moment s\_ delta\_ cor r\] ),byme asuring t he momen ts o fth e c omple x amplitu de $u (t )$, onec an fin d fun ctio n$\ m u(k )$ fora ll i nteg er $ k$ a ndth erefo re,g ive n thatthis func tio n isan al yticalat least near t he real-$k $axi s, fin d the whol e $\mu(k)$ and thus the full st ati stics ofthe $\del ta$ -corre lat e d nois e $\xi (t)$.
Com p a rativ e l ywea kfrequencyn o ise
==== == ==== ======= =================The presence ofnon -$\d e l ta $-c o rr e lat ed fre q u ency | for_Gaussian and_Poisson noises we have,_respectively, $2\mu_G(1)-\mu_G(2)=-2D$_and_$2\mu_P(1)-\mu_P(2) =_\nu[1-\exp(ig)]^2$._The time decay_of the pair_correlator of $\delta u(t)=u(t)-\langle_u \rangle$ is_exponential,_with exponent $\Lambda + \mu(1)\equiv\tilde\Gamma + i\delta\tilde\omega$.
Not only does the pair correlator, Eq (\[eq:delta\_corr\_2nd\_correlator\]), allow_one_to reveal_frequency_noise_where there are no conventional_spectral signatures of it, but_it also_gives an insight into the noise statistics. More_insights_can be gained_from the higher-order moments of $u(t)$. By writing $$\begin{aligned}
\label{eq:ordered_moments}
&&u_F^n(0)_= n!\int_{-\infty}^0dt_1\int_{-\infty}^{t_1}dt_2\ldots \int_{-\infty}^{t_{n-1}}dt_n (-i)^n\nonumber\\
&&\times\exp\left\{\sum_{j=1}^{n}\left[\Lambda t_j+ i(n+1-j)\int_{t_{j}}^{t_{j-1}}dt'_j\xi(t'_j)\right]\right\},\end{aligned}$$_we obtain from_Eq. (\[eq:charact\_fnctnl\])_$$\begin{aligned}
\label{eq:moments_delta_corr}
\langle_u^n\rangle = n!\left(\frac{-iF}{4\omega_F}\right)^n\prod_{j=1}^n\left[j\Lambda +_\mu(j)\right]^{-1}.\end{aligned}$$ From Eq. (\[eq:moments\_delta\_corr\]), by measuring the_moments of the complex amplitude $u(t)$,_one can find function $\mu(k)$ for all_integer $k$ and therefore, given that_this function is analytical at_least near_the real-$k$ axis, find the_whole $\mu(k)$ and_thus the_full statistics of_the $\delta$-correlated noise $\xi(t)$.
Comparatively weak frequency_noise
==================================
The presence of_non-$\delta$-correlated frequency |
+1}}{r_n} =
\sqrt{1 + A_{n}\sin\left(2\theta_{n}\right) + A_{n}^{2}\sin^{2}\theta_{n}}.
\label{eq:4}$$
Using action-angle variables (\[eq:3\]), (\[eq:4\]), the inverse localization length is written in a quite simple form (see details in [@IRT98]), $$\begin{aligned}
l_{\infty}^{-1} \equiv \Lambda = \lim_{L \rightarrow \infty} \frac{1}{L} \sum_{n=1}^{L}
\mbox{ln} \mid \frac{\psi_{n+1}}{\psi_{n}} \mid = \lim_{L \rightarrow
\infty} \frac{1}{L} \sum_{n=1}^{L}
\mbox{ln} \frac{r_{n+1}}{r_{n}} \nonumber \\
= \frac{1}{2}\left< \mbox{ln} \left( 1 + A_n \sin2\theta_n + A_n^2\sin^2\theta_n \right)
\right>_n \;.
\label{eq:5a}\end{aligned}$$ Here the brackets $\left<...\right>_n$ stand for averaging over $n$, i.e. along the trajectory of the map (\[eq:3\]), (\[eq:4\]). It is important to note that the above expression for $\Lambda$ depends only on the angle $\theta_{n}$ and not on the radius $r_{n}$. The above relation is correct for the energies $E$ not very close to the band edges $E=\pm 2$ where $\mu = 0,\pi$. At the band edges, there is additional contribution to the localization length that depends on the ratio $\sin \theta_{n+1}/\sin \theta_{n}$ [@IRT98]. In this paper we consider the case when the energy $E$ is inside the allowed band, $|E| < 2$. Moreover, we also exclude the band center $E=0$ where the localization length has a singular behavior that requires specific treatment (see [@IRT98] and references therein).
Apart from the abovementioned restrictions, the relation (\[eq:5a\]) is valid for [*any*]{} potential $\epsilon_n$. However, | +1}}{r_n } =
\sqrt{1 + A_{n}\sin\left(2\theta_{n}\right) + A_{n}^{2}\sin^{2}\theta_{n } }.
\label{eq:4}$$
Using action - angle variables (\[eq:3\ ]), (\[eq:4\ ]), the inverse localization of function duration is written in a quite simple human body (witness details in [ @IRT98 ]), $ $ \begin{aligned }
l_{\infty}^{-1 } \equiv \Lambda = \lim_{L \rightarrow \infty } \frac{1}{L } \sum_{n=1}^{L }
\mbox{ln } \mid \frac{\psi_{n+1}}{\psi_{n } } \mid = \lim_{L \rightarrow
\infty } \frac{1}{L } \sum_{n=1}^{L }
\mbox{ln } \frac{r_{n+1}}{r_{n } } \nonumber \\
= \frac{1}{2}\left < \mbox{ln } \left (1 + A_n \sin2\theta_n + A_n^2\sin^2\theta_n \right)
\right>_n \; .
\label{eq:5a}\end{aligned}$$ Here the brackets $ \left<... \right>_n$ stall for average over $ n$, i.e. along the trajectory of the map (\[eq:3\ ]), (\[eq:4\ ]). It is important to note that the above expression for $ \Lambda$ count only on the angle $ \theta_{n}$ and not on the spoke $ r_{n}$. The above relation is correct for the energies $ E$ not very close to the band edges $ E=\pm 2 $ where $ \mu = 0,\pi$. At the band edge, there is additional contribution to the localization distance that depends on the ratio $ \sin \theta_{n+1}/\sin \theta_{n}$ [ @IRT98 ]. In this paper we consider the event when the energy $ E$ is inside the allowed band, $ |E| < 2$. Moreover, we also bar the band center $ E=0 $ where the localization length has a singular behavior that requires specific treatment (see [ @IRT98 ] and references therein).
Apart from the abovementioned restrictions, the relation (\[eq:5a\ ]) is valid for [ * any * ] { } potential $ \epsilon_n$. However, | +1}}{r_n} =
\dqrt{1 + A_{n}\sin\left(2\theta_{n}\rigmt) + A_{n}^{2}\sin^{2}\theta_{n}}.
\lcvel{eq:4}$$
Uving adtion-angue variables (\[eq:3\]), (\[eq:4\]), the inverde locaoization length is wrigten in a quite sumplt form (see details in [@IRT98]), $$\begin{zpignzd}
o_{\infty}^{-1} \equiv \Lsmbda = \lhm_{L \rightarrow \ivfcy} \frac{1}{L} \sum_{n=1}^{L}
\mbox{ln} \mid \frac{\psi_{n+1}}{\psy_{n}} \mid = \lim_{L \rightarrjw
\ingey} \fdac{1}{L} \sum_{n=1}^{L}
\mbox{ln} \frac{r_{n+1}}{r_{n}} \nonumbsr \\
= \fgac{1}{2}\left< \mbox{ln} \legt( 1 + A_n \sin2\theta_n + A_n^2\sin^2\thfta_n \right)
\right>_n \;.
\label{fq:5a}\end{aligngs}$$ Hqee the brackdts $\left<...\right>_n$ stand fkr averaging over $n$, i.e. along thd tralectory of jkw mwk (\[eq:3\]), (\[eq:4\]). It iw impjrtant to nobv that dhe abofe expression nor $\Lembdq$ depends only on the angle $\theta_{n}$ and got on tha xadius $r_{n}$. The above rwlqtion is worrdxt wor tie snergifs $X$ not very dlose to thw band edges $E=\pm 2$ wnewv $\mu = 0,\pi$. At tge banq qdges, there is additional contribution uo ths localization length tyat depends on the rajio $\sin \thqta_{n+1}/\sin \theta_{n}$ [@IRT98]. In this paper we consider the wase xhdn uhc dbeggy $E$ is inside the allowed band, $|E| < 2$. Moreover, se akso exclude thc band center $E=0$ whrrf yre localizatiun lenyfh has a singular bejavior jhat rwquires skecifoc treatment (see [@IRT98] and rederences thegein).
Qpart from the aborementioned xestrivtionx, the relation (\[eq:5a\]) is vclid fkr [*any*]{} potejtial $\epsjuon_n$. However, | +1}}{r_n} = \sqrt{1 + A_{n}\sin\left(2\theta_{n}\right) + A_{n}^{2}\sin^{2}\theta_{n}}. action-angle (\[eq:3\]), (\[eq:4\]), inverse localization length simple (see details in $$\begin{aligned} l_{\infty}^{-1} \equiv = \lim_{L \rightarrow \infty} \frac{1}{L} \sum_{n=1}^{L} \mid \frac{\psi_{n+1}}{\psi_{n}} \mid = \lim_{L \rightarrow \infty} \frac{1}{L} \sum_{n=1}^{L} \mbox{ln} \frac{r_{n+1}}{r_{n}} \nonumber \\ \frac{1}{2}\left< \mbox{ln} \left( 1 + A_n \sin2\theta_n + A_n^2\sin^2\theta_n \right) \right>_n \;. \label{eq:5a}\end{aligned}$$ the $\left<...\right>_n$ for over $n$, i.e. along the trajectory of the map (\[eq:3\]), (\[eq:4\]). It is important to note the above expression for $\Lambda$ depends only on angle $\theta_{n}$ and not the radius $r_{n}$. The above is for the $E$ very to the band $E=\pm 2$ where $\mu = 0,\pi$. At the band edges, there is additional contribution to the localization that depends ratio $\sin \theta_{n}$ In paper we consider when the energy $E$ is inside $|E| < 2$. Moreover, we also exclude the center $E=0$ the localization length has a singular that requires specific treatment (see [@IRT98] and references Apart from the abovementioned restrictions, the relation (\[eq:5a\]) is valid for [*any*]{} potential $\epsilon_n$. However, | +1}}{r_n} =
\sqrt{1 + A_{n}\sin\left(2\theta_{n}\righT) + A_{n}^{2}\sin^{2}\thetA_{n}}.
\labEl{eQ:4}$$
UsInG actIon-aNgle variables (\[eQ:3\]), (\[Eq:4\]), thE inverse localization leNgth iS wRItteN In A quitE simple FOrM (SEe dEtAiLs iN [@Irt98]), $$\bEgin{aLigNed}
l_{\infTy}^{-1} \equiv \LamBda = \LiM_{L \rightarrow \INfTy} \frac{1}{L} \sum_{N=1}^{L}
\mBox{ln} \mid \frac{\Psi_{N+1}}{\psi_{n}} \mId = \Lim_{l \RightArrOw
\infTy} \frac{1}{l} \Sum_{n=1}^{L}
\mBox{ln} \frac{R_{n+1}}{R_{N}} \nonumBEr \\
= \frac{1}{2}\lEFT< \mBox{lN} \left( 1 + A_n \sin2\theta_n + A_N^2\SiN^2\Theta_n \right)
\rigHt>_n \;.
\labEl{EQ:5a}\END{alIgnEd}$$ Here the bRaCkets $\LEft<...\righT>_N$ sTAND foR Averaging over $N$, i.e. along the TRajEctory Of The MAp (\[eq:3\]), (\[eq:4\]). it is iMpORtaNt to note thaT the Above exprEssion FOr $\LambdA$ Depends Only on The AngLe $\thETa_{N}$ aNd nOt ON thE RaDiuS $R_{n}$. THe above rElAtIon is CorrECT FOr thE enErgiEs $E$ noT very close to tHe bAnd eDGes $e=\pm 2$ whEre $\mu = 0,\Pi$. At ThE band Edges, tHere iS aDditional contriButiOn to the loCalIzAtiOn LengtH That dePenDs oN the ratIo $\sin \thETa_{n+1}/\SiN \THEtA_{n}$ [@IRT98]. In this paper we CoNSIdEr the casE when tHE eNeRGy $E$ is insIdE thE allOWEd banD, $|E| < 2$. MoREoVer, we alsO excluDE tHe Band cenTeR $E=0$ wherE tHe lOcaLizatIOn leNgth haS a singulAr behAVior that requirES specific treaTMeNT (SeE [@iRT98] aNd rEferences thEreiN).
aparT froM ThE abOVemenTioneD rEStRIctions, the relation (\[eQ:5a\]) Is valiD for [*aNy*]{} potential $\epSilon_n$. HoweVER, | +1}}{r_n} =
\sqrt{1 + A_{n }\sin\left (2\th eta _{n }\ righ t) + A_{n}^{2}\sin ^ {2}\ theta_{n}}.
\label{eq: 4}$$
U s inga ct ion-a ngle va r ia b l es(\ [e q:3 \] ) ,(\[eq :4\ ]), the inverse l oca li zation lengt h i s writtenina quite simp leform ( se e d e tails in [@IR T98]), $$\beg in{aligne d} l_{\in f ty}^{-1 } \e quiv \Lambda = \lim _ {L \rightarrow \i nfty}\f r ac { 1 }{L } \ sum_{n=1}^ {L }
\m b ox{ln}\ mi d \ fra c {\psi_{n+1}}{ \psi_{n}} \ m id = \l im _{L \right arrow
\ i nft y} \frac{1} {L}\sum_{n=1 }^{L}\ mbox{ln } \frac {r_{n+ 1}} {r_ {n}} \ no num be r \\ = \ f rac {1}{2}\l ef t< \mbo x{ln } \ l eft( 1+ A_ n \si n2\theta_n +A_n ^2\s i n^2 \thet a_n \ righ t)
\rig ht>_n\;.
\ la bel{eq:5a}\end{ alig ned}$$ He reth e b ra ckets $\left <.. .\r ight>_n $ stand for a v e r ag ing over $n$, i.e. a l o ng the tra jector y o ft he map ( \[ eq: 3\]) , (\[eq :4\] ) .It is im portan t t onote th at the a bo veexp ressi o n fo r $\La mbda$ de pends only on the an g le $\theta_{n } $a n dn ot o n t he radius $ r_{n } $. T he a b ov e r e latio n isco r re c t for the energies$E $ notveryclose to theband edges $ E =\pm 2$wher e $ \ mu = 0,\pi$. A t the band edge s , thereis ad ditional contribu t i on to th e l oca liz ati o n l ength that de p e ndson the ra tio $\sin\th eta _{n +1} /\ sin \thet a_{n}$ [ @I RT 98 ]. In this paper we c ons id erthe c a se whe n the ene rg y$ E$is insi d et h e al lo we d ba nd, $ |E| < 2$. Mor eover,we also e xcl u de t he b and cen ter $E=0$ whe re the local iz ati on len g t h has asingular behavior thatr equires sp ecifi c tr eatment ( see [@IRT 98] and re ferenc es th er ein ) .
Apa r t f rom t he aboveme n t ion ed re st rict ions, t he relation (\[eq: 5 a\] ) is valid fo r [ *any * ] {} po t en t ial $ \ eps i l on_n$. However, | +1}}{r_n} =
\sqrt{1_+ A_{n}\sin\left(2\theta_{n}\right)_+ A_{n}^{2}\sin^{2}\theta_{n}}.
\label{eq:4}$$
Using action-angle variables_(\[eq:3\]), (\[eq:4\]),_the_inverse localization_length_is written in_a quite simple_form (see details in_[@IRT98]), $$\begin{aligned}
l_{\infty}^{-1} \equiv_\Lambda_ = \lim_{L \rightarrow \infty} \frac{1}{L} \sum_{n=1}^{L}
\mbox{ln} \mid \frac{\psi_{n+1}}{\psi_{n}} \mid =__\lim_{L \rightarrow
\infty}_\frac{1}{L}_\sum_{n=1}^{L}
\mbox{ln}_ \frac{r_{n+1}}{r_{n}} \nonumber \\
_= \frac{1}{2}\left< \mbox{ln} \left(_1 +_A_n \sin2\theta_n + A_n^2\sin^2\theta_n \right)
\right>_n \;.
\label{eq:5a}\end{aligned}$$ Here the_brackets_$\left<...\right>_n$ stand for_averaging over $n$, i.e. along the trajectory of the_map (\[eq:3\]), (\[eq:4\]). It is important_to note that_the_above_expression for $\Lambda$ depends_only on the angle $\theta_{n}$ and_not on the radius $r_{n}$. The_above relation is correct for the energies_$E$ not very close to the_band edges $E=\pm 2$ where_$\mu =_0,\pi$. At the band edges,_there is additional_contribution to_the localization length_that depends on the ratio $\sin_\theta_{n+1}/\sin \theta_{n}$ [@IRT98]._In this paper we consider the_case_when the energy_$E$_is_inside the_allowed band, $|E|_<_2$. Moreover,_we_also exclude the band center $E=0$_where_the localization length has a singular behavior_that requires specific treatment_(see_[@IRT98] and references therein).
Apart_from the abovementioned restrictions, the_relation (\[eq:5a\]) is valid for [*any*]{}_potential $\epsilon_n$._However, |
{2}\kappa$ for a reverse phase slip. The Gibbs free-energy barrier for the two kinds of phase slips, computed by subtracting the Gibbs free energy for the ground state from that of the saddle-point state, is $$\Delta G=\Ecore \left(\frac{16 \sqrt{2}}{3}\pm 4 J \pi\right).$$ The former free-energy is obtained by substituting into for both wires; the latter one is obtained by substituting into for both wires. We note that the Gibbs free-energy barrier heights for parallel phase slips (in both the forward and reverse directions) are just double those of the LAMH result for a single wire. From the barrier heights, we can work out the generated voltage by appealing to the Josephson relation, $V
= (\hbar/2 e) \dot{\Theta}$, and to the fact that each phase slip corresponds to the addition (or subtraction) of $2\pi$ to the phase. Hence, we arrive at the current-voltage relation associated with parallel phase slips at $\delta=0$: $$\begin{aligned}
V_{\delta=0 \, \text{, par}}=\frac{\hbar}{e} \, \Omega \, e^{-\beta
\Ecore \frac{16 \sqrt{2}}{3}} \sinh \left(I/I_0\right),
\label{v:d0:par}\end{aligned}$$ where the prefactor $\Omega$ may be computed using time-dependent Ginzburg-Landau theory or extracted from experiment, and $I_0=4
e/\beta h$.
If we take into account the two lowest excited states, which we ignored earlier, then voltage can also be generated via sequential phase slips (in addition to the parallel ones, treated above). To tackle this case, we construct a diagram in which the vertices represent the metastable and saddle-point solutions of, and the edges represent the corresponding free energy barriers; see Fig. \[0\_diag\]. Pairs of metastable-state vertices are connected via two saddle-point-state vertices, corresponding to a phase slip on either the top or the bottom wire. To go from one metastable state to another, the system must follow the edge out of the starting metastable state leading to the desired saddle-point state. We assume that, once the saddle-point state is reached, the top of the barrier has been passed and the order parameter relaxes to the target metastable state | { 2}\kappa$ for a reverse phase slip. The Gibbs free - energy barrier for the two kind of phase chemise, computed by subtracting the Gibbs spare department of energy for the ground state from that of the saddle - period state, is $ $ \Delta G=\Ecore \left(\frac{16 \sqrt{2}}{3}\pm 4 J \pi\right).$$ The erstwhile free - energy is prevail by substituting into for both wires; the latter one is prevail by substituting into for both wires. We note that the Gibbs free - energy barrier high for parallel phase slips (in both the forward and reverse guidance) are just double those of the LAMH result for a single wire. From the barrier high, we can work out the generated voltage by appealing to the Josephson relation, $ five
= (\hbar/2 e) \dot{\Theta}$, and to the fact that each phase slip corresponds to the addition (or subtraction) of $ 2\pi$ to the phase. Hence, we arrive at the current - voltage relation associated with parallel phase slips at $ \delta=0 $: $ $ \begin{aligned }
V_{\delta=0 \, \text {, par}}=\frac{\hbar}{e } \, \Omega \, e^{-\beta
\Ecore \frac{16 \sqrt{2}}{3 } } \sinh \left(I / I_0\right),
\label{v: d0: par}\end{aligned}$$ where the prefactor $ \Omega$ may be computed using time - dependent Ginzburg - Landau theory or extracted from experiment, and $ I_0=4
e/\beta h$.
If we take into account the two lowest excited state, which we ignored earlier, then electric potential can besides be generated via sequential phase slips (in addition to the parallel one, treated above). To tackle this case, we construct a diagram in which the vertex represent the metastable and saddle - point solutions of, and the edges represent the corresponding detached energy barriers; see Fig. \[0\_diag\ ]. Pairs of metastable - department of state vertices are connect via two saddleback - point - state vertices, correspond to a phase slip on either the top or the bottom wire. To fail from one metastable country to another, the system must follow the boundary out of the starting metastable state leading to the desired saddle - detail state. We bear that, once the saddle - period state is pass, the top of the barrier has been passed and the order parameter relaxes to the aim metastable state | {2}\kapoa$ for a reverse phase suip. The Gibbs ftew-energb barrisr for tfe two kinds of phase slips, rompyted vy subtracting the Gibcs free ejergy foe tht ground state from that of the daddnx-point state, is $$\Delta G=\Ecmre \left(\frac{16 \sxrg{2}}{3}\pl 4 J \pi\right).$$ The former free-energy ys obtaonfd by substitujing pneo fkg noth wires; the latter one is obtzined bj substituting inyo for both wires. We note hhat the Gibbs free-eneggy barrier heidyts for paraulel phase slips (in bojh the forward and reverse direcgions) are just eoyblf those of tie LAMR result for a singla wire. Grom the barricr hemghtw, we can work out the generated voltage by appeaniug to the Josephson rwlqtion, $V
= (\htar/2 d) \dog{\Thtta}$, ahd to hhe fact that each phase slip corresponds tp eye addition (od subtwastion) of $2\pi$ to the phase. Hence, we arrivt at fhe current-voltage relarion associated with karallel prase slips at $\delta=0$: $$\begin{aligned}
V_{\delta=0 \, \text{, par}}=\fsac{\hber}{d} \, \Imcga \, w^{-\bfta
\Ecore \frac{16 \sqrt{2}}{3}} \sinh \left(I/I_0\right),
\label{v:d0:par}\qhd{slpgned}$$ where the pvefactor $\Omega$ may bf vjmputed using time-dzlehdent Ginzburg-Landwu theoty or wxtracted frok experiment, and $I_0=4
e/\beta h$.
If we take intj account the two llwest excitzd stajes, whoch we ignored earlier, chen vkltage can wlso be gsverated via sequdntpal [hase slips (in addition to the paraolel ones, tfeatgd abovq). To tacklf this case, we construct a dicgram in which hhe vertices represent the metastable and sadcla-popnt solutnons on, and the edges represent the corres'ondine free eneggy barrixrs; see Fig. \[0\_dyag\]. Pairs of kgtastable-statx verticef arw cobnected xia two saddle-loint-statv rertices, xorresponding to a phxae slip on eithzx uhw top or the bpttum rige. Vo go xrom one metdstacle xtate to anouker, ghe xystem must follow tve esge out of the stattlng metasjable staee leading to the desired saddlt-point stete. We asfume that, once the saddle-point state is rewched, the to[ of the barrier has been passed and the order parametec relaxes to the targej metastable state | {2}\kappa$ for a reverse phase slip. The barrier the two of phase slips, free for the ground from that of saddle-point state, is $$\Delta G=\Ecore \left(\frac{16 4 J \pi\right).$$ The former free-energy is obtained by substituting into for both the latter one is obtained by substituting into for both wires. We note the free-energy heights parallel phase slips (in both the forward and reverse directions) are just double those of the result for a single wire. From the barrier we can work out generated voltage by appealing to Josephson $V = e) and the fact that phase slip corresponds to the addition (or subtraction) of $2\pi$ to the phase. Hence, we arrive at current-voltage relation parallel phase at $$\begin{aligned} \, \text{, par}}=\frac{\hbar}{e} \, e^{-\beta \Ecore \frac{16 \sqrt{2}}{3}} \sinh the prefactor $\Omega$ may be computed using time-dependent theory or from experiment, and $I_0=4 e/\beta h$. we take into account the two lowest excited which we ignored earlier, then voltage can also be generated via sequential phase slips (in the parallel ones, treated To tackle this we a in the vertices the metastable and saddle-point solutions of, and the edges represent the free energy barriers; see Fig. \[0\_diag\]. Pairs of metastable-state vertices via saddle-point-state vertices, corresponding a phase slip on the or the bottom wire. from metastable the must the edge out of starting metastable state leading to desired saddle-point state. We state is reached, the top of the barrier been passed and the order parameter relaxes the target metastable state | {2}\kappa$ for a reverse phase slip. the Gibbs frEe-eneRgy BarRiEr foR the Two kinds of phasE SlipS, computed by subtracting The GiBbS Free ENeRgy foR the groUNd STAte FrOm ThaT oF ThE saddLe-pOint staTe, is $$\Delta G=\ecoRe \Left(\frac{16 \sqrt{2}}{3}\PM 4 J \Pi\right).$$ The ForMer free-energY is ObtainEd By sUBstitUtiNg intO for boTH wires; The latter OnE Is obtaINed by suBSTiTutiNg into for both wireS. we NOte that the GibbS free-eNeRGy BARriEr hEights for pArAllel PHase sliPS (iN BOTh tHE forward and reVerse directIOns) Are jusT dOubLE those Of the lAmh reSult for a sinGle wIre. From thE barriER heightS, We can woRk out tHe gEneRateD VoLtAge By APpeALiNg tO The josephsoN rElAtion, $v
= (\hbaR/2 E) \DOT{\TheTa}$, aNd to The faCt that each phaSe sLip cORreSpondS to thE addItIon (or SubtraCtion) Of $2\Pi$ to the phase. HenCe, we Arrive at tHe cUrRenT-vOltagE RelatiOn aSsoCiated wIth paraLLel PhASE SlIps at $\delta=0$: $$\begin{aliGnED}
v_{\dElta=0 \, \text{, Par}}=\fraC{\HbAr}{E} \, \omega \, e^{-\beTa
\ecoRe \frAC{16 \Sqrt{2}}{3}} \sInh \lEFt(i/I_0\right),
\lAbel{v:d0:PAr}\EnD{aligneD}$$ wHere thE pRefActOr $\OmeGA$ may Be compUted usinG time-DEpendent GinzbuRG-Landau theory OR eXTRaCTed fRom Experiment, aNd $I_0=4
e/\BEta h$.
if we TAkE inTO accoUnt thE tWO lOWest excited states, whIcH we ignOred eArlier, then volTage can alsO BE GenerateD via SEqUEntial phase sliPs (in aDdition to tHE paralleL ones, Treated aBove). To tacKLE this casE, we ConStrUct A DIaGram in which thE VErtiCeS represEnt The metaStaBle And SadDlE-point solUtions of, AnD tHe EdGes RepreSEnt the coRrEspOnDinG free ENergy bArrieRs; seE FIg. \[0\_DIag\]. pairs of MEtASTablE-sTaTe veRtiCeS are cOnneCTed Via two sAddle-poinT-stATe veRtIcEs, correSponding to a phAsE slip on eitHeR thE top or THE bottom wIre. To go from one metastablE State to AnoTher, tHe syStem must fOllOw the eDge OUt of thE startIng meTaStaBLE statE LEaDinG tO the desireD SAddLe-poiNt StatE. We assuMe that, once the saddlE-PoiNt state is reacHed, The tOP Of The BArRIer HaS BeeN PAssed and the ordeR parameter ReLAxEs to the tarGEt mEtAstable State | {2}\kappa$ for a reverse p hase slip. TheGib bsfr ee-e nerg y barrier fort he t wo kinds of phase slip s, co mp u tedb ysubtr actingt he G ibb sfr eeen e rg y for th e groun d state fr omth at of the sa d dl e-point st ate , is $$\Delt a G =\Ecor e\le f t(\fr ac{ 16 \s qrt{2} } {3}\pm 4 J \pi\ ri g ht).$$ The for m e rfree -energy is obtain e db y substituting intofo r b o t h w ire s; the lat te r one is obta i ne d b y s u bstituting in to for both wir es. We n ote that t he Gi bb s fr ee-energy b arri er height s forp arallel phase s lips ( inbot h th e f or war da ndr ev ers e di rections )ar e jus t do u b l e tho seof t he LA MH result for asing l e w ire.Fromtheba rrier heigh ts, w ecan work out th e ge nerated v olt ag e b yappea l ing to th e J osephso n relat i on, $ V = ( \hbar/2 e) \dot{\T he t a }$ , and to the f a ct t h at eachph ase sli p corre spon d sto the a dditio n ( or subtra ct ion) o f$2\ pi$ to t h e ph ase. H ence, we arri v e at the curre n t-voltage rel a ti o n a s soci ate d with para llel phas e sl i ps at $\del ta=0$ :$ $\ b egin{aligned}
V_{\d el ta=0 \ , \te xt{, par}}=\f rac{\hbar} { e } \, \Ome ga \ , e ^ {-\beta
\Ecore \fra c{16 \sqrt { 2}}{3}}\sinh \left(I /I_0\righ t ) ,
\label {v: d0: par }\e n d {a ligned}$$ whe r e the p refacto r $ \Omega$ ma y b e c omp ut ed usingtime-dep en de nt G inz burg- L andau th eo ryor ex tract e d from expe rime nt ,a nd$I_0=4e /\ b e ta h $.
If w e t ak e int o ac c oun t the t wo lowest ex c ited s ta tes, wh ich we ignore dearlier, t he n v oltage c an alsobe generated via sequen t ial pha seslips (in addition to the p ara l lel on es, tr eated a bov e ) . Tot a ck leth is case, w e con struc ta di agram i n which the vertic e s r epresent themet asta b l eand sa d dle -p o int s olutions of, an d the edge sr ep resent the cor re spondin g freeenerg y barrie rs; see F ig. \[0\_ di ag\] . Pai rs of meta stable-s tate vert i ces a r econne cte d viatw o s addle -point - sta te ve rtices ,corres pondi ng to a ph ase slip on either thetop or thebot tom wire. To gofrom onemeta stable sta tetoanoth er, the s yste m m ust follo w th e edge out of th e st arting meta s t a ble stat e l e adingto t he desired saddle - point state. W e as s u metha t , on ce the saddle-po int s t a te is re ac hed, the to p of the b a rrier has b een pa ssed an d th e order par ame ter relax esto the tar ge tm etasta blest ate | {2}\kappa$ for_a reverse_phase slip. The Gibbs_free-energy barrier_for_the two_kinds_of phase slips,_computed by subtracting_the Gibbs free energy_for the ground_state_from that of the saddle-point state, is $$\Delta G=\Ecore \left(\frac{16 \sqrt{2}}{3}\pm 4 J \pi\right).$$_The_former free-energy_is_obtained_by substituting into for both_wires; the latter one is_obtained by_substituting into for both wires. We note that_the_Gibbs free-energy barrier_heights for parallel phase slips (in both the forward_and reverse directions) are just double_those of the_LAMH_result_for a single wire._From the barrier heights, we can_work out the generated voltage by_appealing to the Josephson relation, $V
= (\hbar/2_e) \dot{\Theta}$, and to the fact_that each phase slip corresponds_to the_addition (or subtraction) of $2\pi$_to the phase._Hence, we_arrive at the_current-voltage relation associated with parallel phase_slips at $\delta=0$:_$$\begin{aligned}
V_{\delta=0 \, \text{, par}}=\frac{\hbar}{e} \, \Omega_\,_e^{-\beta
\Ecore \frac{16 \sqrt{2}}{3}}_\sinh_\left(I/I_0\right),
\label{v:d0:par}\end{aligned}$$_where the_prefactor $\Omega$ may_be_computed using_time-dependent_Ginzburg-Landau theory or extracted from experiment,_and_$I_0=4
e/\beta h$.
If we take into account the_two lowest excited states,_which_we ignored earlier, then_voltage can also be generated_via sequential phase slips (in addition_to the_parallel ones,_treated above). To tackle this case, we construct a diagram in_which the vertices represent the metastable_and saddle-point solutions of,_and the_edges_represent the corresponding_free_energy barriers;_see Fig. \[0\_diag\]. Pairs of metastable-state vertices are_connected via_two saddle-point-state vertices, corresponding to a_phase slip on either_the_top or the bottom wire. To_go from one metastable state to_another, the system must follow_the_edge_out of the starting metastable_state leading to the desired saddle-point_state. We assume_that, once the saddle-point state is reached,_the_top of the barrier has been_passed_and the order parameter relaxes to_the_target_metastable state |
{H}^A$ to $\mathcal{H}^B$ and density operators on $\mathcal{H}^B \otimes \mathcal{H}^A$, which is essential for FIG. 5 and Theorem \[thm:stabprop\].
Completely positive maps $\Lambda$ with $0 \leq \text{Tr}(\Lambda(\rho)) \leq \text{Tr}(\rho)$ have an operational interpretation: the associated channels can ‘fail’ or ‘abort’ the computation by yielding 0. For example, let $\Lambda$ be the channel that measures in the $\sigma_Z$ basis and postselects on obtaining $\ket{0}$. Then $\Lambda(\ket{1}\bra{1}) = 0$, and $\Lambda(\ket{+}\bra{+}) = \frac{1}{2} \ket{0}\bra{0}$.
Let $\Lambda$ be a completely positive map from $\mathcal{H}^A$ to $\mathcal{H}^B$. Let $\ket{\text{Bell}_A} \in \mathcal{H}^A \otimes \mathcal{H}^A$ be a Bell state for $\mathcal{H}^A$, i.e. if $\{\ket{i}\}$ are a basis for $\mathcal{H}^A$ then:
$$\ket{\text{Bell}_A} = \frac{1}{\sqrt{\text{dim}(\mathcal{H}^A)}} \sum_i \ket{i} \otimes \ket{i}$$
The **un-normalized Choi state** $\phi_\Lambda$ of $\Lambda$ is the resulting state when $\Lambda$ is applied to one half of $\ket{\text{Bell}_A}$.
$$\phi_\Lambda = (\Lambda \otimes I)(\ket{\text{Bell}_A}\bra{\text{Bell}_A} ) \in \mathcal{H}^B \otimes \mathcal{H}^A$$
$\text{Tr}(\phi_\Lambda$) of $\Lambda$ can be less than 1 if $\Lambda$ is not trace preserving. Let $\bar \phi_\Lambda = \phi_\Lambda/\text{Tr}(\phi_\Lambda)$ be the **normalized Choi state** with trace 1. This distinction is crucial.
To calculate the output of a channel $\Lambda(\rho)$ given its Choi state $\phi_\Lambda$ we compute: $$\Lambda(\rho) = \text{dim}(\mathcal | { H}^A$ to $ \mathcal{H}^B$ and density operators on $ \mathcal{H}^B \otimes \mathcal{H}^A$, which is essential for FIG. 5 and Theorem \[thm: stabprop\ ].
Completely incontrovertible map $ \Lambda$ with $ 0 \leq \text{Tr}(\Lambda(\rho) ) \leq \text{Tr}(\rho)$ have an operational interpretation: the consort channels can ‘ fail ’ or ‘ abort ’ the computation by yield 0. For exercise, let $ \Lambda$ be the groove that measures in the $ \sigma_Z$ basis and postselects on prevail $ \ket{0}$. Then $ \Lambda(\ket{1}\bra{1 }) = 0 $, and $ \Lambda(\ket{+}\bra{+ }) = \frac{1}{2 } \ket{0}\bra{0}$.
Let $ \Lambda$ be a completely positivist map from $ \mathcal{H}^A$ to $ \mathcal{H}^B$. lease $ \ket{\text{Bell}_A } \in \mathcal{H}^A \otimes \mathcal{H}^A$ be a Bell state for $ \mathcal{H}^A$, i.e. if $ \{\ket{i}\}$ are a basis for $ \mathcal{H}^A$ then:
$ $ \ket{\text{Bell}_A } = \frac{1}{\sqrt{\text{dim}(\mathcal{H}^A) } } \sum_i \ket{i } \otimes \ket{i}$$
The * * un - normalize Choi state * * $ \phi_\Lambda$ of $ \Lambda$ is the resulting state when $ \Lambda$ is put on to one half of $ \ket{\text{Bell}_A}$.
$ $ \phi_\Lambda = (\Lambda \otimes I)(\ket{\text{Bell}_A}\bra{\text{Bell}_A }) \in \mathcal{H}^B \otimes \mathcal{H}^A$$
$ \text{Tr}(\phi_\Lambda$) of $ \Lambda$ can be less than 1 if $ \Lambda$ is not trace preserving. Let $ \bar \phi_\Lambda = \phi_\Lambda/\text{Tr}(\phi_\Lambda)$ be the * * normalize Choi state * * with trace 1. This distinction is crucial.
To calculate the end product of a channel $ \Lambda(\rho)$ given its Choi state $ \phi_\Lambda$ we compute: $ $ \Lambda(\rho) = \text{dim}(\mathcal | {H}^A$ ho $\mathcal{H}^B$ and density operators on $\mcrhcal{H}^U \otimea \mathcau{H}^A$, which is essential for FMG. 5 abd Thtjrem \[thm:stabprop\].
Comoletely plsitive naps $\Oambda$ witi $0 \leq \tcrt{Tr}(\Lzlbda(\xhi)) \leq \text{Tr}(\rhp)$ have an mperational indefpxetation: the associated channels can ‘fail’ ot ‘wbort’ the compotatipg by jitlding 0. For example, let $\Lambda$ be the chennel that measires in the $\sigma_Z$ basis ajd plstselects on obtalning $\ket{0}$. Tyen $\Jqmbda(\ket{1}\bra{1}) = 0$, and $\Lambda(\ket{+}\bra{+}) = \frzc{1}{2} \ket{0}\bra{0}$.
Let $\Lambda$ be a complegely 'ositive mak rrlk $\mathcal{H}^A$ to $\mwthcal{H}^B$. Let $\ket{\text{Tell}_A} \im \mathcal{H}^A \otlmes \katycal{H}^A$ be a Bell statx for $\mathcal{H}^A$, i.e. is $\{\ket{i}\}$ ara c basis for $\mathcal{H}^A$ tyen:
$$\kej{\text{Tell}_X} = \ffac{1}{\aqct{\tsxt{dim}(\latical{H}^A)}} \sum_i \ket{i} \otimew \ket{i}$$
The **un-normalieed Bnoi state** $\phj_\Lambdw$ jf $\Lambda$ is the resulting state when $\Ldmbsa$ is applied to one haof of $\ket{\text{Bell}_A}$.
$$\phi_\Pambda = (\Lwmbda \otimes I)(\ket{\text{Bell}_A}\bra{\text{Bell}_A} ) \in \mathcan{H}^B \ovioes \mwgycwl{H}^A$$
$\text{Tr}(\phi_\Lambda$) of $\Lambda$ can be less thag 1 of $\Lambda$ is not trace presrrgimd. Let $\bar \phi_\Uambda = \pgi_\Lambda/\text{Tr}(\phi_\Lwmbda)$ bg the **bormalizeq Chpi state** with trace 1. This dustinction if crucial.
To calculace the outpuc of a chanmel $\Lambda(\rho)$ given its Choj state $\phi_\Pambda$ we zompute: $$\Lambda(\rhu) = \next{gim}(\mathcal | {H}^A$ to $\mathcal{H}^B$ and density operators on \mathcal{H}^A$, is essential FIG. 5 and $\Lambda$ $0 \leq \text{Tr}(\Lambda(\rho)) \text{Tr}(\rho)$ have an interpretation: the associated channels can ‘fail’ ‘abort’ the computation by yielding 0. For example, let $\Lambda$ be the channel measures in the $\sigma_Z$ basis and postselects on obtaining $\ket{0}$. Then $\Lambda(\ket{1}\bra{1}) = and = \ket{0}\bra{0}$. $\Lambda$ be a completely positive map from $\mathcal{H}^A$ to $\mathcal{H}^B$. Let $\ket{\text{Bell}_A} \in \mathcal{H}^A \otimes \mathcal{H}^A$ a Bell state for $\mathcal{H}^A$, i.e. if $\{\ket{i}\}$ a basis for $\mathcal{H}^A$ $$\ket{\text{Bell}_A} = \frac{1}{\sqrt{\text{dim}(\mathcal{H}^A)}} \sum_i \ket{i} \ket{i}$$ **un-normalized Choi $\phi_\Lambda$ $\Lambda$ the resulting state $\Lambda$ is applied to one half of $\ket{\text{Bell}_A}$. $$\phi_\Lambda = (\Lambda \otimes I)(\ket{\text{Bell}_A}\bra{\text{Bell}_A} ) \in \mathcal{H}^B \otimes $\text{Tr}(\phi_\Lambda$) of be less 1 $\Lambda$ not trace preserving. \phi_\Lambda = \phi_\Lambda/\text{Tr}(\phi_\Lambda)$ be the **normalized trace 1. This distinction is crucial. To calculate output of channel $\Lambda(\rho)$ given its Choi state we compute: $$\Lambda(\rho) = \text{dim}(\mathcal | {H}^A$ to $\mathcal{H}^B$ and density opErators on $\mAthcaL{H}^B \OtiMeS \matHcal{h}^A$, which is essenTIal fOr FIG. 5 and Theorem \[thm:stabProp\].
COmPLeteLY pOsitiVe maps $\LAMbDA$ WitH $0 \lEq \TexT{TR}(\laMbda(\rHo)) \lEq \text{TR}(\rho)$ have an OpeRaTional interpREtAtion: the asSocIated channelS caN ‘fail’ oR ‘aBorT’ The coMpuTatioN by yieLDing 0. FoR example, lEt $\lAmbda$ bE The chanNEL tHat mEasures in the $\sigma_z$ BaSIs and postselecTs on obTaINiNG $\Ket{0}$. theN $\Lambda(\ket{1}\BrA{1}) = 0$, and $\LAMbda(\ket{+}\BRa{+}) = \FRAC{1}{2} \keT{0}\Bra{0}$.
Let $\Lambda$ bE a completelY PosItive mAp FroM $\MathcaL{H}^A$ to $\MaTHcaL{H}^B$. Let $\ket{\teXt{BeLl}_A} \in \mathCal{H}^A \oTImes \matHCal{H}^A$ be A Bell sTatE foR $\matHCaL{H}^a$, i.e. If $\{\KEt{i}\}$ ARe A baSIs fOr $\mathcaL{H}^a$ tHen:
$$\keT{\texT{bELL}_A} = \frAc{1}{\sQrt{\tExt{diM}(\mathcal{H}^A)}} \sum_I \keT{i} \otIMes \Ket{i}$$
THe **un-nOrmaLiZed ChOi statE** $\phi_\LAmBda$ of $\Lambda$ is thE resUlting staTe wHeN $\LaMbDa$ is aPPlied tO onE haLf of $\ket{\Text{BelL}_a}$.
$$\phI_\LAMBDa = (\lambda \otimes I)(\ket{\teXt{bELl}_a}\bra{\text{bell}_A} ) \iN \MaThCAl{H}^B \otimEs \MatHcal{h}^a$$
$\Text{TR}(\phi_\lAmBda$) of $\LamBda$ can BE lEsS than 1 if $\laMbda$ is NoT trAce PreseRVing. let $\bar \Phi_\LambdA = \phi_\LAMbda/\text{Tr}(\phi_\LAMbda)$ be the **normALiZED CHOi stAte** With trace 1. ThIs diSTincTion IS cRucIAl.
To cAlculAtE ThE Output of a channel $\LamBdA(\rho)$ giVen itS Choi state $\phi_\lambda$ we coMPUTe: $$\Lambda(\Rho) = \tEXt{DIm}(\mathcal | {H}^A$ to $\mathcal{H}^B$and densit y ope rat ors o n $\ math cal{H}^B \otim e s \m athcal{H}^A$, which is esse nt i al f o rFIG.5 and T h eo r e m \ [t hm :st ab p ro p\].
Co mpletel y positive ma ps $\Lambda$ w i th $0 \leq \ tex t{Tr}(\Lambd a(\ rho))\l eq\ text{ Tr} (\rho )$ hav e an op erational i n terpre t ation:t h easso ciated channels c a n‘ fail’ or ‘abor t’ the c o mp u t ati onby yieldin g0. Fo r exampl e ,l e t $\ L ambda$ be the channel th a t m easure sint he $\s igma_ Z$ bas is and post sele cts on ob tainin g $\ket{ 0 }$. The n $\La mbd a(\ ket{ 1 }\ br a{1 }) = 0 $ ,and $\L ambda(\k et {+ }\bra {+}) = \ frac {1} {2}\ket{ 0}\bra{0}$.
Let $\L a mbd a$ be a co mple te ly po sitive mapfr om $\mathcal{H} ^A$to $\math cal {H }^B $. Let$ \ket{\ tex t{B ell}_A} \in \m a thc al { H } ^A \otimes \mathcal{ H} ^ A $be a Bel l stat e f or $\mathca l{ H}^ A$,i . e. if $\{ \ ke t{i}\}$are ab as is for $\ ma thcal{ H} ^A$ th en:
$ $\ke t{\tex t{Bell}_ A} =\ frac{1}{\sqrt{ \ text{dim}(\ma t hc a l {H } ^A)} } \ sum_i \ket{ i} \ o time s \k e t{ i}$ $
The **un -n o rm a lized Choi state**$\ phi_\L ambda $ of $\Lambda $ is the r e s u lting st atew he n $\Lambda$ isappli ed to oneh alf of $ \ket{ \text{Be ll}_A}$.$ $\phi_\L amb da= ( \La m b da \otimes I)(\ k e t{\t ex t{Bell} _A} \bra{\t ext {Be ll} _A} ) \in \mat hcal{H}^ B\o ti me s \ mathc a l{H}^A$$
$\t ex t{T r}(\p h i_\Lam bda$) of$\ La m bda $ can b e l e s s th an 1 if$\L am bda$is n o t t race pr eserving. Le t $\b ar \ phi_\La mbda = \phi_\ La mbda/\text {T r}( \phi_\ L a mbda)$ b e the **normalized Choi state** wi th tr ace1. This d ist inctio n i s cruci al.
T o cal cu lat e the o u t pu t o fa channel$ \ Lam bda(\ rh o)$given i ts Choi state $\ph i _\L ambda$ we com put e: $ $ \ La mbd a (\ r ho) = \te x t {dim}(\mathcal | {H}^A$ to_$\mathcal{H}^B$ and_density operators on $\mathcal{H}^B_\otimes \mathcal{H}^A$,_which_is essential_for_FIG. 5 and Theorem \[thm:stabprop\].
Completely_positive maps $\Lambda$_with $0 \leq \text{Tr}(\Lambda(\rho))_\leq \text{Tr}(\rho)$ have_an_operational interpretation: the associated channels can ‘fail’ or ‘abort’ the computation by yielding 0._For_example, let_$\Lambda$_be_the channel that measures in_the $\sigma_Z$ basis and postselects_on obtaining_$\ket{0}$. Then $\Lambda(\ket{1}\bra{1}) = 0$, and $\Lambda(\ket{+}\bra{+}) =_\frac{1}{2}_\ket{0}\bra{0}$.
Let $\Lambda$ be_a completely positive map from $\mathcal{H}^A$ to $\mathcal{H}^B$. Let_$\ket{\text{Bell}_A} \in \mathcal{H}^A \otimes \mathcal{H}^A$ be_a Bell state_for_$\mathcal{H}^A$,_i.e. if $\{\ket{i}\}$ are_a basis for $\mathcal{H}^A$ then:
$$\ket{\text{Bell}_A} =_\frac{1}{\sqrt{\text{dim}(\mathcal{H}^A)}} \sum_i \ket{i} \otimes \ket{i}$$
The **un-normalized_Choi state** $\phi_\Lambda$ of $\Lambda$ is the_resulting state when $\Lambda$ is applied_to one half of $\ket{\text{Bell}_A}$.
$$\phi_\Lambda_= (\Lambda_\otimes I)(\ket{\text{Bell}_A}\bra{\text{Bell}_A} ) \in \mathcal{H}^B_\otimes \mathcal{H}^A$$
$\text{Tr}(\phi_\Lambda$) of_$\Lambda$ can_be less than_1 if $\Lambda$ is not trace_preserving. Let $\bar_\phi_\Lambda = \phi_\Lambda/\text{Tr}(\phi_\Lambda)$ be the **normalized_Choi_state** with trace_1._This_distinction is_crucial.
To calculate the_output_of a_channel_$\Lambda(\rho)$ given its Choi state $\phi_\Lambda$_we_compute: $$\Lambda(\rho) = \text{dim}(\mathcal |
EC}(\epsilon_{\rm c}).$$ where $\lceil~\rceil$ represents the ceiling function and $\lambda_{\rm EC}(\epsilon_{\rm c})$ is the cost of error correction to achieve $\epsilon_{\rm c}$-correctness. For simplicity, we will replace the right-hand side by a slightly pessimistic bound as $$n_{\rm fin}\leq n_Z(1-h\left (\frac{f(k_X,n_X,n_{\rm tot})}{n_Z}\right))
-{\rm log_2} \frac{2}{\epsilon_{\rm PA}} - \lambda_{\rm EC}(\epsilon_{\rm c}).
\label{keyrate}$$
Bounds on phase errors {#bounds}
----------------------
In this subsection, we discuss the specific methods to obtain $f(k_X,n_X,n_{\rm tot})$ in Eq. (\[f\]) including a method based on the Bernoulli sampling, and a more conventional method based on the simple random sampling. We also introduce a third, rather convoluted method, which will help to elucidate the difference between the former two methods.
Before discussing each of the methods, we first derive general statistical properties. Since the $Z$-labeled phase error and the $X$-labeled bit error are obtained by identical measurements, the procedure to obtain those errors is equivalent to the following steps after discarding the rounds with no-detection (i.e., with Bob failing to receive a qubit): (a) Alice and Bob further discard each of the remaining rounds with probability $1-\tilde{p}_Z^2-\tilde{p}_X^2$. (b) They make $X$-basis measurements on the remaining $n_{\rm tot}$ rounds and obtain $k_{\rm tot}$ errors. (c) Finally, they label each of the $n_{\rm tot}$ rounds as $Z$ or $X$ with probability $p_Z$ and $p_X$ (see [Eq. (\[pdef\])]{}), respectively, and obtain $k_{\rm ph}$ phase errors in $Z$-labeled rounds and $k_X=k_{\rm tot}-k_{\rm ph}$ bit errors in $X$-labeled rounds. In this procedure, since $k_X$ errors are sampled from $k_{\rm tot}$ errors with a fixed probability $p_X$, it follows a binomial distribution if $k_{\rm tot}$ and $n_{\rm | EC}(\epsilon_{\rm c}).$$ where $ \lceil~\rceil$ represents the ceiling function and $ \lambda_{\rm EC}(\epsilon_{\rm c})$ is the cost of error discipline to achieve $ \epsilon_{\rm c}$-correctness. For ease, we will replace the right - bridge player english by a slightly pessimistic bound as $ $ n_{\rm fin}\leq n_Z(1 - h\left (\frac{f(k_X, n_X, n_{\rm tot})}{n_Z}\right) )
-{\rm log_2 } \frac{2}{\epsilon_{\rm PA } } - \lambda_{\rm EC}(\epsilon_{\rm c }).
\label{keyrate}$$
bounce on phase error { # bounds }
----------------------
In this subsection, we discourse the specific methods to obtain $ f(k_X, n_X, n_{\rm tot})$ in Eq. (\[f\ ]) admit a method based on the Bernoulli sample distribution, and a more conventional method based on the simple random sampling. We also introduce a third, preferably convoluted method, which will help to clear the difference between the former two methods.
Before discussing each of the method, we first derive general statistical properties. Since the $ Z$-labeled phase error and the $ X$-labeled moment error are obtained by identical measurements, the procedure to obtain those error is equivalent to the following steps after discarding the rounds with no - detection (i.e., with Bob failing to receive a qubit ): (a) Alice and Bob further discard each of the remaining rounds with probability $ 1-\tilde{p}_Z^2-\tilde{p}_X^2$. (b) They do $ X$-basis measurements on the remaining $ n_{\rm tot}$ rounds and receive $ k_{\rm tot}$ errors. (c) Finally, they tag each of the $ n_{\rm tot}$ rounds as $ Z$ or $ X$ with probability $ p_Z$ and $ p_X$ (see [ Eq. (\[pdef\ ]) ] { }), respectively, and obtain $ k_{\rm ph}$ phase errors in $ Z$-labeled turn and $ k_X = k_{\rm tot}-k_{\rm ph}$ bit errors in $ X$-labeled rounds. In this procedure, since $ k_X$ errors are sampled from $ k_{\rm tot}$ error with a fixed probability $ p_X$, it follows a binomial distribution if $ k_{\rm tot}$ and $ n_{\rm | EC}(\fpsilon_{\rm c}).$$ where $\lceil~\rgeil$ represents jhw ceilmng fundtion ana $\lambda_{\rm EC}(\epsilon_{\rm c})$ is vhe xost if error correction to achieve $\vpsilon_{\rm c}$-cocrectness. For simplicity, we wimp re'lece the right-hakd side by d slightly pesvioidtic bound as $$n_{\rm fin}\leq n_Z(1-h\left (\fwac{f(k_X,n_C,n_{\gm tot})}{n_Z}\right))
-{\ri lob_2} \frad{2}{\vpwilon_{\rm PA}} - \lambda_{\rm EC}(\epsiloh_{\rm c}).
\lebel{keyrate}$$
Bouncs on phase errors {#bounds}
----------------------
Ij thls subsection, we dlscuss the wpecydic methods go obtain $f(k_X,n_X,n_{\rm tot})$ in Eq. (\[f\]) including a method basea on che Bernouloi salkling, and a nore bonventional method bdsed on the simple rakdom vampling. We also introdure a third, rather cogvoluted kechod, which will help ro eluchdata thd diwfedeice betwefn vhe former fwo methods.
Vefore discussing escr of the methoss, we siwst derive general statistical propertits. Sihce the $Z$-labeled phase wrror and the $X$-labelef bit errjr are obtained by identical measurements, the prowedurx go ibbain rhlse errors is equivalent to the following ste[a sfner discarding thc rounds with no-deyeftojn (i.e., with Boc failnhg to receive a qubih): (a) Alise ane Bob furuher ciscard each of the remainibg rounds winh peobability $1-\tilde{p}_Z^2-\cilde{p}_X^2$. (b) Thzy makg $X$-basos measurements on the xemainjng $n_{\rm tot}$ rounds aha obtain $k_{\rm tot}$ ergors. (c) Finally, they label each of the $n_{\cm toc}$ rounds as $E$ or $X$ rith probahilitn $p_Z$ and $p_X$ (see [Eq. (\[pfef\])]{}), rgspecthvely, and lbtain $k_{\rm ph}$ phase errors in $Z$-labeled rounds atd $n_X=k_{\rm toc}-k_{\rm pm}$ bit errors in $X$-labeled rounbs. In thns prozedure, sinbe $k_X$ errmrs are sam[led from $k_{\rm jot}$ errors wivh a fixeq pribabulity $p_B$, it follows a ninomial bnstributiin if $k_{\rm tot}$ and $k_{\rm | EC}(\epsilon_{\rm c}).$$ where $\lceil~\rceil$ represents the ceiling $\lambda_{\rm c})$ is cost of error For we will replace right-hand side by slightly pessimistic bound as $$n_{\rm fin}\leq (\frac{f(k_X,n_X,n_{\rm tot})}{n_Z}\right)) -{\rm log_2} \frac{2}{\epsilon_{\rm PA}} - \lambda_{\rm EC}(\epsilon_{\rm c}). \label{keyrate}$$ Bounds on errors {#bounds} ---------------------- In this subsection, we discuss the specific methods to obtain tot})$ Eq. including method based on the Bernoulli sampling, and a more conventional method based on the simple random We also introduce a third, rather convoluted method, will help to elucidate difference between the former two Before each of methods, first general statistical properties. the $Z$-labeled phase error and the $X$-labeled bit error are obtained by identical measurements, the procedure to those errors to the steps discarding rounds with no-detection Bob failing to receive a qubit): Bob further discard each of the remaining rounds probability $1-\tilde{p}_Z^2-\tilde{p}_X^2$. They make $X$-basis measurements on the $n_{\rm tot}$ rounds and obtain $k_{\rm tot}$ errors. Finally, they label each of the $n_{\rm tot}$ rounds as $Z$ or $X$ with probability $p_X$ (see [Eq. (\[pdef\])]{}), and obtain $k_{\rm phase in rounds $k_X=k_{\rm tot}-k_{\rm bit errors in $X$-labeled rounds. In this procedure, since $k_X$ errors sampled from $k_{\rm tot}$ errors with a fixed probability $p_X$, a distribution if $k_{\rm and $n_{\rm | EC}(\epsilon_{\rm c}).$$ where $\lceil~\rceIl$ represenTs the CeiLinG fUnctIon aNd $\lambda_{\rm EC}(\epSIlon_{\Rm c})$ is the cost of error corRectiOn TO achIEvE $\epsiLon_{\rm c}$-cORrECTneSs. foR siMpLIcIty, we WilL replacE the right-hAnd SiDe by a slightlY PeSsimistic bOunD as $$n_{\rm fin}\leq N_Z(1-h\Left (\frAc{F(k_X,N_x,n_{\rm tOt})}{n_z}\righT))
-{\rm log_2} \FRac{2}{\epsIlon_{\rm PA}} - \lAmBDa_{\rm EC}(\EPsilon_{\rM C}).
\LaBel{kEyrate}$$
Bounds on phaSE eRRors {#bounds}
----------------------
In thIs subsEcTIoN, WE diScuSs the speciFiC methODs to obtAIn $F(K_x,N_X,n_{\RM tot})$ in Eq. (\[f\]) inclUding a methoD BasEd on thE BErnOUlli saMplinG, aND a mOre conventiOnal Method basEd on thE Simple rANdom samPling. WE alSo iNtroDUcE a ThiRd, RAthER cOnvOLutEd method, WhIcH will Help TO ELUcidAte The dIfferEnce between thE foRmer TWo mEthodS.
BefoRe diScUssinG each oF the mEtHods, we first deriVe geNeral statIstIcAl pRoPertiES. Since The $z$-laBeled phAse erroR And ThE $x$-LAbEled bit error are obtAiNED bY identicAl measUReMeNTs, the proCeDurE to oBTAin thOse eRRoRs is equiValent TO tHe FollowiNg Steps aFtEr dIscArdinG The rOunds wIth no-detEctioN (I.e., with Bob failiNG to receive a quBIt): (A) aLiCE and bob Further discArd eACh of The rEMaIniNG rounDs witH pRObABility $1-\tilde{p}_Z^2-\tilde{p}_x^2$. (b) they maKe $X$-baSis measuremenTs on the remAINIng $n_{\rm toT}$ rouNDs ANd obtain $k_{\rm tot}$ ErrorS. (c) Finally, tHEy label eAch of The $n_{\rm toT}$ rounds as $z$ OR $X$ with prObaBilIty $P_Z$ aND $P_X$ (See [Eq. (\[pdef\])]{}), respECTiveLy, And obtaIn $k_{\Rm ph}$ phaSe eRroRs iN $Z$-lAbEled roundS and $k_X=k_{\rM tOt}-K_{\rM pH}$ biT erroRS in $X$-labeLeD roUnDs. IN this PRoceduRe, sinCe $k_X$ ErRoRS arE sampleD FrOM $K_{\rm tOt}$ ErRors WitH a Fixed ProbABilIty $p_X$, it Follows a bInoMIal dIsTrIbution If $k_{\rm tot}$ and $n_{\rM | EC}(\epsilon_{\rm c}).$$where $\lc eil~\ rce il$ r epre sent s the ceilingf unct ion and $\lambda_{\rmEC}(\ ep s ilon _ {\ rm c} )$ is t h ec o stof e rro rc or recti onto achi eve $\epsi lon _{ \rm c}$-corr e ct ness. Forsim plicity, wewil l repl ac e t h e rig ht- handside b y a sli ghtly pes si m isticb ound as $ $n _{\r m fin}\leq n_Z(1 - h\ l eft (\frac{f(k _X,n_X ,n _ {\ r m to t}) }{n_Z}\rig ht ))
-{ \ rm log_ 2 }\ f r ac{ 2 }{\epsilon_{\ rm PA}} - \ l amb da_{\r mEC} ( \epsil on_{\ rm c}) .
\label{k eyra te}$$
Bo unds o n phasee rrors { #bound s}--- ---- - -- -- --- -- - --- -
Int his subsect io n, we d iscu s s t he s pec ific meth ods to obtain $f (k_X , n_X ,n_{\ rm to t})$ i n Eq. (\[f\ ]) in cl uding a methodbase d on theBer no ull isampl i ng, an d a mo re conv entiona l me th o d ba sed on the simplera n d om samplin g. Wea ls oi ntroduce a th ird, r ather con v ol uted met hod, w h ic hwill he lp to el uc ida tethe d i ffer ence b etween t he fo r mer two method s .
Before dis c us s i ng each of the method s, w e fir st d e ri veg enera l sta ti s ti c al properties. Sinc ethe $Z $-lab eled phase er ror and th e $ X$-label ed b i te rror are obtai ned b y identica l measure ments , the pr ocedure t o obtain t hos e e rro rsi s e quivalent tot h e fo ll owing s tep s after di sca rdi ngth e roundswith no- de te ct io n ( i.e., with Bob f ail in g t o rec e ive aqubit ): ( a) A l ice and Bo b f u r ther d is card ea ch of t he r e mai ning ro unds with pr o babi li ty $1-\ti lde{p}_Z^2-\t il de{p}_X^2$ .(b) Theym a ke $X$-b asis measurements on th e remain ing $n_{ \rmtot}$ rou nds and o bta i n $k_{ \rm to t}$ e rr ors . (c) F i n al ly, t hey labele a chof th e$n_{ \rm tot }$ rounds as $Z$ o r $X $ with probab ili ty $ p _ Z$ an d $ p _X$ ( s ee[ E q. (\[pdef\])]{ }), respec ti v el y, and obt a in$k _{\rm p h}$ pha se er r ors in$Z$-label ed rounds a nd $ k _ X=k _{\rm tot} -k_{\rmph}$ bite rrors in $X$- lab eled r ou nds . Inthis p r oce dure, since $ k_X$ e rrors a re sampl ed from $k_{\rm tot}$ e rrorswitha f ixed prob abi l ity $p_X$, i t fo llows a bi nom ial dist rib u tionif $ k _{ \rm tot}$ and $n_{\rm | EC}(\epsilon_{\rm_c}).$$ where_$\lceil~\rceil$ represents the ceiling_function and_$\lambda_{\rm_EC}(\epsilon_{\rm c})$_is_the cost of_error correction to_achieve $\epsilon_{\rm c}$-correctness. For_simplicity, we will_replace_the right-hand side by a slightly pessimistic bound as $$n_{\rm fin}\leq n_Z(1-h\left (\frac{f(k_X,n_X,n_{\rm_tot})}{n_Z}\right))
-{\rm_log_2} \frac{2}{\epsilon_{\rm_PA}}_-_\lambda_{\rm EC}(\epsilon_{\rm c}).
\label{keyrate}$$
Bounds on_phase errors {#bounds}
----------------------
In this subsection,_we discuss_the specific methods to obtain $f(k_X,n_X,n_{\rm tot})$ in_Eq. (\[f\])_including a method_based on the Bernoulli sampling, and a more conventional_method based on the simple random_sampling. We also_introduce_a_third, rather convoluted method,_which will help to elucidate the_difference between the former two methods.
Before_discussing each of the methods, we first_derive general statistical properties. Since the_$Z$-labeled phase error and the_$X$-labeled bit_error are obtained by identical_measurements, the procedure_to obtain_those errors is_equivalent to the following steps after_discarding the rounds_with no-detection (i.e., with Bob failing_to_receive a qubit):_(a)_Alice_and Bob_further discard each_of_the remaining_rounds_with probability $1-\tilde{p}_Z^2-\tilde{p}_X^2$. (b) They make_$X$-basis_measurements on the remaining $n_{\rm tot}$ rounds_and obtain $k_{\rm tot}$_errors._(c) Finally, they label_each of the $n_{\rm tot}$_rounds as $Z$ or $X$ with_probability $p_Z$_and $p_X$_(see [Eq. (\[pdef\])]{}), respectively, and obtain $k_{\rm ph}$ phase errors in $Z$-labeled_rounds and $k_X=k_{\rm tot}-k_{\rm ph}$ bit_errors in $X$-labeled rounds._In this_procedure,_since $k_X$ errors_are_sampled from_$k_{\rm tot}$ errors with a fixed probability_$p_X$, it_follows a binomial distribution if $k_{\rm_tot}$ and $n_{\rm |
the current best feasible reference path.
The search process of PPNP resembles that in IDWS; PPNP here also maintains a priority queue $\mathbf{Q}$, and the search process also contains iterative rounds of loco-state examination. However, PPNP is subtly different from IDWS in RPGP. IDWS explores a tiny fraction of the loco-state space and finds a reference RW path, while PPNP investigates the loco-state space comprehensively and trims off redundant loco-states. While the search order in IDWS follows TECO, PWSO and VWNO, PPNP does not employ them. Instead, the order in PPNP is controlled by pruning strategies, and loco-states may re-enter the priority queue in PPNP. Throughout the search process, To skip redundant loco-states, PPNP explores the following pruning strategies. 1) *Infeasible Loco-State Pruning* (ILSP). If $\text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{\text{s}}, \gamma^{\text{v}}_{st}) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > C$, every RW path from $st_{\text{s}}$ to the destination via $st$ is infeasible. Thus, $st$ is removed. 2) *Suboptimal Loco-State Pruning* (SLSP). If $\text{l}_{\min}(\gamma^{\text{v}}_{\text{s}}, \gamma^{\text{v}}_{st}) + \text{l}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > \tilde{L}$, any RW path from $st_{\text{s}}$ to the destination via $st$ is longer than the reference RW path. $st$ is thereby removed. 3) *Unpromising Loco-State Locking* (ULSL). If $\text{c}_\text{c}(st_{\text{s}},st) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > C$, currently it is not likely to find any feasible RW path via $st$. Therefore, PPNP pauses the search expanded from $st$. Note that $st$ cannot be removed yet | the current best feasible reference path.
The search process of PPNP resemble that in IDWS; PPNP here besides maintain a priority queue $ \mathbf{Q}$, and the search procedure besides contains iterative rounds of loco - state of matter examination. However, PPNP is subtly different from IDWS in RPGP. IDWS research a tiny fraction of the loco - department of state space and finds a reference book RW path, while PPNP investigate the loco - state outer space comprehensively and trims off redundant loco - states. While the search club in IDWS follows TECO, PWSO and VWNO, PPNP does not employ them. Instead, the order in PPNP is control by pruning strategies, and loco - states may re - enter the priority queue in PPNP. Throughout the search process, To cut redundant loco - states, PPNP explores the following pruning strategies. 1) * Infeasible Loco - State Pruning * (ILSP). If $ \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{\text{s } }, \gamma^{\text{v}}_{st }) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st }, \gamma^{\text{v}}_{\text{t } }) > C$, every RW way from $ st_{\text{s}}$ to the finish via $ st$ is infeasible. Thus, $ st$ is removed. 2) * Suboptimal Loco - State Pruning * (SLSP). If $ \text{l}_{\min}(\gamma^{\text{v}}_{\text{s } }, \gamma^{\text{v}}_{st }) + \text{l}_{\min}(\gamma^{\text{v}}_{st }, \gamma^{\text{v}}_{\text{t } }) > \tilde{L}$, any RW path from $ st_{\text{s}}$ to the destination via $ st$ is longer than the reference RW path. $ st$ is thereby removed. 3) * Unpromising Loco - State Locking * (ULSL). If $ \text{c}_\text{c}(st_{\text{s}},st) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st }, \gamma^{\text{v}}_{\text{t } }) > C$, presently it is not probable to find any feasible RW path via $ st$. Therefore, PPNP hesitate the search expanded from $ st$. Note that $ st$ cannot be removed so far | thf current best feasible veference path.
Thg wearch proceas of PPVP resembles that in IDWS; PPIP hwre aoso maintains a priorigy queue $\lathbf{Q}$, qnd uhe search process also containa itexavive rounds of koco-state axamination. Hofexex, PPNP is subtly different from IDWS in RPGL. LDWS explores w timr frzbtlon of the loco-state space and fjnds a geference RW path, while PPNP investigates tje llco-state space comorehensiveli anq trims off rddundant loco-states. Whjle the search order in IDWS foulows TECO, PWSO abd GFNO, PPNP doxs not employ them. Instead, the orcer in PPNP is coitrooled by pruning stratxgies, and loco-states may re-endex the priority queue un PPNP. Thrmughuyt ghe sxardh profesa, To skip dedundant lico-states, PPNP explprqw the followihg prugigg strategies. 1) *Infeasible Loco-State Prutinf* (ILSP). If $\text{c}^{\alpha}_{\min}(\tamma^{\text{v}}_{\text{s}}, \gamma^{\jext{v}}_{st}) + \tqxt{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > C$, every RW petf fxim $st_{\gwxh{s}}$ to the destination via $st$ is infeasible. Thta, $xt$ is removed. 2) *Smboptimal Loco-Statr Origing* (SLSP). If $\jext{l}_{\miu}(\fajma^{\text{v}}_{\text{s}}, \gammw^{\text{v}}_{sj}) + \texr{l}_{\min}(\gammw^{\texy{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > \tilde{O}$, any RW patk feom $st_{\text{s}}$ to the destinatiou via $xt$ is longer than the refereuce RW path. $st$ is thereby ddmoved. 3) *Unpromiskng Lmco-State Locking* (ULSL). If $\tqxt{c}_\text{c}(wt_{\tert{s}},st) + \tdxt{c}^{\slpha}_{\myn}(\gamma^{\texh{v}}_{st}, \ndmma^{\text{v}}_{\text{t}}) > C$, furreutly ht is not pikely to find any feasible RW 'eth via $st$. Thgrexorv, PPNP pabses tme search expanqed from $st$. Noje that $sc$ cannut be remoned yet | the current best feasible reference path. The of resembles that IDWS; PPNP here $\mathbf{Q}$, the search process contains iterative rounds loco-state examination. However, PPNP is subtly from IDWS in RPGP. IDWS explores a tiny fraction of the loco-state space finds a reference RW path, while PPNP investigates the loco-state space comprehensively and off loco-states. the order in IDWS follows TECO, PWSO and VWNO, PPNP does not employ them. Instead, the order PPNP is controlled by pruning strategies, and loco-states re-enter the priority queue PPNP. Throughout the search process, skip loco-states, PPNP the pruning 1) *Infeasible Loco-State (ILSP). If $\text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{\text{s}}, \gamma^{\text{v}}_{st}) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > C$, every RW path from $st_{\text{s}}$ to the destination $st$ is $st$ is 2) Loco-State (SLSP). If $\text{l}_{\min}(\gamma^{\text{v}}_{\text{s}}, \text{l}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > \tilde{L}$, any RW to the destination via $st$ is longer than reference RW $st$ is thereby removed. 3) *Unpromising Locking* (ULSL). If $\text{c}_\text{c}(st_{\text{s}},st) + \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > currently it is not likely to find any feasible RW path via $st$. Therefore, PPNP search expanded from $st$. that $st$ cannot removed | the current best feasible refErence path.
the seArcH prOcEss oF PPNp resembles that IN IDWs; PPNP here also maintains A prioRiTY queUE $\mAthbf{q}$, and the SEaRCH prOcEsS alSo COnTains IteRative rOunds of locO-stAtE examination. hOwEver, PPNP is SubTly different FroM IDWS iN RpGP. idWS exPloRes a tIny fraCTion of The loco-stAtE Space aND finds a REFeRencE RW path, while PPNP iNVeSTigates the loco-State sPaCE cOMPreHenSively and tRiMs off REdundanT LoCO-STatES. While the searCh order in IDws foLlows TeCo, PWso and VWnO, PPNp dOEs nOt employ theM. InsTead, the orDer in PpnP is conTRolled bY pruniNg sTraTegiES, aNd LocO-sTAteS MaY re-ENteR the prioRiTy Queue In PPnp. tHRougHouT the SearcH process, To skiP reDundANt lOco-stAtes, PpNP eXpLores The folLowinG pRuning strategieS. 1) *InfEasible LoCo-STaTe PRuNing* (IlsP). If $\teXt{c}^{\AlpHa}_{\min}(\gaMma^{\text{V}}_{\TexT{s}}, \GAMMa^{\Text{v}}_{st}) + \text{c}^{\alpha}_{\mIn}(\GAMmA^{\text{v}}_{st}, \Gamma^{\tEXt{V}}_{\tEXt{t}}) > C$, everY Rw paTh frOM $St_{\texT{s}}$ to THe DestinatIon via $ST$ iS iNfeasibLe. thus, $st$ Is RemOveD. 2) *SuboPTimaL Loco-STate PrunIng* (SLsp). If $\text{l}_{\min}(\gamMA^{\text{v}}_{\text{s}}, \gaMMa^{\TEXt{V}}_{St}) + \teXt{l}_{\Min}(\gamma^{\texT{v}}_{st}, \GAmma^{\Text{V}}_{\TeXt{t}}) > \TIlde{L}$, Any RW PaTH fROm $st_{\text{s}}$ to the destiNaTion viA $st$ is Longer than the Reference Rw PATh. $st$ is thErebY ReMOved. 3) *UnpromisinG Loco-state LockiNG* (ULSL). If $\tExt{c}_\tExt{c}(st_{\teXt{s}},st) + \text{C}^{\ALpha}_{\min}(\gAmmA^{\teXt{v}}_{St}, \gAMMa^{\Text{v}}_{\text{t}}) > C$, cuRREntlY iT is not lIkeLy to finD anY feAsiBle rW Path via $st$. thereforE, PpNp pAuSes The seARch expanDeD frOm $St$. NOte thAT $st$ canNot be RemoVeD yET | the current best feasible reference path .
The s earc h pr ocess of PPNPr esem bles that in IDWS; PPN P her ea lsom ai ntain s a pri o ri t y qu eu e$\m at h bf {Q}$, an d the s earch proc ess a lso contains it erative ro und s of loco-st ate exami na tio n . How eve r, PP NP iss ubtlydifferent f r om IDW S in RPG P . I DWSexplores a tiny f r ac t ion of the loc o-stat es pa c e an d f inds a ref er enceR W path, wh i l e PP N P investigate s the loco- s tat e spac ecom p rehens ively a n d t rims off re dund ant loco- states . Whilet he sear ch ord erinIDWS fo ll ows T E CO, PW SOa ndVWNO, PP NP d oes n ot e m p l o y th em. Ins tead, the order in PP NP i s co ntrol led b y pr un ing s trateg ies,an d loco-states m ay r e-enter t hepr ior it y que u e in P PNP . T hrougho ut thes ear ch p r oc ess, To skip redun da n t l oco-stat es, PP N Pex p lores th efol lowi n g prun ings tr ategies. 1) *I n fe as ible Lo co -State P run ing * (IL S P).If $\t ext{c}^{ \alph a }_{\min}(\gamm a ^{\text{v}}_{ \ te x t {s } }, \ gam ma^{\text{v }}_{ s t})+ \t e xt {c} ^ {\alp ha}_{ \m i n} ( \gamma^{\text{v}}_{ st }, \ga mma^{ \text{v}}_{\t ext{t}}) > C $ , everyRW p a th from $st_{\tex t{s}} $ to the d e stinatio n via $st$ is infeasib l e . Thus,$st $ i s r emo v e d. 2) *Suboptim a l Loc o- State P run ing* (S LSP ).If$\t ex t{l}_{\mi n}(\gamm a^ {\ te xt {v} }_{\t e xt{s}},\g amm a^ {\t ext{v } }_{st} ) + \ text {l }_ { \mi n}(\gam m a^ { \ text {v }} _{st },\g amma^ {\te x t{v }}_{\te xt{t}}) > \t i lde{ L} $, any RW path from $s t_ {\text{s}} $tothe de s t inationvia $st$ is longer than the ref ere nce R W pa th. $st$isthereb y r e moved. 3) *U nprom is ing L oco-S t a te Lo ck ing* (ULSL ) . If $\te xt {c}_ \text{c }(st_{\text{s}},st ) +\text{c}^{\al pha }_{\ m i n} (\g a mm a ^{\ te x t{v } } _{st}, \gamma^{ \text{v}}_ {\ t ex t{t}}) > C $ , c ur rentlyit is n ot li k ely tofind anyfeasibleRW pat h via $st$. The refore,PPNP paus e s the se archexp andedfr om$st$. Notet hat $st$ canno tbe rem ovedye t | the_current best_feasible reference path.
The search_process of_PPNP_resembles that_in_IDWS; PPNP here_also maintains a_priority queue $\mathbf{Q}$, and_the search process_also_contains iterative rounds of loco-state examination. However, PPNP is subtly different from IDWS in_RPGP._IDWS explores_a_tiny_fraction of the loco-state space_and finds a reference RW_path, while_PPNP investigates the loco-state space comprehensively and trims_off_redundant loco-states. While_the search order in IDWS follows TECO, PWSO and_VWNO, PPNP does not employ them._Instead, the order_in_PPNP_is controlled by pruning_strategies, and loco-states may re-enter the_priority queue in PPNP. Throughout the_search process, To skip redundant loco-states, PPNP_explores the following pruning strategies. 1)_*Infeasible Loco-State Pruning* (ILSP). If_$\text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{\text{s}}, \gamma^{\text{v}}_{st})_+ \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) > C$,_every RW path_from $st_{\text{s}}$_to the destination_via $st$ is infeasible. Thus, $st$_is removed. 2)_*Suboptimal Loco-State Pruning* (SLSP). If $\text{l}_{\min}(\gamma^{\text{v}}_{\text{s}},_\gamma^{\text{v}}_{st})_+ \text{l}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}})_>_\tilde{L}$,_any RW_path from $st_{\text{s}}$_to_the destination_via_$st$ is longer than the reference_RW_path. $st$ is thereby removed. 3) *Unpromising_Loco-State Locking* (ULSL). If_$\text{c}_\text{c}(st_{\text{s}},st)_+ \text{c}^{\alpha}_{\min}(\gamma^{\text{v}}_{st}, \gamma^{\text{v}}_{\text{t}}) >_C$, currently it is not_likely to find any feasible RW_path via_$st$. Therefore,_PPNP pauses the search expanded from $st$. Note that $st$ cannot_be removed yet |
respect to $u$) the metric can simply be written as $${{\rm d}}s^2 = - {{\rm d}}t^2 + P^2 {{\rm d}}x^2 + Q^2 {{\rm d}}y^2 + {{\rm d}}z^2\. \label{E5}$$ This form of the homogeneous [*pp*]{} waves is suitable for physical interpretation. Considering two free test particles standing at fixed $x$, $y$ and $z$, their relative motion in the $x$-direction is given by the function $P(u)$ while it is given by $Q(u)$ in the $y$-direction. The motions are unaffected in the $z$-direction which demonstrate transversality of gravitational waves. The coordinate $u=t-z$ can now be understood as a “retarded time” and the function $d(u)$ as a “profile” of the wave. Note also that functions $P, Q$ may have a higher degree of smoothness than the function $d$ so that relative motions of particles are continuous even in the case of a shock wave (with a step-function profile, $d(u)\sim\Theta(u)$), or an impulsive wave (with a distributional profile, $d(u)\sim\delta(u)$).
**2 Standard sandwich wave**
A sandwich gravitational wave [@BPR; @BP] is constructed from the homogeneous [*pp*]{} solution (\[E1\]), (\[E2\]) if the function $d(u)$ is non-vanishing only on some finite interval of $u$, say $u\in[u_1, u_2]$. In such a case the space-time splits into three regions: a flat region $u<u_1$ (“Beforezone”), a curved region $u_1<u<u_2$ (“Wavezone”), and another flat region $u_2<u$ (“Afterzone”). In the region $u<u_1$ where $d(u)=0$ it is natural to choose solutions of Eqs. (\[E4\]) such that $P=1$ and $Q=1$ so that the metric (\[E5\]) is explicitly written in Minkowski form. The form of the metric (\[E5\]) for $u>u_1$ is then given by solutions of Eqs. (\[E4\]) where the function $P, Q$ are chosen to be continuous up | respect to $ u$) the metric can simply be written as $ $ { { \rm d}}s^2 = - { { \rm d}}t^2 + P^2 { { \rm d}}x^2 + Q^2 { { \rm d}}y^2 + { { \rm d}}z^2\. \label{E5}$$ This human body of the homogeneous [ * pp * ] { } wave is suitable for forcible interpretation. view two free test atom standing at fixed $ x$, $ y$ and $ z$, their proportional apparent motion in the $ x$-direction is given by the routine $ P(u)$ while it is given by $ Q(u)$ in the $ y$-direction. The motions are unmoved in the $ z$-direction which attest transversality of gravitational waves. The coordinate $ u = t - z$ can now be understood as a “ retarded time ” and the function $ d(u)$ as a “ visibility ” of the wave. Note also that functions $ phosphorus, Q$ may have a higher degree of smoothness than the function $ d$ so that relative gesture of particles are continuous even in the case of a shock wave (with a step - function visibility, $ d(u)\sim\Theta(u)$), or an impulsive wave (with a distributional profile, $ d(u)\sim\delta(u)$).
* * 2 Standard sandwich wave * *
A sandwich gravitational wave [ @BPR; @BP ] is constructed from the homogeneous [ * pp * ] { } solution (\[E1\ ]), (\[E2\ ]) if the function $ d(u)$ is non - vanishing only on some finite interval of $ u$, read $ u\in[u_1, u_2]$. In such a casing the space - time splits into three regions: a flat region $ u < u_1 $ (“ Beforezone ”), a curved area $ u_1 < uracil < u_2 $ (“ Wavezone ”), and another flat area $ u_2 < u$ (“ Afterzone ”). In the area $ u < u_1 $ where $ d(u)=0 $ it is natural to choose solutions of Eqs. (\[E4\ ]) such that $ P=1 $ and $ Q=1 $ so that the metric (\[E5\ ]) is explicitly write in Minkowski form. The form of the metric (\[E5\ ]) for $ u > u_1 $ is then move over by solutions of Eqs. (\[E4\ ]) where the function $ P, Q$ are chosen to be continuous up | redpect to $u$) the metric cak simply be writjeb as $${{\rk d}}s^2 = - {{\rm d}}t^2 + O^2 {{\rm d}}x^2 + Q^2 {{\rm d}}y^2 + {{\rm d}}z^2\. \labep{E5}$$ This form of the homogeneojs [*pp*]{} wavvs is suirablt for physical invsrpretabnon. Ckksidexiig two free tesj particles vtanding at fifea $r$, $y$ and $z$, their relative motion in tre $x$-dirrchion is given fy tnq fuhbtlon $P(u)$ while it is given by $Q(u)$ ih the $y$-virection. The mptions are unaffected in tje $z$-firection which delonstrate ttznsdwrsality of eravitational waves. Thg coordinate $u=t-z$ can now be undefstoob as a “retatbwd hhme” and the funcnion $d(u)$ as a “igofile” mf the eave. Note also thet fynctions $P, Q$ may have a higher degree os smoothnasa than the functiin $d$ so thad reuqtixe joviohs of oarvicles are dontinuous wven in the case of a whock wave (wifh a see[-function profile, $d(u)\sim\Theta(u)$), or an impllsibe wave (with a distriburional profile, $d(u)\sim\dglta(u)$).
**2 Stanqard sandwich wave**
A sandwich gravitational wave [@B[R; @BP] ks einstrjxtfd from the homogeneous [*pp*]{} solution (\[E1\]), (\[E2\]) if thq finbtion $d(u)$ is non-vakishing only on soke fogite interval of $u$, say $u\in[u_1, u_2]$. In such a fase thg spacw-time splyts onto three regions: a flat rwgion $u<u_1$ (“Befjeezone”), a curved reyion $u_1<u<u_2$ (“Wavzzone”), snd amother flat region $u_2<u$ (“Ayterzohe”). In the rfgion $u<u_1$ sfere $d(u)=0$ it is nagursl to choose solutions of Eqf. (\[E4\]) such vhat $'=1$ and $Q=1$ ro tnat thq metric (\[E5\]) is ew[licitly written ij Minnofski form. Hhe form of the metric (\[E5\]) for $u>u_1$ is then given bf smlutions of Eas. (\[E4\]) where thq function $P, Q$ are chjsen go be contpnuous up | respect to $u$) the metric can simply as d}}s^2 = {{\rm d}}t^2 + {{\rm + {{\rm d}}z^2\. This form of homogeneous [*pp*]{} waves is suitable for interpretation. Considering two free test particles standing at fixed $x$, $y$ and $z$, relative motion in the $x$-direction is given by the function $P(u)$ while it given $Q(u)$ the The motions are unaffected in the $z$-direction which demonstrate transversality of gravitational waves. The coordinate $u=t-z$ now be understood as a “retarded time” and function $d(u)$ as a of the wave. Note also functions Q$ may a degree smoothness than the $d$ so that relative motions of particles are continuous even in the case of a shock wave a step-function or an wave a profile, $d(u)\sim\delta(u)$). **2 wave** A sandwich gravitational wave [@BPR; from the homogeneous [*pp*]{} solution (\[E1\]), (\[E2\]) if function $d(u)$ non-vanishing only on some finite interval $u$, say $u\in[u_1, u_2]$. In such a case space-time splits into three regions: a flat region $u<u_1$ (“Beforezone”), a curved region $u_1<u<u_2$ (“Wavezone”), flat region $u_2<u$ (“Afterzone”). the region $u<u_1$ $d(u)=0$ is to solutions of (\[E4\]) such that $P=1$ and $Q=1$ so that the metric (\[E5\]) explicitly written in Minkowski form. The form of the metric $u>u_1$ then given by of Eqs. (\[E4\]) where function Q$ are chosen to up | respect to $u$) the metric can simPly be writtEn as $${{\rM d}}s^2 = - {{\Rm d}}T^2 + P^2 {{\Rm d}}x^2 + q^2 {{\rm d}}Y^2 + {{\rm d}}z^2\. \label{E5}$$ ThiS Form Of the homogeneous [*pp*]{} waveS is suItABle fOR pHysicAl interPReTATioN. COnSidErINg Two frEe tEst partIcles standIng At Fixed $x$, $y$ and $z$, tHEiR relative mOtiOn in the $x$-direCtiOn is giVeN by THe funCtiOn $P(u)$ wHile it IS given By $Q(u)$ in the $Y$-dIRectioN. the motiONS aRe unAffected in the $z$-dirECtIOn which demonstRate trAnSVeRSAliTy oF gravitatiOnAl wavES. The cooRDiNATE $u=t-Z$ Can now be underStood as a “retARdeD time” aNd The FUnctioN $d(u)$ as A “pROfiLe” of the wave. note Also that fUnctioNS $P, Q$ may hAVe a highEr degrEe oF smOothNEsS tHan ThE FunCTiOn $d$ SO thAt relatiVe MoTions Of paRTICLes aRe cOntiNuous Even in the case Of a ShocK WavE (with A step-FuncTiOn proFile, $d(u)\Sim\ThEtA(u)$), or an impulsive Wave (With a distRibUtIonAl ProfiLE, $d(u)\sim\DelTa(u)$).
**2 standarD sandwiCH waVe**
a SANdWich gravitational wAvE [@bpR; @bP] is consTructeD FrOm THe homogeNeOus [*Pp*]{} soLUTion (\[E1\]), (\[e2\]) if tHE fUnction $d(U)$ is non-VAnIsHing onlY oN some fInIte IntErval OF $u$, saY $u\in[u_1, u_2]$. in such a cAse thE Space-time splitS Into three regiONs: A FLaT RegiOn $u<U_1$ (“Beforezone”), A curVEd reGion $U_1<U<u_2$ (“wavEZone”), aNd anoThER fLAt region $u_2<u$ (“Afterzone”). in The regIon $u<u_1$ Where $d(u)=0$ it is naTural to choOSE SolutionS of EQS. (\[E4\]) SUch that $P=1$ and $Q=1$ so That tHe metric (\[E5\]) iS ExplicitLy wriTten in MiNkowski foRM. the form oF thE meTriC (\[E5\]) fOR $U>u_1$ Is then given by SOLutiOnS of Eqs. (\[E4\]) WheRe the fuNctIon $p, Q$ aRe cHoSen to be coNtinuous Up | respect to $u$) the metri c can simp ly be wr itt en as$${{ \rm d}}s^2 = - {{\r m d}}t^2 + P^2 {{\rm d }}x^2 + Q^2{ {\ rm d} }y^2 +{ {\ r m d} }z ^2 \.\l a be l{E5} $$This fo rm of thehom og eneous [*pp* ] {} waves issui table for ph ysi cal in te rpr e tatio n.Consi dering two fr ee test p ar t icless tanding a tfixe d $x$, $y$ and $z $ ,t heir relativemotion i n t h e $x $-d irection i sgiven by thef un c t i on$ P(u)$ while i t is givenb y $ Q(u)$in th e $y$-d irect io n . T he motionsareunaffecte d in t h e $z$-d i rection which de mon stra t etr ans ve r sal i ty of gra vitation al w aves. The c o o rdin ate $u= t-z$can now be un der stoo d as a “r etard ed t im e” an d thefunct io n $d(u)$ as a “ prof ile” of t hewa ve. N ote a l so tha t f unc tions $ P, Q$ m a y h av e a h igher degree of sm oo t h ne ss thanthe fu n ct io n $d$ soth atrela t i ve mo tion s o f partic les ar e c on tinuous e ven in t hecas e ofa sho ck wav e (witha ste p -function prof i le, $d(u)\sim \ Th e t a( u )$), or an impulsi ve w a ve ( with adis t ribut ional p r of i le, $d(u)\sim\delta (u )$).
**2 S tandard sandw ich wave** A sandwic h gr a vi t ational wave [ @BPR; @BP] is c o nstructe d fro m the ho mogeneous [ *pp*]{}sol uti on(\[ E 1 \] ), (\[E2\]) i f thefu nction$d( u)$ isnon -va nis hin gonly on s ome fini te i nt er val of $ u $, say $ u\ in[ u_ 1,u_2]$ . In su ch acase t he spa ce-time sp l i ts i nt othre e r eg ions: a f l atregion$u<u_1$ ( “Be f orez on e” ), a cu rved region $ u_ 1<u<u_2$ ( “W ave zone”) , and anot her flat region $u_2<u$ (“After zon e”).In t he region $u <u_1$whe r e $d(u )=0$ i t isna tur a l to c h o os e s ol utions ofE q s.(\[E4 \] ) su ch that $P=1$ and $Q=1$ s o th at the metric (\ [E5\ ] ) i s e x pl i cit ly wri t t en in Minkowski form. The f o rm of the me t ric ( \[E5\]) for $u >u_1$ is then given by solution sof E q s . ( \[E4\]) wh ere thefunction$ P, Q$ ar e cho sen to be c ont inuou s up | respect_to $u$)_the metric can simply_be written_as_$${{\rm d}}s^2_=_- {{\rm d}}t^2_+ P^2 {{\rm_d}}x^2 + Q^2 {{\rm_d}}y^2 + {{\rm_d}}z^2\._\label{E5}$$ This form of the homogeneous [*pp*]{} waves is suitable for physical interpretation. Considering_two_free test_particles_standing_at fixed $x$, $y$ and_$z$, their relative motion in_the $x$-direction_is given by the function $P(u)$ while it_is_given by $Q(u)$_in the $y$-direction. The motions are unaffected in the_$z$-direction which demonstrate transversality of gravitational_waves. The coordinate_$u=t-z$_can_now be understood as_a “retarded time” and the function_$d(u)$ as a “profile” of the_wave. Note also that functions $P, Q$_may have a higher degree of_smoothness than the function $d$_so that_relative motions of particles are_continuous even in_the case_of a shock_wave (with a step-function profile, $d(u)\sim\Theta(u)$),_or an impulsive_wave (with a distributional profile, $d(u)\sim\delta(u)$).
**2_Standard_sandwich wave**
A sandwich_gravitational_wave_[@BPR; @BP]_is constructed from_the_homogeneous [*pp*]{}_solution_(\[E1\]), (\[E2\]) if the function $d(u)$_is_non-vanishing only on some finite interval of_$u$, say $u\in[u_1, u_2]$._In_such a case the_space-time splits into three regions:_a flat region $u<u_1$ (“Beforezone”), a_curved region_$u_1<u<u_2$ (“Wavezone”),_and another flat region $u_2<u$ (“Afterzone”). In the region $u<u_1$ where_$d(u)=0$ it is natural to choose_solutions of Eqs. (\[E4\])_such that_$P=1$_and $Q=1$ so_that_the metric_(\[E5\]) is explicitly written in Minkowski form._The form_of the metric (\[E5\]) for $u>u_1$_is then given by_solutions_of Eqs. (\[E4\]) where the function_$P, Q$ are chosen to be_continuous up |
ignores the space $X$. We need a similar result in our setup.
\[lem:forget\_X\] The functor $p_X^*\colon {\mathrm{KK}}^G\to{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$ maps ${{\mathcal{CC}}},{{\mathcal{CI}}},{\langle{{\mathcal{CI}}}\rangle}
\subseteq{\mathrm{KK}}^G$ to the corresponding subcategories in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$. If $f\in{\mathrm{KK}}^G(A,B)$ is a weak equivalence or vanishes for compact subgroups, so does $p_X^*(f)$. If ${{\mathsf{D}}}\in{\mathrm{KK}}^G({{\mathsf{P}}},{{\star}})$ is a Dirac morphism for $G$, then $p_X^*({{\mathsf{D}}})\in{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}(C_0(X,{{\mathsf{P}}}),C_0(X))$ is a Dirac morphism for $G{\mathbin{\ltimes}}X$. There are natural isomorphisms $$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A,
\qquad
(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A.$$
Recall from Section \[sec:preliminaries\] that the functor $p_X^*$ is compatible with restriction and induction. This implies $p_X^*({{\mathcal{CC}}})\subseteq{{\mathcal{CC}}}$ and $p_X^*({{\mathcal{CI}}})\subseteq{{\mathcal{CI}}}$. The same holds for ${\langle{{\mathcal{CI}}}\rangle}$ because $p_X^*$ is triangulated and commutes with direct sums. This implies the assertions about weak equivalences and Dirac morphisms. Now yields $$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A
\cong (G{\mathbin{\ltimes}}X) {\mathbin{\ltimes_{\mathrm{r}}}}(p_X^*({{\mathsf{P}}}) | ignores the space $ X$. We need a similar resultant role in our apparatus.
\[lem: forget\_X\ ] The functor $ p_X^*\colon { \mathrm{KK}}^G\to{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$ maps $ { { \mathcal{CC}}},{{\mathcal{CI}}},{\langle{{\mathcal{CI}}}\rangle }
\subseteq{\mathrm{KK}}^G$ to the corresponding subcategories in $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}$. If $ f\in{\mathrm{KK}}^G(A, B)$ is a weak comparison or vanishes for compact subgroups, indeed does $ p_X^*(f)$. If $ { { \mathsf{D}}}\in{\mathrm{KK}}^G({{\mathsf{P}}},{{\star}})$ is a Dirac morphism for $ G$, then $ p_X^*({{\mathsf{D}}})\in{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}(C_0(X,{{\mathsf{P}}}),C_0(X))$ is a Dirac morphism for $ G{\mathbin{\ltimes}}X$. There be natural isomorphism $ $ (G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A,
\qquad
(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A.$$
Recall from Section \[sec: preliminaries\ ] that the functor $ p_X^*$ is compatible with limitation and induction. This implies $ p_X^*({{\mathcal{CC}}})\subseteq{{\mathcal{CC}}}$ and $ p_X^*({{\mathcal{CI}}})\subseteq{{\mathcal{CI}}}$. The like holds for $ { \langle{{\mathcal{CI}}}\rangle}$ because $ p_X^*$ is triangulated and commute with direct sums. This implies the assertions about weak equivalences and Dirac morphisms. Now succumb $ $ (G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A
\cong (G{\mathbin{\ltimes}}X) { \mathbin{\ltimes_{\mathrm{r}}}}(p_X^*({{\mathsf{P } } } ) | igjores the space $X$. We need a similar result in onr setul.
\[lem:forgdt\_X\] The functor $p_X^*\colon {\mathcm{KK}}^T\to{\maukrm{KK}}^{G{\mathbin{\ltimes}}X}$ maps ${{\matjcal{CC}}},{{\marhcao{XI}}},{\langle{{\mavgcal{CI}}}\rangle}
\dubsztxq{\mathrm{KK}}^G$ to jhe correspotding subcategmrked in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$. If $f\in{\iathrm{KL}}^G(W,B)$ is a weak ezuivsjencs or vanishes for compact subgroupa, so dots $p_X^*(f)$. If ${{\mathsf{D}}}\im{\mathrm{KK}}^G({{\mathsf{P}}},{{\star}})$ is a Dirwc morphism for $G$, tjen $p_X^*({{\mathsd{D}}})\in{\iqthrm{KK}}^{G{\mathcin{\ltimes}}X}(B_0(R,{{\mathsf{P}}}),C_0(X))$ js a Dirac morphism for $G{\mathbiv{\ltimzs}}X$. There atz nahoral isomorpiisms $$(D{\mathbin{\ltimcx}}X){\mathtin{\ltimrs^{\mathbb{L}}_{\mathrk{r}}}}A\rong G{\mathbin{\ltimes^{\mathbb{N}}_{\mathrm{r}}}}A,
\qquad
(D{\mathbin{\ldijes}}X){\mathbin{\ltimes^{\narhrm{Ots}}_{\madhrm{f}}}}Q\covg F{\methgin{\ltiles^{\jathrm{Obs}}_{\mzthrm{r}}}}A.$$
Recaol from Section \[sec:ptejpkinaries\] thaf the sugctor $p_X^*$ is compatible with restriction dnd induction. This implies $p_X^*({{\mathcal{CC}}})\subseteq{{\mwthcal{CC}}}$ wnd $p_X^*({{\mathcal{CI}}})\subseteq{{\mathcal{CI}}}$. The same holds fmr ${\laigue{{\mcbmcal{ZU}}}\rwngle}$ because $p_X^*$ is triangulated and commutes rjtn cirect sums. Thls implies the assrrhipgs about weak equivcmehces and Dirac morohisms. Gow yuelds $$(G{\mauhbin{\ktimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\nathrm{r}}}}A
\coug (T{\mathbin{\ltimes}}X) {\machbin{\ltimes_{\mcthrm{r}}}}(k_X^*({{\mathxf{P}}}) | ignores the space $X$. We need a in setup. \[lem:forget\_X\] functor $p_X^*\colon {\mathrm{KK}}^G\to{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$ corresponding in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$. If is a weak or vanishes for compact subgroups, so $p_X^*(f)$. If ${{\mathsf{D}}}\in{\mathrm{KK}}^G({{\mathsf{P}}},{{\star}})$ is a Dirac morphism for $G$, then $p_X^*({{\mathsf{D}}})\in{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}(C_0(X,{{\mathsf{P}}}),C_0(X))$ is a morphism for $G{\mathbin{\ltimes}}X$. There are natural isomorphisms $$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A, \qquad (G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A.$$ Recall Section that functor is compatible with restriction and induction. This implies $p_X^*({{\mathcal{CC}}})\subseteq{{\mathcal{CC}}}$ and $p_X^*({{\mathcal{CI}}})\subseteq{{\mathcal{CI}}}$. The same holds for ${\langle{{\mathcal{CI}}}\rangle}$ $p_X^*$ is triangulated and commutes with direct sums. implies the assertions about equivalences and Dirac morphisms. Now $$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A (G{\mathbin{\ltimes}}X) {\mathbin{\ltimes_{\mathrm{r}}}}(p_X^*({{\mathsf{P}}}) | ignores the space $X$. We need a siMilar resulT in ouR seTup.
\[LeM:forGet\_X\] the functor $p_X^*\coLOn {\maThrm{KK}}^G\to{\mathrm{KK}}^{G{\mathBin{\ltImES}}X}$ maPS ${{\mAthcaL{CC}}},{{\mathCAl{ci}}},{\LanGlE{{\mAthCaL{cI}}}\RanglE}
\suBseteq{\mAthrm{KK}}^G$ to The CoRresponding sUBcAtegories iN ${\maThrm{KK}}^{G{\mathbIn{\lTimes}}X}$. if $F\in{\MAthrm{kK}}^G(a,B)$ is a Weak eqUIvalenCe or vanisHeS For comPAct subgROUpS, so dOes $p_X^*(f)$. If ${{\mathsf{D}}}\in{\MAtHRm{KK}}^G({{\mathsf{P}}},{{\stAr}})$ is a DIrAC mORPhiSm fOr $G$, then $p_X^*({{\mAtHsf{D}}})\iN{\Mathrm{Kk}}^{g{\mATHBin{\LTimes}}X}(C_0(X,{{\mathsF{P}}}),C_0(X))$ is a DiraC MorPhism fOr $g{\maTHbin{\ltImes}}X$. thERe aRe natural isOmorPhisms $$(G{\maThbin{\lTImes}}X){\maTHbin{\ltiMes^{\matHbb{l}}_{\maThrm{R}}}}a\cOnG G{\mAtHBin{\LTiMes^{\MAthBb{L}}_{\mathrM{r}}}}a,
\qQuad
(G{\MathBIN{\LTimeS}}X){\mAthbIn{\ltiMes^{\mathrm{Obs}}_{\mAthRm{r}}}}A\COng g{\mathBin{\ltImes^{\MaThrm{OBs}}_{\mathRm{r}}}}A.$$
REcAll from Section \[sEc:prEliminariEs\] tHaT thE fUnctoR $P_X^*$ is coMpaTibLe with rEstrictIOn aNd INDUcTion. This implies $p_X^*({{\mAtHCAl{cC}}})\subsetEq{{\mathCAl{cC}}}$ ANd $p_X^*({{\mathCaL{CI}}})\SubsETEq{{\matHcal{ci}}}$. THe same hoLds for ${\LAnGlE{{\mathcaL{Ci}}}\ranglE}$ bEcaUse $P_X^*$ is tRIangUlated And commuTes wiTH direct sums. ThiS Implies the assERtIONs ABout WeaK equivalencEs anD diraC morPHiSms. nOw yieLds $$(G{\mAtHBiN{\Ltimes}}X){\mathbin{\ltimeS^{\mAthbb{L}}_{\MathrM{r}}}}A
\cong (G{\mathbIn{\ltimes}}X) {\mATHBin{\ltimeS_{\matHRm{R}}}}(P_X^*({{\mathsf{P}}}) | ignores the space $X$. We need a si milar re sul tin o ur s etup.
\[lem:f o rget \_X\] The functor $p_X ^*\co lo n {\m a th rm{KK }}^G\to { \m a t hrm {K K} }^{ G{ \ ma thbin {\l times}} X}$ maps $ {{\ ma thcal{CC}}}, { {\ mathcal{CI }}} ,{\langle{{\ mat hcal{C I} }}\ r angle }
\sub seteq{ \ mathrm {KK}}^G$to the co r respond i n gsubc ategories in ${\m a th r m{KK}}^{G{\mat hbin{\ lt i me s } }X} $.If $f\in{\ ma thrm{ K K}}^G(A , B) $ i s a weak equivale nce or vani s hes for c om pac t subgr oups, s o do es $p_X^*(f )$.If ${{\ma thsf{D } }}\in{\ m athrm{K K}}^G( {{\ mat hsf{ P }} }, {{\ st a r}} ) $isa Di rac morp hi sm for$G$, t h e n $p _X^ *({{ \math sf{D}}})\in{\ mat hrm{ K K}} ^{G{\ mathb in{\ lt imes} }X}(C_ 0(X,{ {\ mathsf{P}}}),C_ 0(X) )$ is a D ira cmor ph ism f o r $G{\ mat hbi n{\ltim es}}X$. The re a r enatural isomorphis ms $ $( G{\mathb in{\lt i me s} } X){\math bi n{\ ltim e s ^{\ma thbb { L} }_{\math rm{r}} } }A \c ong G{\ ma thbin{ \l tim es^ {\mat h bb{L }}_{\m athrm{r} }}}A, \qquad
(G{ \ mathbin{\ltim e s} } X ){ \ math bin {\ltimes^{\ math r m{Ob s}}_ { \m ath r m{r}} }}A\c on g G { \mathbin{\ltimes^{\ ma thrm{O bs}}_ {\mathrm{r}}} }A.$$
Rec a l l from Se ctio n \ [ sec:preliminar ies\] that thef unctor $ p_X^* $ is com patible w i t h restri cti onand in d u ct ion. This imp l i es $ p_ X^*({{\ mat hcal{CC }}} )\s ubs ete q{ {\mathcal {CC}}}$an d$p _X ^*( {{\ma t hcal{CI} }} )\s ub set eq{{\ m athcal {CI}} }$.Th es ame holdsf or $ {\la ng le {{\m ath ca l{CI} }}\r a ngl e}$ bec ause $p_X ^*$ is t ri an gulated and commutes w ith direct s ums . This i mplies t he assertions about wea k equiva len ces a nd D irac morp his ms. No w y ields$$(G{\ mathb in {\l t i mes}} X ) {\ mat hb in{\ltimes ^ { \ma thbb{ L} }_{\ mathrm{ r}}}}A
\cong (G{ \ mat hbin{\ltimes} }X) {\m a t hb in{ \ lt i mes _{ \ mat h r m{r}}}}(p_X^*({ {\mathsf{P }} } ) | ignores_the space $X$._We need a similar_result in_our_setup.
\[lem:forget\_X\] The_functor_$p_X^*\colon {\mathrm{KK}}^G\to{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$ maps_${{\mathcal{CC}}},{{\mathcal{CI}}},{\langle{{\mathcal{CI}}}\rangle}
\subseteq{\mathrm{KK}}^G$_to the corresponding subcategories_in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$. If_$f\in{\mathrm{KK}}^G(A,B)$_is a weak equivalence or vanishes for compact subgroups, so does $p_X^*(f)$. If ${{\mathsf{D}}}\in{\mathrm{KK}}^G({{\mathsf{P}}},{{\star}})$_is_a Dirac_morphism_for $G$,_then $p_X^*({{\mathsf{D}}})\in{\mathrm{KK}}^{G{\mathbin{\ltimes}}X}(C_0(X,{{\mathsf{P}}}),C_0(X))$ is a Dirac_morphism for $G{\mathbin{\ltimes}}X$. There are_natural isomorphisms_$$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A,
\qquad
(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A\cong G{\mathbin{\ltimes^{\mathrm{Obs}}_{\mathrm{r}}}}A.$$
Recall from_Section \[sec:preliminaries\]_that the functor $p_X^*$_is compatible with restriction and induction. This implies $p_X^*({{\mathcal{CC}}})\subseteq{{\mathcal{CC}}}$_and $p_X^*({{\mathcal{CI}}})\subseteq{{\mathcal{CI}}}$. The same holds for_${\langle{{\mathcal{CI}}}\rangle}$ because $p_X^*$ is_triangulated_and_commutes with direct sums._This implies the assertions about weak_equivalences and Dirac morphisms. Now yields_$$(G{\mathbin{\ltimes}}X){\mathbin{\ltimes^{\mathbb{L}}_{\mathrm{r}}}}A
\cong (G{\mathbin{\ltimes}}X) {\mathbin{\ltimes_{\mathrm{r}}}}(p_X^*({{\mathsf{P}}}) |
----------------------------
We use a two-step algorithm for radiative transfer. First, we evolve all simulation volumes solving the radiative transfer equation in a low angular resolution mode simultaneously with the equations of hydrodynamics. To correct for low resolution, we apply a high angular resolution transfer filter iteratively to our solutions at $z=3$ to improve on the equilibrium position of HI, HeI and HeII ionization fronts.
We developed a UVB radiative transfer module for the parallel AMR cosmology N-body Eulerian hydrodynamics code *Enzo*. The fluid flow equations are solved with the PPM scheme [@colella.84] on a comoving grid (see for @oshea......04 for details), and chemical abundances for various ionization and molecular states of hydrogen and helium are solved with a chemical reaction network with 9 species and 28 reactions [@anninos...97; @abel...97]. Each level of grid refinement has twice the spatial resolution of the previous level. We solve the radiative transfer equation with a photon conserving scheme self-consistently at each level of resolution carrying fluxes explicitly from parent grids to all of its subgrids. On each grid, radiative transfer is computed with a timestep which is usually significantly smaller than the hydro timestep and is adjusted adaptively to obtain the best balance between accuracy and the speed of calculations. Due to complexity of the setup we have not included point source radiation in our models yet, and also we limit transport of background radiation to simple sweeps along the xyz-coordinate axis. We then post-process *Enzo* output with much higher angular resolution transfer. Below we describe these two steps in detail.
Stage one: time stepping and parallelization of the base grid
-------------------------------------------------------------
By definition, radiation propagates at the speed of light. In an explicit advection numerical scheme the Courant condition would necessarily require prohibitively small timesteps to guarantee stability. However, our photon conservation technique is inherently stable, and from this standpoint there is no need to take very small timesteps, but accuracy of the solution is an entirely different issue. In many cases where the supply of ionizing photons greatly exceeds the recombination rate I-fronts can propagate at or close to the speed of light, and even in this regime our photon conservation will give an accurate solution as long as the radiation-chemistry timestep is small enough. How do we ensure that we provide timesteps small enough to balance accuracy and speed of calculations?
Let us | ----------------------------
We use a two - step algorithm for radiative transfer. First, we develop all model volumes solving the radiative transportation equation in a broken angular resolution manner simultaneously with the equations of hydrodynamics. To correct for abject resolution, we apply a gamey angular resolution transfer filter iteratively to our solution at $ z=3 $ to improve on the equilibrium position of HI, HeI and HeII ionization fronts.
We developed a UVB radiative transfer faculty for the parallel AMR cosmology N - body Eulerian hydrodynamics code * Enzo *. The fluid flow equality are solved with the PPM scheme [ @colella.84 ] on a comoving power system (see for @oshea...... 04 for detail), and chemical abundances for various ionization and molecular states of hydrogen and helium are solved with a chemical reaction net with 9 species and 28 reactions [ @anninos... 97; @abel... 97 ]. Each level of grid refinement has twice the spatial settlement of the previous level. We solve the radiative transfer equation with a photon conserving scheme self - consistently at each level of resolution carry fluxes explicitly from parent grids to all of its subgrids. On each power system, radiative transportation is computed with a timestep which is usually significantly smaller than the hydro timestep and is align adaptively to receive the best balance between accuracy and the focal ratio of calculations. Due to complexity of the setup we have not included point source radiation in our model yet, and also we limit transport of background radiation to elementary sweeps along the xyz - coordinate axis. We then post - process * Enzo * output with much high angular resolution transfer. Below we identify these two steps in detail.
Stage one: time stepping and parallelization of the nucleotide grid
-------------------------------------------------------------
By definition, radiotherapy propagates at the speed of light. In an denotative advection numerical scheme the Courant condition would necessarily require prohibitively small timesteps to undertake stability. However, our photon conservation technique is inherently static, and from this standpoint there is no need to take very small timesteps, but accuracy of the solution is an wholly different issue. In many cases where the provision of ionizing photons greatly exceeds the recombination rate I - fronts can circulate at or near to the speed of light, and even in this regimen our photon conservation will give an accurate solution as long as the radiotherapy - chemistry timestep is small enough. How do we ensure that we provide timesteps humble enough to balance accuracy and speed of calculations?
Let us | ----------------------------
We kse a two-step algorithm nor radiative trcbsfer. Hirst, ws evolve all simulation volumes solvmng rhe rqdiative transfer equagion in a low angylar eesolution mode simultansluslv xith the equatipns of hydsodynamics. To wofrzct for low resolution, we apply a hidh angukag resolution ttansftr siltsg lteratively to our solutions at $a=3$ to imkrove on the equilobrium position of HI, HeI wnd JeII ionization frlnts.
We deveoopeq a UVB radiagive transfer module fkr the parallel AMR cosmology N-cody Zulerian hyeridyjdmics code *Xnzo*. Tre fluid flow equatimns are solved with tme PPK sxheme [@colella.84] on a cokoving grid (see fot @oshea......04 fos betails), and chemical qbyndanwes xor xqriuus ipnjzatioj aid moleculad states of hydrogen and heliuk wgr solved witg a chqmycal reaction network with 9 species and 28 rsactions [@anninos...97; @abel...97]. Eqch level of grid reflnement hws twice the spatial resolution of the previous lavel. Xe soovc thd rwdiative transfer equation with a photon consqdvonn scheme self-confistently ay favr level of rerolution darrying fluxes exolicitli from parent gwids to all of its subgrids. On wach grid, rabiarive transfer is clmputed witk a tikestel which is usually signnficanfly smaller than the fydro timestep avd ps agjusted adaptively to obtayn the bewt bclance bdtwegn accuwacy and tje spcad of calculations. Due jo com[lexity of the setup we have not included point source tadhatpon in oux modeks yet, and alfo we limit trcnsport jf bazkground rzdiatioi to simple fweeps along dje xyz-coordiiate axis. We rhen post-pruzess *Enzo* outpit with mlck higher qngular resolution trxhsfer. Below we btsctube these two xteos yn dxtail.
Fdage one: tima stdppkmg ana parallelidatkon pf the base grid
-------------------------------------------------------------
By dafinjtion, radiation prppwgates aj the speqd of light. Im an explicit adveftion nuoericsl fcheme the Courant condition wkuld necedsavily require [rohlbityvely smalk timesteps to guarantee stability. Howeter, our photon conservarion technique is iuhtrently stablx, and srom this standpoint there is no need to take yery small timesteps, bht accgracy of the solution is an entirely different issue. In many cases where the sypply of ionizing lhotpns ggeacly exceeqs tie recombination rste I-fronts can propagate at or close tm che speed of light, and even im ghis regime oor photon conservation wilm give am accurate solution as long as yhe radiation-chemistry tlmestep iw smaml enolfh. How do we ensure yhat wz provude timwsteis small enougf tp nalance acfurary cnd speed of calculations?
Let us | ---------------------------- We use a two-step algorithm for First, evolve all volumes solving the low resolution mode simultaneously the equations of To correct for low resolution, we a high angular resolution transfer filter iteratively to our solutions at $z=3$ to on the equilibrium position of HI, HeI and HeII ionization fronts. We developed UVB transfer for parallel AMR cosmology N-body Eulerian hydrodynamics code *Enzo*. The fluid flow equations are solved with the scheme [@colella.84] on a comoving grid (see for for details), and chemical for various ionization and molecular of and helium solved a reaction network with species and 28 reactions [@anninos...97; @abel...97]. Each level of grid refinement has twice the spatial resolution of previous level. the radiative equation a conserving scheme self-consistently level of resolution carrying fluxes explicitly to all of its subgrids. On each grid, transfer is with a timestep which is usually smaller than the hydro timestep and is adjusted to obtain the best balance between accuracy and the speed of calculations. Due to complexity setup we have not point source radiation our yet, also limit transport background radiation to simple sweeps along the xyz-coordinate axis. We then *Enzo* output with much higher angular resolution transfer. Below we two in detail. Stage time stepping and parallelization the grid ------------------------------------------------------------- By definition, at speed an advection scheme the Courant condition necessarily require prohibitively small timesteps guarantee stability. However, our stable, and from this standpoint there is no to take very small timesteps, but accuracy the solution is an entirely different issue. In many cases where the of ionizing exceeds the recombination rate I-fronts can propagate at close to the speed light, and even in this regime our photon conservation give accurate solution long as the timestep is small How do we we provide small to speed of calculations? Let us | ----------------------------
We use a two-step algorithm for Radiative tRansfEr. FIrsT, wE evoLve aLl simulation voLUmes Solving the radiative traNsfer EqUAtioN In A low aNgular rESoLUTioN mOdE siMuLTaNeousLy wIth the eQuations of HydRoDynamics. To coRReCt for low reSolUtion, we apply A hiGh anguLaR reSOlutiOn tRansfEr filtER iteraTively to oUr SOlutioNS at $z=3$ to iMPRoVe on The equilibrium posITiON of HI, HeI and HeIi ionizAtIOn FROntS.
We Developed a uVb radiATive traNSfER MOduLE for the parallEl AMR cosmolOGy N-Body EuLeRiaN HydrodYnamiCs COde *enzo*. The fluiD floW equationS are soLVed with THe PPM scHeme [@coLelLa.84] oN a coMOvInG grId (SEe fOR @oSheA......04 For Details), aNd ChEmicaL abuNDANCes fOr vArioUs ionIzation and molEcuLar sTAteS of hyDrogeN and HeLium aRe solvEd witH a Chemical reactioN netWork with 9 sPecIeS anD 28 rEactiONs [@anniNos...97; @AbeL...97]. Each leVel of grID reFiNEMEnT has twice the spatiaL rESOlUtion of tHe prevIOuS lEVel. We solVe The RadiATIve trAnsfER eQuation wIth a phOToN cOnserviNg Scheme SeLf-cOnsIstenTLy at Each leVel of resOlutiON carrying fluxeS Explicitly froM PaRENt GRids To aLl of its subgRids. oN eacH griD, RaDiaTIve trAnsfeR iS CoMPuted with a timestep wHiCh is usUally Significantly Smaller thaN THE hydro tiMestEP aND is adjusted adaPtiveLy to obtain THe best baLance Between aCcuracy anD THe speed oF caLcuLatIonS. dUe To complexity oF THe seTuP we have Not IncludeD poInt SouRce RaDiation in Our modelS yEt, AnD aLso We limIT transpoRt Of bAcKgrOund rADiatioN to siMple SwEePS alOng the xYZ-cOORdinAtE aXis. WE thEn Post-pRoceSS *EnZo* outpuT with much HigHEr anGuLaR resoluTion transfer. BElOw we descriBe TheSe two sTEPs in detaIl.
Stage one: time stepping aND parallEliZatioN of tHe base griD
-------------------------------------------------------------
By DefiniTioN, RadiatIon proPagatEs At tHE Speed OF LiGht. in An explicit ADVecTion nUmEricAl schemE the Courant conditiON woUld necessarilY reQuirE PRoHibITiVEly SmALl tIMEsteps to guarantEe stabilitY. HOWeVer, our photON coNsErvatioN techniQue is INherentLy stable, aNd from thiS sTandPOInt There is no nEed to takE very smalL TimesTEpS, but aCcuRacy of ThE soLutioN is an eNTirEly diFferenT iSsue. In Many cAsEs where tHe supply of ionizing photoNs greaTly exCeeDs the recoMbiNAtiOn rate I-frOnts Can propagaTe aT or Close To tHE speeD of lIGhT, anD Even iN thiS Regime our PHoTon CONsErvation wilL GIVe aN accuRatE SolutiOn as Long as the radiatioN-Chemistry timesTep iS SMalL enOUgh. HOw Do we ensure that We pRoVIDe timestEpS small enougH to balanCe ACcuraCy and sPeed of CalculaTIOnS?
let us | -------------------------- --
We use a tw o-s tep a lgor ithm for radiative tran sfer. First, we evolve allsi m ulat i on volu mes sol v in g the r ad iat iv e t ransf erequatio n in a low an gu lar resoluti o nmode simul tan eously withthe equat io nso f hyd rod ynami cs. To correc t for low r e soluti o n, we a p p ly a h igh angular resol u ti o n transfer fil ter it er a ti v e lytoour soluti on s at$ z=3$ to im p r o veo n the equilib rium positi o n o f HI,He I a n d HeII ioni za t ion fronts.
W e de veloped a UVB r a diative transfe r modu lefor the pa ra lle lA MRc os mol o gyN-body E ul er ian h ydro d y n a mics co de * Enzo* . The fluid f low equ a tio ns ar e sol vedwi th th e PPMschem e[@colella.84] o n acomovinggri d(se efor @ o shea.. ... .04 for de tails), and c h e m ic al abundances forva r i ou s ioniza tion a n dmo l ecular s ta tes ofh y droge n an d h elium ar e solv e dwi th a ch em ical r ea cti onnetwo r k wi th 9 s pecies a nd 28 reactions [@an n inos...97; @a b el . . .9 7 ]. E ach level of g ridr efin emen t h ast wicethe s pa t ia l resolution of thepr evious leve l. We solve t he radiati v e transfer equ a ti o n with a photo n con serving sc h eme self -cons istently at eachl e vel of r eso lut ion ca r r yi ng fluxes exp l i citl yfrom pa ren t grids to al l o f i ts subgrids . On eac hgr id ,rad iativ e transfe risco mpu ted w i th a t imest ep w hi ch isusually si g n ific an tl y sm all er than the hyd ro time step andisa djus te dadaptiv ely to obtain t he best ba la nce betwe e n accurac y and the speed of calc u lations . D ue to com plexity o f t he set upw e have not i nclud ed po i n t sou r c erad ia tion in ou r mod els y et , an d alsowe limit transport ofbackground ra dia tion t osim p le swe ep s al o n g the xyz-coord inate axis .W ethen post- p roc es s *Enzo * outpu t wit h much h igher ang ular reso lu tion t ran sfer. Belo w we des cribe the s e two st eps i n d etail.
Sta ge on e: tim e st eppin g andpa rallel izati on of thebase grid
------------- ------ ----- --- --------- --- - --- --------- ---- -----
Bydef ini tion, ra d iatio n pr o pa gat e s atthes peed of l i gh t.I n a n explicita d v ect ion n ume r ical s chem e the Courant con d ition would ne cess a r ily re q uire p rohibitively s mal lt i mestepsto guaranteestabilit y. Howev er, ou r phot on cons e r va t ion te chni que is inher ent ly stable, a nd from t hisst andpoi nt the r e is n o need to take v ery s m a ll ti m est eps,bu t accur a cy o f the solu tion is anentire ly d iffer ent iss ue . In m any c ases where the suppl y ofionizin gphot ons great ly e x c eedsthere com binationr a te I- fr o nts can prop ag ateat or clo s e to the sp e ed of l ig ht, a nd eve n i n this regim e o ur ph o t on conserv a tion wi l l giv e an a ccurat e solut i onas long a s t h e radiatio n-chemist r y t im este p is sma ll eno ug h.Ho w do we ensurethat weprovidet i mestep s s m alle noughto balance acc uracy a n d speed o f c alculat i ons ?
Let us | ----------------------------
We use_a two-step_algorithm for radiative transfer._First, we_evolve_all simulation_volumes_solving the radiative_transfer equation in_a low angular resolution_mode simultaneously with_the_equations of hydrodynamics. To correct for low resolution, we apply a high angular resolution_transfer_filter iteratively_to_our_solutions at $z=3$ to improve_on the equilibrium position of_HI, HeI_and HeII ionization fronts.
We developed a UVB radiative_transfer_module for the_parallel AMR cosmology N-body Eulerian hydrodynamics code *Enzo*. The_fluid flow equations are solved with_the PPM scheme_[@colella.84]_on_a comoving grid (see_for @oshea......04 for details), and chemical_abundances for various ionization and molecular_states of hydrogen and helium are solved_with a chemical reaction network with_9 species and 28 reactions_[@anninos...97; @abel...97]._Each level of grid refinement_has twice the_spatial resolution_of the previous_level. We solve the radiative transfer_equation with a_photon conserving scheme self-consistently at each_level_of resolution carrying_fluxes_explicitly_from parent_grids to all_of_its subgrids._On_each grid, radiative transfer is computed_with_a timestep which is usually significantly smaller_than the hydro timestep_and_is adjusted adaptively to_obtain the best balance between_accuracy and the speed of calculations._Due to_complexity of_the setup we have not included point source radiation in our_models yet, and also we limit_transport of background radiation_to simple_sweeps_along the xyz-coordinate_axis._We then_post-process *Enzo* output with much higher angular_resolution transfer._Below we describe these two steps_in detail.
Stage one: time_stepping_and parallelization of the base grid
-------------------------------------------------------------
By_definition, radiation propagates at the speed_of light. In an explicit_advection_numerical_scheme the Courant condition would_necessarily require prohibitively small timesteps to_guarantee stability. However,_our photon conservation technique is inherently stable,_and_from this standpoint there is no_need_to take very small timesteps, but_accuracy_of_the solution is an entirely_different issue. In many cases where_the supply of ionizing photons greatly exceeds the recombination_rate I-fronts can_propagate at or close to_the_speed_of light, and even in this regime our photon conservation_will give_an accurate solution_as long as the radiation-chemistry timestep is small enough. How_do we ensure that we provide timesteps_small enough to balance accuracy and speed of calculations?
Let us |
{\Pi}(\bar{\theta}; {\mathcal{S}}_{k,d}).\end{aligned}$$
Using, the statement of the can be alternatively represented as $$\begin{aligned}
\label{dual-norm-characterization}
({\|\beta\|}_{k\square d}^\star)^2
=({\|\bar{\beta}\|}_{k\square d}^\star)^2
= \sup_{A\in {\overline{\texttt{BD}}}(k,d)} \, \bar{\beta}^{{\sf T}}A \bar{\beta}\end{aligned}$$ where $\bar{\beta}$ is nonnegative and non-increasing, and ${\overline{\texttt{BD}}}(k,d)$ is the set of block diagonal matrices with $d$ blocks exactly covering the first $k$ rows and columns and zero elsewhere, where on each block of size $q$, all of the entries are equal to $\frac{1}{q}$. Note that if the input is not a sorted nonnegative vector, then we need to consider ${\texttt{BD}}(k,d)\equiv\{PAP^{{\sf T}}:P\in \mathcal{P}_\pm,A\in{\overline{\texttt{BD}}}(k,d)\}$, where $\mathcal{P}_\pm $ is the set of signed permutation matrices. This brings us to $$\begin{aligned}
\label{dual-norm-var}
({\|\beta\|}_{k\square d}^\star)^2
= \sup_{A\in {\texttt{BD}}(k,d)} \, \beta^{{\sf T}}A \beta.\end{aligned}$$ The aforementioned representations, in,, and, all depend on an efficient characterization of combinatorial sets such as ${\bar{\texttt{P}}}(k,d)$ or ${\texttt{BD}}(k,d)$. below shows that ${\texttt{BD}}(k,d)$ is of exponential size, which renders direct optimization inefficient.
\[lem:BD-size\] $|{\texttt{BD}}(k,d)| < (\frac{2epd}{k})^k$.
is proved in.
Next, we review a dynamic programming approach to reformulate the above in terms of a quadratic program.
### A Dynamic Program and a QCQP Representation {#sec:qcqp}
Consider a non-negative sorted vector $\bar{\beta}$ with $\bar{\beta}_1\geq \cdots \geq \bar{\beta | { \Pi}(\bar{\theta }; { \mathcal{S}}_{k, d}).\end{aligned}$$
Using, the statement of the can be alternatively represented as $ $ \begin{aligned }
\label{dual - norm - word picture }
({ \|\beta\|}_{k\square d}^\star)^2
=( { \|\bar{\beta}\|}_{k\square d}^\star)^2
= \sup_{A\in { \overline{\texttt{BD}}}(k, d) } \, \bar{\beta}^{{\sf T}}A \bar{\beta}\end{aligned}$$ where $ \bar{\beta}$ is nonnegative and non - increasing, and $ { \overline{\texttt{BD}}}(k, d)$ is the hardening of block diagonal matrices with $ d$ block exactly covering the beginning $ k$ course and columns and zero elsewhere, where on each pulley of size $ q$, all of the entries are equal to $ \frac{1}{q}$. Note that if the stimulation is not a sorted nonnegative vector, then we need to regard $ { \texttt{BD}}(k, d)\equiv\{PAP^{{\sf T}}:P\in \mathcal{P}_\pm, A\in{\overline{\texttt{BD}}}(k, d)\}$, where $ \mathcal{P}_\pm $ is the set of sign permutation matrices. This fetch us to $ $ \begin{aligned }
\label{dual - norm - var }
({ \|\beta\|}_{k\square d}^\star)^2
= \sup_{A\in { \texttt{BD}}(k, d) } \, \beta^{{\sf T}}A \beta.\end{aligned}$$ The aforementioned representation, in, , and, all depend on an efficient characterization of combinatorial sets such as $ { \bar{\texttt{P}}}(k, d)$ or $ { \texttt{BD}}(k, d)$. downstairs shows that $ { \texttt{BD}}(k, d)$ is of exponential size, which renders direct optimization inefficient.
\[lem: BD - size\ ] $ |{\texttt{BD}}(k, d)| < (\frac{2epd}{k})^k$.
is proved in.
Next, we review a active programming approach to reformulate the above in terms of a quadratic program.
# # # A Dynamic Program and a QCQP Representation { # sec: qcqp }
regard a non - minus sorted vector $ \bar{\beta}$ with $ \bar{\beta}_1\geq \cdots \geq \bar{\beta | {\Pi}(\har{\theta}; {\mathcal{S}}_{k,d}).\end{allgned}$$
Using, the sjarement of ths can be alternatively represented ad $$\vegin{qligned}
\label{dual-norm-chxracterizwtion}
({\|\bete\|}_{k\square d}^\star)^2
=({\|\bar{\beta}\|}_{k\squadc d}^\stcr)^2
= \sup_{A\in {\oyerline{\textdt{BD}}}(k,d)} \, \bar{\betd}^{{\sw C}}A \bar{\beta}\end{aligned}$$ where $\bar{\beta}$ if nonnebahive and non-insreaxyng, znd ${\overline{\texttt{BD}}}(k,d)$ is the set kf blocn diagonal mayrices with $d$ blocks exactpy clvering the first $n$ rows and xoluibs and zero dlsewhere, where on eacg block of size $q$, all of the engries are equal ti $\frww{1}{q}$. Note thav if tre input is kpt a smrted npnnegative vecbor, tien qe need to consider ${\txxttt{BD}}(k,d)\equiv\{PAP^{{\sf J}}:P\in \mathcdl{'}_\pm,A\in{\overline{\texttt{BE}}}(k,e)\}$, whete $\madhcau{P}_\pm $ ia vhe set ov smgned permufation matruces. This brings us tj $$\begin{aligned}
\mabel{dtaj-norm-var}
({\|\beta\|}_{k\square d}^\star)^2
= \sup_{A\it {\tsxttt{BD}}(k,d)} \, \beta^{{\sf T}}A \bera.\end{aligned}$$ The aforgmentioned representations, in,, and, all depend on an efficiend chacaztexldatiub lf combinatorial sets such as ${\bar{\texttt{P}}}(k,d)$ or ${\fectnt{BD}}(k,d)$. below showf that ${\textyt{HD}}(l,q)$ is of expongntial sizs, which renders digect opjimizarion inefsicirnt.
\[lem:BD-size\] $|{\texttt{BD}}(k,d)| < (\frqc{2epd}{k})^k$.
is prjced in.
Next, we revizw a dynamic probrammong approach to reformuuate the above ln terms kw a quadratic prugrsm.
### A Dynamic Program and a QSQP Reprewentction {#sez:qcqk}
Considqr a non-nehativc sorted vector $\bar{\heta}$ cith $\tar{\beta}_1\geq \cdots \geq \bar{\beta | {\Pi}(\bar{\theta}; {\mathcal{S}}_{k,d}).\end{aligned}$$ Using, the statement of the alternatively as $$\begin{aligned} ({\|\beta\|}_{k\square d}^\star)^2 =({\|\bar{\beta}\|}_{k\square \bar{\beta}^{{\sf \bar{\beta}\end{aligned}$$ where $\bar{\beta}$ nonnegative and non-increasing, ${\overline{\texttt{BD}}}(k,d)$ is the set of block matrices with $d$ blocks exactly covering the first $k$ rows and columns and elsewhere, where on each block of size $q$, all of the entries are to Note if input is not a sorted nonnegative vector, then we need to consider ${\texttt{BD}}(k,d)\equiv\{PAP^{{\sf T}}:P\in \mathcal{P}_\pm,A\in{\overline{\texttt{BD}}}(k,d)\}$, where $ is the set of signed permutation matrices. brings us to $$\begin{aligned} ({\|\beta\|}_{k\square d}^\star)^2 = \sup_{A\in {\texttt{BD}}(k,d)} \beta^{{\sf \beta.\end{aligned}$$ The representations, and, depend on an characterization of combinatorial sets such as ${\bar{\texttt{P}}}(k,d)$ or ${\texttt{BD}}(k,d)$. below shows that ${\texttt{BD}}(k,d)$ is of exponential size, renders direct \[lem:BD-size\] $|{\texttt{BD}}(k,d)| (\frac{2epd}{k})^k$. proved Next, we review programming approach to reformulate the above a quadratic program. ### A Dynamic Program and QCQP Representation Consider a non-negative sorted vector $\bar{\beta}$ $\bar{\beta}_1\geq \cdots \geq \bar{\beta | {\Pi}(\bar{\theta}; {\mathcal{S}}_{k,d}).\end{alIgned}$$
Using, The stAteMenT oF the Can bE alternatively REpreSented as $$\begin{aligned}
\laBel{duAl-NOrm-cHArActerIzation}
({\|\BEtA\|}_{K\SquArE d}^\StaR)^2
=({\|\bAR{\bEta}\|}_{k\sQuaRe d}^\star)^2
= \Sup_{A\in {\overLinE{\tExttt{BD}}}(k,d)} \, \bar{\BEtA}^{{\sf T}}A \bar{\beTa}\eNd{aligned}$$ wheRe $\bAr{\beta}$ Is NonNEgatiVe aNd non-IncreaSIng, and ${\Overline{\tExTTt{BD}}}(k,d)$ IS the set OF BlOck dIagonal matrices wiTH $d$ BLocks exactly coVering ThE FiRST $k$ rOws And columns AnD zero ELsewherE, WhERE On eACh block of size $Q$, all of the enTRieS are eqUaL to $\FRac{1}{q}$. NoTe thaT iF The Input is not a SortEd nonnegaTive veCTor, then WE need to ConsidEr ${\tExtTt{BD}}(K,D)\eQuIv\{PaP^{{\SF T}}:P\IN \mAthCAl{P}_\Pm,A\in{\oveRlInE{\textTt{BD}}}(K,D)\}$, WHEre $\mAthCal{P}_\Pm $ is tHe set of signed PerMutaTIon MatriCes. ThIs brInGs us tO $$\begin{AlignEd}
\Label{dual-norm-vaR}
({\|\betA\|}_{k\square d}^\StaR)^2
= \sUp_{A\In {\TexttT{bD}}(k,d)} \, \beTa^{{\sF T}}A \Beta.\end{Aligned}$$ tHe aFoREMEnTioned representatiOnS, IN,, aNd, all depEnd on aN EfFiCIent charAcTerIzatION of coMbinAToRial sets Such as ${\BAr{\TeXttt{P}}}(k,d)$ Or ${\Texttt{bD}}(K,d)$. bEloW showS That ${\Texttt{bD}}(k,d)$ is of ExponENtial size, which REnders direct oPTiMIZaTIon iNefFicient.
\[lem:Bd-sizE\] $|{\TextTt{BD}}(K,D)| < (\fRac{2EPd}{k})^k$.
iS provEd IN.
NEXt, we review a dynamic pRoGrammiNg appRoach to reformUlate the abOVE In terms oF a quADrATic program.
### A DynAmic PRogram and a qcQP RepreSentaTion {#sec:qCqp}
ConsidER A non-negaTivE soRteD veCTOr $\Bar{\beta}$ with $\baR{\BEta}_1\gEq \Cdots \geQ \baR{\beta | {\Pi}(\bar{\theta}; {\mat hcal{S}}_{ k,d}) .\e nd{ al igne d}$$
Using, the s t atem ent of the can be alte rnati ve l y re p re sente d as $$ \ be g i n{a li gn ed}
\ l ab el{du al- norm-ch aracteriza tio n}
({\|\be t a\ |}_{k\squa red}^\star)^2 =({\| \b ar{ \ beta} \|} _{k\s quared }^\sta r)^2
= \sup_{ A \in {\o v e rl ine{ \texttt{BD}}}(k,d ) }\ , \bar{\beta}^ {{\sfT} } A\ b ar{ \be ta}\end{al ig ned}$ $ where$ \b a r { \be t a}$ is nonneg ative and n o n-i ncreas in g,a nd ${\ overl in e {\t exttt{BD}}} (k,d )$ is the set o f blockd iagonal matri ces wi th $ d $bl ock se xac t ly co v eri ng the f ir st $k$rows a n d col umn s an d zer o elsewhere,whe re o n ea ch bl ock o f si ze $q$, all o f the e ntries are equa l to $\frac{1 }{q }$ . N ot e tha t if th e i npu t is no t a sor t edno n n e ga tive vector, thenwe n ee d to con sider$ {\ te x ttt{BD}} (k ,d) \equ i v \{PAP ^{{\ s fT}}:P\in \math c al {P }_\pm,A \i n{\ove rl ine {\t exttt { BD}} }(k,d) \}$, whe re $\ m athcal{P}_\pm$ is the set o f s i g ne d per mut ation matri ces. This bri n gs us to $$ \begi n{ a li g ned}
\label{dual-no rm -var} ( {\|\beta\|}_{ k\square d } ^ \ star)^2 = \ s up_{A\in {\tex ttt{B D}}(k,d)}\ , \beta^ {{\sf T}}A \b eta.\end{ a l igned}$$ Th e a for eme n t io ned represent a t ions ,in,, an d,all dep end on an ef fi cient cha racteriz at io nof co mbina t orial se ts su ch as ${\b a r{\tex ttt{P }}}( k, d) $ or ${\tex t tt { B D}}( k, d) $. b elo wshows tha t ${ \texttt {BD}}(k,d )$i s of e xp onentia l size, which r enders dir ec t o ptimiz a t ion inef ficient.
\[lem:BD-size \ ] $|{\t ext tt{BD }}(k ,d)| < (\ fra c{2epd }{k } )^k$.
is pr ovedin .
N e xt, w e re vie wa dynamicp r ogr ammin gappr oach to reformulate the a b ove in terms ofa q uadr a t ic pr o gr a m.
# # # A D ynamic Programand a QCQP R e pr esentation {#s ec :qcqp}
Consid er an on-nega tive sort ed vector $ \bar { \ bet a}$ with $ \bar{\be ta}_1\geq \cdot s \ geq \ bar {\beta | {\Pi}(\bar{\theta};_{\mathcal{S}}_{k,d}).\end{aligned}$$
Using, the_statement of the can_be alternatively_represented_as $$\begin{aligned}
\label{dual-norm-characterization}
__ ({\|\beta\|}_{k\square_d}^\star)^2
_ =({\|\bar{\beta}\|}_{k\square d}^\star)^2
_ _=_\sup_{A\in {\overline{\texttt{BD}}}(k,d)} \, \bar{\beta}^{{\sf T}}A \bar{\beta}\end{aligned}$$ where $\bar{\beta}$ is nonnegative and non-increasing, and ${\overline{\texttt{BD}}}(k,d)$_is_the set_of_block_diagonal matrices with $d$ blocks_exactly covering the first $k$_rows and_columns and zero elsewhere, where on each block_of_size $q$, all_of the entries are equal to $\frac{1}{q}$. Note that if_the input is not a sorted_nonnegative vector, then_we_need_to consider ${\texttt{BD}}(k,d)\equiv\{PAP^{{\sf T}}:P\in_\mathcal{P}_\pm,A\in{\overline{\texttt{BD}}}(k,d)\}$, where $\mathcal{P}_\pm $ is the_set of signed permutation matrices. This_brings us to $$\begin{aligned}
\label{dual-norm-var}
_({\|\beta\|}_{k\square d}^\star)^2
=_\sup_{A\in {\texttt{BD}}(k,d)} \, \beta^{{\sf T}}A_\beta.\end{aligned}$$ The_aforementioned representations, in,, and, all_depend on an_efficient characterization_of combinatorial sets_such as ${\bar{\texttt{P}}}(k,d)$ or ${\texttt{BD}}(k,d)$. below_shows that ${\texttt{BD}}(k,d)$_is of exponential size, which renders_direct_optimization inefficient.
\[lem:BD-size\] $|{\texttt{BD}}(k,d)|_<_(\frac{2epd}{k})^k$.
is_proved in.
Next,_we review a_dynamic_programming approach_to_reformulate the above in terms of_a_quadratic program.
### A Dynamic Program and a_QCQP Representation {#sec:qcqp}
Consider a_non-negative_sorted vector $\bar{\beta}$ with_$\bar{\beta}_1\geq \cdots \geq \bar{\beta |
distribution, that is, in the large-$k$ behavior of the probabilities ${{\mathbb P}}^{y}\{N \geq
k\}$.
\[theorem.killed-number.weak\] If the offspring distribution has mean $1$, positive variance $\sigma^2$, and finite third moment then $$\label{eqn.killed-number.weak}
\sum_{k=1}^{m} k{{\mathbb P}}(N\geq k) \sim 2C_{7}y\sqrt{m} \quad
\text{ where } \;\; C_7=\frac{\sigma}{\sqrt{6\pi}}.$$
The proof, which will use a form of Karamata’s Tauberian theorem, will be given in section \[ssec.killed-number.weak\].
If we knew that the sequence $k{{\mathbb P}}^{y} (N\geq k)$ were monotone then we could conclude from that ${{\mathbb P}}^{y}
(N\geq k)\sim C_{7}y/k^{3/2}$. However, it seems unlikely that monotonicity of the sequence $k{{\mathbb P}}^{y} (N\geq k)$ holds in general. Thus, to obtain sharp asymptotic results about the individual probabilities ${\mathbb{P}}^{y} (N=k)$, we will impose more restrictive hypotheses on the offspring distribution that will allow us to avoid the use of Karamata’s theorem. In its place, we will use a result of Flajolet and Odlyzko [@FO90] that allows one to extract information about the asymptotic behavior of the coefficients of a power series from information about its behavior on the circle $\Gamma$ of convergence. Our hypotheses are most conveniently formulated in terms of the functions $$\label{eqn.h.kappa}
h (s)=2[\Psi (1-s)- (1-s)] \quad \text{and} \quad
\kappa (s) =\int_{0}^{s} h (s')\,ds',$$ where $\Psi$ is the probability generating function of the offspring distribution. Since the power series for a probability generating function has radius of convergence $1$, the function $h$ extends to an analytic function $h (z)$ in the disk of radius $1$ centered at $z=1$, as does its integral $\kappa$. If the offspring distribution has finite support then the functions $h$ and $\kappa$ are polynomials | distribution, that is, in the large-$k$ behavior of the probabilities $ { { \mathbb P}}^{y}\{N \geq
k\}$.
\[theorem.killed - number.weak\ ] If the offspring distribution has mean $ 1 $, incontrovertible discrepancy $ \sigma^2 $, and finite third moment then $ $ \label{eqn.killed - number.weak }
\sum_{k=1}^{m } k{{\mathbb P}}(N\geq k) \sim 2C_{7}y\sqrt{m } \quad
\text { where } \;\; C_7=\frac{\sigma}{\sqrt{6\pi}}.$$
The proof, which will use a shape of Karamata ’s Tauberian theorem, will be given in section \[ssec.killed - number.weak\ ].
If we know that the sequence $ k{{\mathbb P}}^{y } (N\geq k)$ were monotone then we could reason from that $ { { \mathbb P}}^{y }
(N\geq k)\sim C_{7}y / k^{3/2}$. However, it seem unlikely that monotonicity of the succession $ k{{\mathbb P}}^{y } (N\geq k)$ holds in general. Thus, to obtain astute asymptotic results about the individual probability $ { \mathbb{P}}^{y } (N = k)$, we will impose more restrictive hypotheses on the young distribution that will allow us to avoid the use of Karamata ’s theorem. In its stead, we will use a result of Flajolet and Odlyzko [ @FO90 ] that give up one to distill information about the asymptotic demeanor of the coefficients of a power series from information about its demeanor on the circle $ \Gamma$ of convergence. Our hypotheses are most conveniently formulated in terms of the functions $ $ \label{eqn.h.kappa }
h (s)=2[\Psi (1 - s)- (1 - s) ] \quad \text{and } \quad
\kappa (s) = \int_{0}^{s } h (s')\,ds',$$ where $ \Psi$ is the probability generating function of the offspring distribution. Since the power series for a probability generating function has spoke of overlap $ 1 $, the function $ h$ extends to an analytic function $ planck's constant (z)$ in the disk of radius $ 1 $ centered at $ z=1 $, as suffice its integral $ \kappa$. If the offspring distribution has finite support then the functions $ h$ and $ \kappa$ are polynomial | didtribution, that is, in tht large-$k$ behavior of the probagilities ${{\mathbb P}}^{y}\{N \geq
k\}$.
\[theorem.killev-numver.weqk\] If the offspring dirtributioj has meqn $1$, kositive variance $\sigma^2$, akb finjbe thnrv moment then $$\lsbel{eqn.kilned-number.weak}
\sbm_{k=1}^{m} k{{\mathbb P}}(N\geq k) \sim 2C_{7}y\sqrt{m} \quaq
\tect{ where } \;\; C_7=\frac{\figms}{\fqrt{6\lp}}.$$
Tme proof, which will use a form or Karameta’s Tauberian yheorem, will be given in sfctiln \[ssec.killed-number.aeak\].
If we kbew eyat the sequdnce $k{{\mathbb P}}^{y} (N\geq k)$ were monotone then we could covcludz from that ${{\mqthht P}}^{y}
(N\geq k)\smm C_{7}y/k^{3/2}$. However, it seems unnikely yhat monotoniclty oh thw sequence $k{{\mathbb P}}^{y} (N\geq k)$ holds in ggneral. Thuv, co obtain sharp asymproric rgsultv abuyt ghe iidibidual prkbabilitiea ${\mathbb{P}}^{y} (B=k)$, we will impose mprq restrictive gypothqsqs on the offspring distribution that wpll zllow us to avoid the uwe of Karamata’s theorgm. In its [lace, we will use a result of Flajolet and Odlyzkm [@FO90] vhxt colows inf to extract information about the asymptotic genanior of the coefflcients of a power sfrogs from informxtion cgoht its behavior on the citcle $\Gqmma$ of cjnvetgence. Our hypotheses are mist convenieutlt formulated in texms of the fbnctioms $$\lanel{eqn.h.kappa}
h (s)=2[\Psi (1-r)- (1-s)] \suad \text{anf} \quad
\kappa (s) =\int_{0}^{s} h (r')\,ds',$$ wvere $\Psi$ is the probabilitr generatmng fbnction uf tne offfpring dishribubhon. Since the poweg sernes fmr a probahility generating function has cedius of convgrgancv $1$, the fuuction $h$ extends to an analytic fonction $h (z)$ in the disk kf radins $1$ centered at $z=1$, as does lts integral $\kappa$. Is thw ofdspring aistribution hss finite support tyen the functions $m$ and $\iappa$ are polynimiqls | distribution, that is, in the large-$k$ behavior probabilities P}}^{y}\{N \geq \[theorem.killed-number.weak\] If the positive $\sigma^2$, and finite moment then $$\label{eqn.killed-number.weak} k{{\mathbb P}}(N\geq k) \sim 2C_{7}y\sqrt{m} \quad where } \;\; C_7=\frac{\sigma}{\sqrt{6\pi}}.$$ The proof, which will use a form of Karamata’s theorem, will be given in section \[ssec.killed-number.weak\]. If we knew that the sequence P}}^{y} k)$ monotone we could conclude from that ${{\mathbb P}}^{y} (N\geq k)\sim C_{7}y/k^{3/2}$. However, it seems unlikely that monotonicity the sequence $k{{\mathbb P}}^{y} (N\geq k)$ holds in Thus, to obtain sharp results about the individual probabilities (N=k)$, will impose restrictive on offspring distribution that allow us to avoid the use of Karamata’s theorem. In its place, we will use a result Flajolet and that allows to information the asymptotic behavior coefficients of a power series from behavior on the circle $\Gamma$ of convergence. Our are most formulated in terms of the functions h (s)=2[\Psi (1-s)- (1-s)] \quad \text{and} \quad \kappa =\int_{0}^{s} h (s')\,ds',$$ where $\Psi$ is the probability generating function of the offspring distribution. Since series for a probability function has radius convergence the $h$ to an function $h (z)$ in the disk of radius $1$ centered at as does its integral $\kappa$. If the offspring distribution has then functions $h$ and are polynomials | distribution, that is, in the laRge-$k$ behaviOr of tHe pRobAbIlitIes ${{\mAthbb P}}^{y}\{N \geq
k\}$.
\[thEOrem.Killed-number.weak\] If the oFfsprInG DistRIbUtion Has mean $1$, POsITIve VaRiAncE $\sIGmA^2$, and fIniTe third Moment then $$\LabEl{Eqn.killed-numBEr.Weak}
\sum_{k=1}^{m} k{{\MatHbb P}}(N\geq k) \sim 2c_{7}y\sQrt{m} \quAd
\TexT{ Where } \;\; c_7=\frAc{\sigMa}{\sqrt{6\PI}}.$$
The prOof, which wIlL Use a foRM of KaraMATa’S TauBerian theorem, will BE gIVen in section \[ssEc.killEd-NUmBER.weAk\].
IF we knew thaT tHe seqUEnce $k{{\maTHbB p}}^{Y} (n\geQ K)$ were monotone Then we could COncLude frOm ThaT ${{\Mathbb p}}^{y}
(N\geQ k)\SIm C_{7}Y/k^{3/2}$. However, it SeemS unlikely That moNOtoniciTY of the sEquencE $k{{\mAthBb P}}^{y} (n\GeQ k)$ HolDs IN geNErAl. THUs, tO obtain sHaRp AsympTotiC RESUlts AboUt thE indiVidual probabiLitIes ${\mAThbB{P}}^{y} (N=k)$, We wilL impOsE more RestriCtive HyPotheses on the ofFsprIng distriButIoN thAt Will aLLow us tO avOid The use oF KaramaTA’s tHeOREM. IN its place, we will use A rESUlT of FlajoLet and oDlYzKO [@FO90] that aLlOws One tO EXtracT infORmAtion aboUt the aSYmPtOtic behAvIor of tHe CoeFfiCientS Of a pOwer seRies from InforMAtion about its bEHavior on the ciRClE $\gAmMA$ of cOnvErgence. Our hYpotHEses Are mOSt ConVEnienTly foRmULaTEd in terms of the functIoNs $$\labeL{eqn.h.Kappa}
h (s)=2[\Psi (1-s)- (1-s)] \qUad \text{and} \QUAD
\kappa (s) =\iNt_{0}^{s} h (S')\,Ds',$$ WHere $\Psi$ is the prObabiLity generaTIng functIon of The offspRing distrIBUtion. SinCe tHe pOweR seRIEs For a probabiliTY GeneRaTing funCtiOn has raDiuS of ConVerGeNce $1$, the funCtion $h$ exTeNdS tO aN anAlytiC Function $H (z)$ In tHe DisK of raDIus $1$ cenTered At $z=1$, aS dOeS Its IntegraL $\KaPPA$. If tHe OfFsprIng DiStribUtioN Has Finite sUpport theN thE FuncTiOnS $h$ and $\kaPpa$ are polynomIaLs | distribution, that is, in the large -$k$beh avi or oftheprobabilities$ {{\m athbb P}}^{y}\{N \geqk\}$.
\ [the o re m.kil led-num b er . w eak \] I f t he of fspri ngdistrib ution hasmea n$1$, positiv e v ariance $\ sig ma^2$, and f ini te thi rd mo m ent t hen $$\l abel{e q n.kill ed-number .w e ak}
\sum_{ k = 1} ^{m} k{{\mathbb P}}(N \ ge q k) \sim 2C_{7 }y\sqr t{ m }\ q uad
\text{ w he re }\ ;\; C_7 = \f r a c {\s i gma}{\sqrt{6\ pi}}.$$
Th e pr oof, w hi chw ill us e a f or m of Karamata’s Tau berian th eorem, will be given i n sect ion \[ ssec . ki ll ed- nu m ber . we ak\ ] .
If we kn ew t hat t he s e q u e nce$k{ {\ma thbbP}}^{y} (N\ge q k )$ w e remonot one t henwe coul d conc ludefr om that ${{\mat hbbP}}^{y}
( N\g eq k) \s im C_ { 7}y/k^ {3/ 2}$ . Howev er, its eem su n l ik ely that monotonic it y of the seq uence$ k{ {\ m athbb P} }^ {y} (N\ g e q k)$ hol d sin gener al. Th u s, t o obtai nsharpas ymp tot ic re s ults about the ind ividu a l probabilitie s ${\mathbb{P} } ^{ y } ( N =k)$ , w e will impo se m o re r estr i ct ive hypot heses o n t h e offspring distrib ut ion th at wi ll allow us t o avoid th e u se of Ka rama t a’ s theorem. In i ts pl ace, we wi l l use aresul t of Fla jolet and O dlyzko [ @FO 90] th ata l lo ws one to ext r a ct i nf ormatio n a bout th e a sym pto tic b ehavior o f the co ef fi ci en tsof ap ower ser ie s f ro m i nform a tion a boutitsbe ha v ior on the ci r c le $ \G am ma$ofco nverg ence . Ou r hypot heses are mo s t co nv en ientlyformulated in t erms of th efun ctions $ $\label{ eqn.h.kappa}
h (s)= 2 [\Psi ( 1-s )- (1 -s)] \quad \t ext {and}\qu a d
\kapp a (s) = \in t _ {0}^{ s } h (s ') \,ds',$$ w h e re$\Psi $is t he prob ability generating fun ction of theoff spri n g d ist r ib u tio n. Sin c e the power seri es for a p ro b ab ility gene r ati ng functi on hasradiu s of con vergence$1$, thefu ncti o n $h $ extendsto an an alytic fu n ction $h (z)$ in the d is k o f rad ius $1 $ ce ntere d at $ z= 1$, as does i ts integ ral $\kappa$. If the of fsprin g dis tri bution ha s f i nit e support the n the func tio ns$h$ a nd$ \kapp a$ a r epol y nomia ls | distribution,_that is,_in the large-$k$ behavior_of the_probabilities_${{\mathbb P}}^{y}\{N_\geq
k\}$.
\[theorem.killed-number.weak\]_If the offspring_distribution has mean_$1$, positive variance $\sigma^2$,_and finite third_moment_then $$\label{eqn.killed-number.weak}
\sum_{k=1}^{m} k{{\mathbb P}}(N\geq k) \sim 2C_{7}y\sqrt{m} \quad
__\text{ where_}_\;\;_C_7=\frac{\sigma}{\sqrt{6\pi}}.$$
The proof, which will use_a form of Karamata’s Tauberian_theorem, will_be given in section \[ssec.killed-number.weak\].
If we knew that the_sequence_$k{{\mathbb P}}^{y} (N\geq_k)$ were monotone then we could conclude from that_${{\mathbb P}}^{y}
(N\geq k)\sim C_{7}y/k^{3/2}$. However, it_seems unlikely that_monotonicity_of_the sequence $k{{\mathbb P}}^{y}_(N\geq k)$ holds in general. Thus,_to obtain sharp asymptotic results about_the individual probabilities ${\mathbb{P}}^{y} (N=k)$, we will_impose more restrictive hypotheses on the_offspring distribution that will allow_us to_avoid the use of Karamata’s_theorem. In its_place, we_will use a_result of Flajolet and Odlyzko [@FO90]_that allows one_to extract information about the asymptotic_behavior_of the coefficients_of_a_power series_from information about_its_behavior on_the_circle $\Gamma$ of convergence. Our hypotheses_are_most conveniently formulated in terms of the_functions $$\label{eqn.h.kappa}
__h (s)=2[\Psi (1-s)- (1-s)]_\quad \text{and} \quad
_ \kappa (s) =\int_{0}^{s} h_(s')\,ds',$$ where_$\Psi$ is_the probability generating function of the offspring distribution. Since the power_series for a probability generating function_has radius of convergence_$1$, the_function_$h$ extends to_an_analytic function_$h (z)$ in the disk of radius_$1$ centered_at $z=1$, as does its integral_$\kappa$. If the offspring_distribution_has finite support then the functions_$h$ and $\kappa$ are polynomials |
small $c>0$ such that the following is satisfied for all $n\geq 1$:
\(i) $v^2>c$ and $$\begin{aligned}
\label{Var}
\sigma_S^2 \equiv \text{Var}_\mathscr{F}\left( \frac{1}{\sqrt{n}}\sum_{i \in N} q_{i,S}\right) > c.
\end{aligned}$$
\(ii) $\frac{1}{n}\sum_{i \in N}|N(i)| > c$ and $\frac{1}{n}\sum_{j \in N}|N_O(j)| > c$, where $N_O(j)$ denotes the out-neighborhood of $j$, i.e., $N_O(j) = \{i \in N: ij \in E\}$.
\(iii) There exists $C>0$ such that for all $i \in N$, $|N(i)| + |N_O(i)| \le C$, for all $n \ge 1$.
Assumptions \[assump: non deg\](i)-(ii) are conditions that ensure nondegeneracy of the distribution of the test statistics and moment conditions. Assumption \[assump: non deg\](iii) is a bounded degree assumption. This assumption ensures that the graph $G$ is sufficiently sparse. We can relax this assumption with a more complex form of conditions and proofs, but this relaxation provides little additional insights.
\[assump: consistency\] For some sequence $\omega_n \rightarrow 0$ such that $\sqrt{n}\omega_n^2 \rightarrow 0$, $$\begin{aligned}
\label{consistency}
\max_{j \in N} |\hat \mu_{j,0,S} - \mu_{j,0,S}| &=& O_P(\omega_n), \text{ for all } S \subset \mathbb{S}, \text{ and }\\ \notag
\max_{i \in N} |\hat \mu_{i,1} - \mu_{i,1}| &=& O_P(\omega_n).
\end{aligned}$$
Assumption \[assump: consistency\] requires a certain rate of convergence for the estimators. Since our covariates are all discrete, the rate of convergence is typically | small $ c>0 $ such that the following is satisfied for all $ n\geq 1 $:
\(i) $ v^2 > c$ and $ $ \begin{aligned }
\label{Var }
\sigma_S^2 \equiv \text{Var}_\mathscr{F}\left (\frac{1}{\sqrt{n}}\sum_{i \in N } q_{i, S}\right) > c.
\end{aligned}$$
\(ii) $ \frac{1}{n}\sum_{i \in N}|N(i)| > c$ and $ \frac{1}{n}\sum_{j \in N}|N_O(j)| > c$, where $ N_O(j)$ denotes the out - vicinity of $ j$, i.e., $ N_O(j) = \{i \in N: ij \in E\}$.
\(iii) There exist $ C>0 $ such that for all $ i \in N$, $ |N(i)| + |N_O(i)| \le C$, for all $ n \ge 1$.
Assumptions \[assump: non deg\](i)-(ii) are conditions that ensure nondegeneracy of the distribution of the test statistic and moment conditions. Assumption \[assump: non deg\](iii) is a jump degree assumption. This premise ensures that the graph $ G$ is sufficiently sparse. We can relax this presumption with a more complex form of conditions and validation, but this liberalization provides little additional insights.
\[assump: consistency\ ] For some sequence $ \omega_n \rightarrow 0 $ such that $ \sqrt{n}\omega_n^2 \rightarrow 0 $, $ $ \begin{aligned }
\label{consistency }
\max_{j \in N } |\hat \mu_{j,0,S } - \mu_{j,0,S}| & = & O_P(\omega_n), \text { for all } S \subset \mathbb{S }, \text { and } \\ \notag
\max_{i \in N } |\hat \mu_{i,1 } - \mu_{i,1}| & = & O_P(\omega_n).
\end{aligned}$$
Assumption \[assump: consistency\ ] necessitate a certain rate of convergence for the calculator. Since our covariates are all discrete, the rate of overlap is typically | smwll $c>0$ such that the folluwing is satisfnwd for all $n\feq 1$:
\(i) $v^2>c$ and $$\begin{aligned}
\label{Vac}
\sigmq_S^2 \equiv \text{Var}_\mathscf{F}\left( \frwc{1}{\sqrt{n}}\sym_{i \mn N} q_{i,S}\right) > c.
\end{aligned}$$
\(jl) $\frae{1}{n}\wum_{i \in N}|N(i)| > c$ and $\frac{1}{n}\vum_{j \in N}|N_O(j)| > w$, dhzre $N_O(j)$ denotes the out-neighborhood jf $j$, i.e., $N_L(j) = \{i \in N: ij \yn E\}$.
\(piy) Thsge exists $C>0$ such that for all $i \in N$, $|N(p)| + |N_O(i)| \le C$, for akl $n \ge 1$.
Assumptions \[assump: non deg\](i)-(ii) are conditlons that ebsurq nondegenerazy of the distribution of the test statistics and momdnt cpnditions. Qswumojion \[assump: ion ded\](iii) is a bommded dagree axsumption. This asvumption ensures that thx graph $G$ is sufficigntly sparve. We can relax thiw qssumktion witf a oort cpmllex flrm of conditjons and priofs, but this relaxstyin provides ljttle wdqitional insights.
\[assump: consistency\] For soje sequence $\omega_n \righrarrow 0$ such that $\sqrj{n}\omega_n^2 \ryghtarrow 0$, $$\begin{aligned}
\label{consistency}
\maf_{j \in V} |\hcb \oy_{j,0,D} - \mu_{j,0,S}| &=& O_P(\omega_n), \text{ for all } S \subset \mathfg{S}, \tvxt{ and }\\ \notag
\max_{i \in N} |\nah \ko_{i,1} - \mu_{i,1}| &=& O_P(\omeea_n).
\ens{aligned}$$
Assumption \[assump: consustency\] rtquirrs a certain rate of convertence for thv esrimators. Since our covariates are all ciscrete, the rate of couvergehce is typifally | small $c>0$ such that the following is all 1$: \(i) and $$\begin{aligned} \label{Var} N} > c. \end{aligned}$$ $\frac{1}{n}\sum_{i \in N}|N(i)| c$ and $\frac{1}{n}\sum_{j \in N}|N_O(j)| > where $N_O(j)$ denotes the out-neighborhood of $j$, i.e., $N_O(j) = \{i \in N: \in E\}$. \(iii) There exists $C>0$ such that for all $i \in N$, + \le for $n \ge 1$. Assumptions \[assump: non deg\](i)-(ii) are conditions that ensure nondegeneracy of the distribution of test statistics and moment conditions. Assumption \[assump: non is a bounded degree This assumption ensures that the $G$ sufficiently sparse. can this with a more form of conditions and proofs, but this relaxation provides little additional insights. \[assump: consistency\] For some sequence \rightarrow 0$ $\sqrt{n}\omega_n^2 \rightarrow $$\begin{aligned} \max_{j N} |\hat \mu_{j,0,S} &=& O_P(\omega_n), \text{ for all } \text{ and }\\ \notag \max_{i \in N} |\hat - \mu_{i,1}| O_P(\omega_n). \end{aligned}$$ Assumption \[assump: consistency\] requires certain rate of convergence for the estimators. Since covariates are all discrete, the rate of convergence is typically | small $c>0$ such that the followinG is satisfiEd for All $N\geQ 1$:
\(i) $V^2>c$ anD $$\begIn{aligned}
\label{vAr}
\siGma_S^2 \equiv \text{Var}_\mathscR{F}\lefT( \fRAc{1}{\sqRT{n}}\Sum_{i \iN N} q_{i,S}\riGHt) > C.
\ENd{aLiGnEd}$$
\(iI) $\fRAc{1}{N}\sum_{i \In N}|n(i)| > c$ and $\fRac{1}{n}\sum_{j \in n}|N_O(J)| > c$, Where $N_O(j)$ denoTEs The out-neigHboRhood of $j$, i.e., $N_O(J) = \{i \iN N: ij \in e\}$.
\(iIi) THEre exIstS $C>0$ sucH that fOR all $i \iN N$, $|N(i)| + |N_O(i)| \le c$, fOR all $n \gE 1$.
assumptIONs \[AssuMp: non deg\](i)-(ii) are conDItIOns that ensure nOndegeNeRAcY OF thE diStribution Of The teST statisTIcS AND moMEnt conditions. assumption \[aSSumP: non deG\](iIi) iS A boundEd degReE AssUmption. This AssuMption ensUres thAT the graPH $G$ is sufFicienTly SpaRse. WE CaN rElaX tHIs aSSuMptIOn wIth a more CoMpLex foRm of CONDItioNs aNd prOofs, bUt this relaxatIon ProvIDes LittlE addiTionAl InsigHts.
\[assUmp: coNsIstency\] For some sEqueNce $\omega_n \RigHtArrOw 0$ Such tHAt $\sqrt{N}\omEga_N^2 \rightaRrow 0$, $$\begIN{alIgNED}
\LaBel{consistency}
\max_{j \In n} |\HAt \Mu_{j,0,S} - \mu_{j,0,S}| &=& o_P(\omegA_N), \tExT{ For all } S \sUbSet \MathBB{s}, \text{ And }\\ \nOTaG
\max_{i \in N} |\Hat \mu_{i,1} - \MU_{i,1}| &=& o_P(\Omega_n).
\eNd{AligneD}$$
ASsuMptIon \[asSUmp: cOnsistEncy\] requIres a CErtain rate of coNVergence for thE EsTIMaTOrs. SIncE our covariaTes aRE all DiscREtE, thE Rate oF convErGEnCE is typically | small $c>0$ such that the following is s ati sfi ed for all $n\geq 1$:
\ ( i) $ v^2>c$ and $$\begin{al igned } \ l ab el{Va r}
\ si g m a_S ^2 \ equ iv \t ext{V ar} _\maths cr{F}\left ( \ fr ac{1}{\sqrt{ n }} \sum_{i \i n N } q_{i,S}\ri ght ) > c.
\ end{a lig ned}$ $
\(i i ) $\fr ac{1}{n}\ su m _{i \i n N}|N(i ) | > c$and $\frac{1}{n}\ s um _ {j \in N}|N_O( j)| >c$ , w h e re$N_ O(j)$ deno te s the out-nei g hb o r h ood of $j$, i.e., $N_O(j) =\ {i\in N: i j \ i n E\}$ .
\( ii i ) T here exists $C> 0$ such t hat fo r all $i \in N$, $|N(i )|+ | N_O( i )| \ leC$ , fo r a ll$ n \ ge 1$.
As su mptio ns \ [ a s s ump: no n de g\](i )-(ii) are co ndi tion s th at en surenond eg enera cy ofthe d is tribution of th e te st statis tic sand m oment condit ion s.Assumpt ion \[a s sum p: n o ndeg\](iii) is a bo un d e ddegree a ssumpt i on .T his assu mp tio n en s u res t hatt he graph $ G$ iss uf fi ciently s parse. W e c anrelax this assum ption wi th am ore complex fo r m of conditio n sa n dp roof s,but this re laxa t ionprov i de s l i ttleaddit io n al insights.
\[assump :consis tency \] For some s equence $\ o m e ga_n \ri ghta r ro w 0$ such that$\sqr t{n}\omega _ n^2 \rig htarr ow 0$, $ $\begin{a l i gned}
\ lab el{ con s i st ency}
\ma x _ {j \ in N} |\h at\mu_{j, 0,S } - \m u_{ j, 0,S}| &=& O_P(\om eg a_ n) ,\te xt{ f o r all }S\su bs et\math b b{S},\text { an d}\ \ \n otag
\ m a x_{i \ in N}|\h at \mu_ {i,1 } -\mu_{i, 1}| &=& O _P( \ omeg a_ n) .
\ end{aligned}$ $
Assumptio n\[a ssump: c onsisten cy\] requires a certain rate of co nverg ence for theest imator s.S ince o ur cov ariat es ar e all d i s cr ete ,the rate o f con verge nc e is typica lly | small_$c>0$ such_that the following is_satisfied for_all_$n\geq 1$:
\(i)_$v^2>c$_and $$\begin{aligned}
_ \label{Var}
_ \sigma_S^2_\equiv \text{Var}_\mathscr{F}\left( \frac{1}{\sqrt{n}}\sum_{i_\in_N} q_{i,S}\right) > c.
\end{aligned}$$
\(ii) $\frac{1}{n}\sum_{i \in N}|N(i)| > c$ and_$\frac{1}{n}\sum_{j_\in N}|N_O(j)|_>_c$,_where $N_O(j)$ denotes the out-neighborhood_of $j$, i.e., $N_O(j) =_\{i \in_N: ij \in E\}$.
\(iii) There exists $C>0$ such_that_for all $i_\in N$, $|N(i)| + |N_O(i)| \le C$, for all_$n \ge 1$.
Assumptions \[assump: non deg\](i)-(ii)_are conditions that_ensure_nondegeneracy_of the distribution of_the test statistics and moment conditions._Assumption \[assump: non deg\](iii) is a_bounded degree assumption. This assumption ensures that_the graph $G$ is sufficiently sparse._We can relax this assumption_with a_more complex form of conditions_and proofs, but_this relaxation_provides little additional_insights.
\[assump: consistency\] For some sequence $\omega_n_\rightarrow 0$ such_that $\sqrt{n}\omega_n^2 \rightarrow 0$, $$\begin{aligned}
__ \label{consistency}
___\max_{j \in_N} |\hat \mu_{j,0,S}_-_\mu_{j,0,S}| &=&_O_P(\omega_n),_\text{ for all } S \subset_\mathbb{S},_\text{ and }\\ \notag
_\max_{i \in N} |\hat_\mu_{i,1}_- \mu_{i,1}| &=& O_P(\omega_n).
_ \end{aligned}$$
Assumption \[assump:_consistency\] requires a certain rate of_convergence for_the estimators._Since our covariates are all discrete, the rate of convergence is_typically |
before in Section 3 and in agreement with the root identity for $\zeta_{k}$ at $\mu=0$ and Result 2 for the value of $d_{\zeta_{k}}(s_{0},0)$.
#### Case 2 - $\mu=-1$:
In this case, since the roots $\rho_{i}=\beta_{i}+i\gamma_{i}$ of $\zeta_{k}$ are mirror-symmetric in the critical line, we have, on invoking the result just proved for $r_{\zeta_{k}}(s_{0},0)$, that
$$\begin{aligned}
r_{\zeta_{k}}(s_{0},-1) & = & -\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}(s_{0}-\rho_{i})^{1}\\
\\ & = & -(s_{0}-\frac{1}{2})r_{\zeta_{k}}(s_{0},0)+i\cdot\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}\gamma_{i}\\
\\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \int_{0}^{T}t\,\textrm{d}N(t)-\int_{0}^{\tilde{T}}\tilde{t}\,\textrm{d}N(\tilde{t})\right\} \end{aligned}$$
Using the notation from [\[]{}1[\]]{} under which $N_{i}(T)$ represents the i’th integral of $N(T)$ and similarly for $\check{N}(T)$ and $S(T)$, this becomes
$$\begin{aligned}
& & r_{\zeta_{k}}(s_{0},-1)\nonumber \\
\nonumber \\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{ TN(T)-N_{1}(T)\right\} -\left\{ \tilde{T}N(\tilde{T})-N_{1}(\tilde{T})\right\} \right\} \nonumber \\
\nonumber \\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{ \begin{array}{cc}
\left\{ T\check{N}(T)-\check{N}_{1}(T)\ | before in Section 3 and in agreement with the root identity for $ \zeta_{k}$ at $ \mu=0 $ and Result 2 for the value of $ d_{\zeta_{k}}(s_{0},0)$.
# # # # subject 2 - $ \mu=-1 $:
In this subject, since the roots $ \rho_{i}=\beta_{i}+i\gamma_{i}$ of $ \zeta_{k}$ are mirror - symmetric in the critical line, we have, on invoking the solution just proved for $ r_{\zeta_{k}}(s_{0},0)$, that
$ $ \begin{aligned }
r_{\zeta_{k}}(s_{0},-1) & = & -\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}(s_{0}-\rho_{i})^{1}\\
\\ & = & -(s_{0}-\frac{1}{2})r_{\zeta_{k}}(s_{0},0)+i\cdot\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}\gamma_{i}\\
\\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\ { \int_{0}^{T}t\,\textrm{d}N(t)-\int_{0}^{\tilde{T}}\tilde{t}\,\textrm{d}N(\tilde{t})\right\ } \end{aligned}$$
use the note from [ \[]{}1[\ ] ] { } under which $ N_{i}(T)$ represents the i’th integral of $ N(T)$ and similarly for $ \check{N}(T)$ and $ S(T)$, this becomes
$ $ \begin{aligned }
& & r_{\zeta_{k}}(s_{0},-1)\nonumber \\
\nonumber \\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\ { \left\ { TN(T)-N_{1}(T)\right\ } -\left\ { \tilde{T}N(\tilde{T})-N_{1}(\tilde{T})\right\ } \right\ } \nonumber \\
\nonumber \\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\ { \left\ { \begin{array}{cc }
\left\ { T\check{N}(T)-\check{N}_{1}(T)\ | bevore in Section 3 and in xgreement with jhw root identjty for $\xeta_{k}$ at $\mu=0$ and Result 2 for vhe calue of $d_{\zeta_{k}}(s_{0},0)$.
#### Case 2 - $\mu=-1$:
Iv this cade, since the eoots $\rho_{i}=\usta_{i}+i\gamma_{i}$ or $\zetc_{k}$ are mirror-symketric in dhe critical lhnd, ce have, on invoking the result just [roved gog $r_{\zeta_{k}}(s_{0},0)$, that
$$\fegim{wlighvd}
v_{\zeta_{k}}(s_{0},-1) & = & -\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}(a_{0}-\rho_{i})^{1}\\
\\ & = & -(s_{0}-\frac{1}{2})r_{\zeta_{k}}(x_{0},0)+i\cdot\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{n}\}}\gamla_{i}\\
\\ & = & i\cdot\undersft{z,\tilde{z}\rithtaweow\infty}{Clim}\,\ueft\{ \int_{0}^{T}t\,\uertrm{d}N(t)-\int_{0}^{\tjlde{T}}\tilde{t}\,\textrm{d}N(\tilde{t})\right\} \dnd{alngned}$$
Using jkw nljation from [\[]{}1[\]]{} undeg which $N_{i}(T)$ rciresentv the i’yh integral of $N(T)$ ane similarly for $\check{I}(T)$ and $S(T)$, this becomgs
$$\begin{alicnzd}
& & r_{\zeta_{k}}(s_{0},-1)\nonumber \\
\ninumbgr \\ & = & i\cait\uvdedsxt{z,\filde{z}\gigitarrow\inftg}{Clim}\,\left\{ \lwft\{ TN(T)-N_{1}(T)\right\} -\left\{ \tyode{T}N(\tilde{T})-N_{1}(\tjlde{T})\rygrt\} \right\} \nonumber \\
\nonumber \\ & = & i\cdot\undtrset{a,\tilde{z}\rightarrow\infty}{Coim}\,\left\{ \left\{ \begin{arrwy}{cc}
\left\{ E\check{N}(T)-\check{N}_{1}(T)\ | before in Section 3 and in agreement root for $\zeta_{k}$ $\mu=0$ and Result $d_{\zeta_{k}}(s_{0},0)$. Case 2 - In this case, the roots $\rho_{i}=\beta_{i}+i\gamma_{i}$ of $\zeta_{k}$ are in the critical line, we have, on invoking the result just proved for that $$\begin{aligned} r_{\zeta_{k}}(s_{0},-1) & = & -\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}(s_{0}-\rho_{i})^{1}\\ \\ & = & -(s_{0}-\frac{1}{2})r_{\zeta_{k}}(s_{0},0)+i\cdot\sum_{\{s_{0}-roots\:\rho_{i}\: \\ = i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \end{aligned}$$ Using the notation from [\[]{}1[\]]{} under which $N_{i}(T)$ represents the i’th integral of $N(T)$ and for $\check{N}(T)$ and $S(T)$, this becomes $$\begin{aligned} & r_{\zeta_{k}}(s_{0},-1)\nonumber \\ \nonumber \\ = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{ TN(T)-N_{1}(T)\right\} \tilde{T}N(\tilde{T})-N_{1}(\tilde{T})\right\} \nonumber \\ \\ = i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{ \begin{array}{cc} T\check{N}(T)-\check{N}_{1}(T)\ | before in Section 3 and in agreeMent with thE root IdeNtiTy For $\zEta_{k}$ At $\mu=0$ and Result 2 fOR the Value of $d_{\zeta_{k}}(s_{0},0)$.
#### Case 2 - $\mu=-1$:
In This cAsE, SincE ThE rootS $\rho_{i}=\beTA_{i}+I\GAmmA_{i}$ Of $\ZetA_{k}$ ARe MirroR-syMmetric In the critiCal LiNe, we have, on inVOkIng the resuLt jUst proved for $R_{\zeTa_{k}}(s_{0},0)$, thAt
$$\BegIN{aligNed}
R_{\zeta_{K}}(s_{0},-1) & = & -\sum_{\{s_{0}-ROots\:\rhO_{i}\: of\:\zeta_{k}\}}(S_{0}-\rHO_{i})^{1}\\
\\ & = & -(s_{0}-\fraC{1}{2})R_{\zeta_{k}}(s_{0},0)+I\CDoT\sum_{\{S_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}\GAmMA_{i}\\
\\ & = & i\cdot\underseT{z,\tildE{z}\RIgHTArrOw\iNfty}{Clim}\,\leFt\{ \Int_{0}^{T}t\,\TExtrm{d}N(T)-\InT_{0}^{\TILde{t}}\Tilde{t}\,\textrm{d}n(\tilde{t})\righT\} \End{AligneD}$$
USinG The notAtion FrOM [\[]{}1[\]]{} unDer which $N_{i}(T)$ ReprEsents the I’th intEGral of $N(t)$ And simiLarly fOr $\cHecK{N}(T)$ aND $S(t)$, tHis BeCOmeS
$$\BeGin{ALigNed}
& & r_{\zeta_{K}}(s_{0},-1)\NoNumbeR \\
\nonUMBER \\ & = & i\cdOt\uNderSet{z,\tIlde{z}\rightarrOw\iNfty}{cLim}\,\Left\{ \lEft\{ TN(t)-N_{1}(T)\rIgHt\} -\lefT\{ \tilde{t}N(\tilDe{t})-N_{1}(\tilde{T})\right\} \riGht\} \nOnumber \\
\noNumBeR \\ & = & i\cDoT\undeRSet{z,\tiLde{Z}\riGhtarroW\infty}{CLIm}\,\lEfT\{ \LEFt\{ \Begin{array}{cc}
\left\{ T\cHeCK{n}(T)-\Check{N}_{1}(T)\ | before in Section 3 and i n agreemen t wit h t hero ot i dent ity for $\zeta _ {k}$ at $\mu=0$ and Result 2 fo rt he v a lu e of$d_{\ze t a_ { k }}( s_ {0 },0 )$ .
####Cas e 2 - $ \mu=-1$:
Inth is case, sin c ethe roots$\r ho_{i}=\beta _{i }+i\ga mm a_{ i }$ of $\ zeta_ {k}$ a r e mirr or-symmet ri c in th e critic a l l ine, we have, on invo k in g the result ju st pro ve d f o r $r _{\ zeta_{k}}( s_ {0},0 ) $, that
$ $ \ b egi n {aligned}
r_{ \zeta_{k}}( s _{0 },-1)&= & -\sum_ {\{s_ {0 } -ro ots\:\rho_{ i}\: of\:\zet a_{k}\ } }(s_{0} - \rho_{i })^{1} \\\\& =& - (s _{0 }- \ fra c {1 }{2 } )r_ {\zeta_{ k} }( s_{0} ,0)+ i \ c d ot\s um_ {\{s _{0}- roots\:\rho_{ i}\ : of \ :\z eta_{ k}\}} \gam ma _{i}\ \
\\ & = &i\ cdot\underset{z ,\ti lde{z}\ri ght ar row \i nfty} { Clim}\ ,\l eft \{ \int _{0}^{T } t\, \t e x t rm {d}N(t)-\int_{0}^{ \t i l de {T}}\til de{t}\ , \t ex t rm{d}N(\ ti lde {t}) \ r ight\ } \e n d{ aligned} $$
Us i ng t he nota ti on fro m[\[ ]{} 1[\]] { } un der wh ich $N_{ i}(T) $ represents th e i’th integra l o f $N ( T)$and similarlyfor$ \che ck{N } (T )$a nd $S (T)$, t h is becomes
$$\begin{a li gned} & & r_{\zeta_{k} }(s_{0},-1 ) \ n onumber\\
\ n on u mber \\ & = &i\cdo t\underset { z,\tilde {z}\r ightarro w\infty}{ C l im}\,\le ft\ { \ lef t\{ T N( T)-N_{1}(T)\r i g ht\} - \left\{ \t ilde{T} N(\ til de{ T}) -N _{1}(\til de{T})\r ig ht \} \ rig ht\}\ nonumber \ \
\ no num ber \ \ & = & i\cd ot\u nd er s et{ z,\tild e {z } \ righ ta rr ow\i nft y} {Clim }\,\ l eft \{ \lef t\{ \begi n{a r ray} {c c}
\left\ { T\check{N}( T) -\check{N} _{ 1}( T)\ | before_in Section_3 and in agreement_with the_root_identity for_$\zeta_{k}$_at $\mu=0$ and_Result 2 for_the value of $d_{\zeta_{k}}(s_{0},0)$.
####_Case 2 -_$\mu=-1$:
In_this case, since the roots $\rho_{i}=\beta_{i}+i\gamma_{i}$ of $\zeta_{k}$ are mirror-symmetric in the critical line,_we_have, on_invoking_the_result just proved for $r_{\zeta_{k}}(s_{0},0)$,_that
$$\begin{aligned}
r_{\zeta_{k}}(s_{0},-1) & = & -\sum_{\{s_{0}-roots\:\rho_{i}\:_of\:\zeta_{k}\}}(s_{0}-\rho_{i})^{1}\\
\\ &_= & -(s_{0}-\frac{1}{2})r_{\zeta_{k}}(s_{0},0)+i\cdot\sum_{\{s_{0}-roots\:\rho_{i}\: of\:\zeta_{k}\}}\gamma_{i}\\
\\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{_\int_{0}^{T}t\,\textrm{d}N(t)-\int_{0}^{\tilde{T}}\tilde{t}\,\textrm{d}N(\tilde{t})\right\}_\end{aligned}$$
Using the notation_from [\[]{}1[\]]{} under which $N_{i}(T)$ represents the i’th integral_of $N(T)$ and similarly for $\check{N}(T)$_and $S(T)$, this_becomes
$$\begin{aligned}
_&_ & r_{\zeta_{k}}(s_{0},-1)\nonumber \\
\nonumber_\\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{_TN(T)-N_{1}(T)\right\} -\left\{ \tilde{T}N(\tilde{T})-N_{1}(\tilde{T})\right\} \right\} \nonumber \\
\nonumber_\\ & = & i\cdot\underset{z,\tilde{z}\rightarrow\infty}{Clim}\,\left\{ \left\{ \begin{array}{cc}
\left\{_T\check{N}(T)-\check{N}_{1}(T)\ |
g}})-\pi_{\mathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\mathfrak{g}}]),\ \ U_{k}:=Z_{k}-[\pi_{\mathfrak{g}},X_k].$$ Using Lemma \[Lemma\_domain\_of\_flowfixed\] and inequality (\[EQ\_X\_n\_nfixed\]), these terms can be bounded by $$\begin{aligned}
\label{EQ_Vfixed}
\|V_k\|_{s,r_{k+1}}&\leq C\|\pi_{\mathfrak{g}}\|_{s+2,r_k}\|X_k\|_{0,r_k}\|X_k\|_{s+2,r_k}\leq Ct_k^{s+2-2\alpha},\\
\label{EQ_Ufixed}\|\varphi_{X_k}^{*}(U_{k})\|_{s,r_{k+1}}&\leq C(\|U_k\|_{s,r_k}+\|U_k\|_{0,r_k}\|X_k\|_{s+1,r_{k}})\leq\\
\nonumber&\leq C(\|U_k\|_{s,r_k}+t_k^{s+1-\alpha}\|U_k\|_{0,r_k}).\end{aligned}$$ To compute the $C^s$-norm for $U_k$, we rewrite it as $$\begin{aligned}
U_k&=Z_k-[\pi_{\mathfrak{g}},X_k]=[\pi_{\mathfrak{g}},h_1^{r_k}(Z_k)]+h_2^{r_k}([\pi_{\mathfrak{g}},Z_k])-[\pi_{\mathfrak{g}},X_k]=\\
&=[\pi_{\mathfrak{g}},(I-S_k)h_1^{r_k}(Z_k)]-\frac{1}{2}h_2^{r_k}([Z_k,Z_k]).\end{aligned}$$ By tameness of the Lie bracket, the first term can be bounded by $$\begin{aligned}
\|[\pi_{\mathfrak{g}},(I-S_k)& h_1^{r_k}(Z_k)]\|_{s,r_k}\leq C \|(I-S_k)h_1^{r_k}(Z_k)\|_{s+1,r | g}})-\pi_{\mathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\mathfrak{g}}]),\ \ U_{k}:=Z_{k}-[\pi_{\mathfrak{g}},X_k].$$ Using Lemma \[Lemma\_domain\_of\_flowfixed\ ] and inequality (\[EQ\_X\_n\_nfixed\ ]), these terms can be bounded by $ $ \begin{aligned }
\label{EQ_Vfixed }
\|V_k\|_{s, r_{k+1}}&\leq C\|\pi_{\mathfrak{g}}\|_{s+2,r_k}\|X_k\|_{0,r_k}\|X_k\|_{s+2,r_k}\leq Ct_k^{s+2 - 2\alpha},\\
\label{EQ_Ufixed}\|\varphi_{X_k}^{*}(U_{k})\|_{s, r_{k+1}}&\leq C(\|U_k\|_{s, r_k}+\|U_k\|_{0,r_k}\|X_k\|_{s+1,r_{k}})\leq\\
\nonumber&\leq C(\|U_k\|_{s, r_k}+t_k^{s+1-\alpha}\|U_k\|_{0,r_k}).\end{aligned}$$ To compute the $ C^s$-norm for $ U_k$, we rewrite it as $ $ \begin{aligned }
U_k&=Z_k-[\pi_{\mathfrak{g}},X_k]=[\pi_{\mathfrak{g}},h_1^{r_k}(Z_k)]+h_2^{r_k}([\pi_{\mathfrak{g}},Z_k])-[\pi_{\mathfrak{g}},X_k]=\\
& =[ \pi_{\mathfrak{g}},(I - S_k)h_1^{r_k}(Z_k)]-\frac{1}{2}h_2^{r_k}([Z_k, Z_k]).\end{aligned}$$ By jejunity of the Lie bracket, the inaugural term can be bounded by $ $ \begin{aligned }
\|[\pi_{\mathfrak{g}},(I - S_k) & h_1^{r_k}(Z_k)]\|_{s, r_k}\leq C \|(I - S_k)h_1^{r_k}(Z_k)\|_{s+1,r | g}})-\pi_{\lathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\matmfrak{g}}]),\ \ U_{k}:=Z_{k}-[\pi_{\majhdrak{g}},X_n].$$ Usinf Lemma \[Uemma\_domain\_of\_flowfixed\] and iiequqlity (\[EQ\_X\_n\_nfixed\]), these termr can be hounded vy $$\btgin{aligned}
\label{EQ_Vfixed}
\|V_k\|_{s,r_{k+1}}&\les C\|\pi_{\kethfrak{g}}\|_{s+2,r_k}\|X_k\|_{0,r_k}\|W_k\|_{s+2,r_k}\leq Cd_k^{s+2-2\alpha},\\
\label{EX_Uwired}\|\varphi_{X_k}^{*}(U_{k})\|_{s,r_{k+1}}&\leq C(\|U_k\|_{s,r_k}+\|U_k\|_{0,r_k}\|X_k\|_{s+1,r_{k}})\lqq\\
\nonumneg&\leq C(\|U_k\|_{s,r_k}+t_k^{s+1-\wlphs}\|T_k\|_{0,r_k}).\snd{aligned}$$ To compute the $C^s$-norm fkr $U_k$, wt rewrite it as $$\bebin{aligned}
U_k&=Z_k-[\pi_{\mathfrak{g}},X_n]=[\pi_{\mwthfrak{g}},h_1^{r_k}(Z_k)]+h_2^{r_k}([\pi_{\lathfrak{g}},Z_k])-[\kj_{\maeyfrak{g}},X_k]=\\
&=[\pi_{\maghfrak{g}},(I-S_k)h_1^{r_k}(Z_k)]-\frac{1}{2}h_2^{r_k}([E_k,Z_k]).\end{aligned}$$ By tameness of thd Lie bracket, tye figvt term can be bjunded by $$\benpn{alignad}
\|[\pi_{\matnfrak{g}},(I-S_k)& h_1^{r_k}(Z_l)]\|_{s,r_n}\lew C \|(I-S_k)h_1^{r_k}(Z_k)\|_{s+1,r | g}})-\pi_{\mathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\mathfrak{g}}]),\ \ U_{k}:=Z_{k}-[\pi_{\mathfrak{g}},X_k].$$ Using Lemma \[Lemma\_domain\_of\_flowfixed\] and these can be by $$\begin{aligned} \label{EQ_Vfixed} \nonumber&\leq To compute the for $U_k$, we it as $$\begin{aligned} U_k&=Z_k-[\pi_{\mathfrak{g}},X_k]=[\pi_{\mathfrak{g}},h_1^{r_k}(Z_k)]+h_2^{r_k}([\pi_{\mathfrak{g}},Z_k])-[\pi_{\mathfrak{g}},X_k]=\\ &=[\pi_{\mathfrak{g}},(I-S_k)h_1^{r_k}(Z_k)]-\frac{1}{2}h_2^{r_k}([Z_k,Z_k]).\end{aligned}$$ By of the Lie bracket, the first term can be bounded by $$\begin{aligned} \|[\pi_{\mathfrak{g}},(I-S_k)& C \|(I-S_k)h_1^{r_k}(Z_k)\|_{s+1,r | g}})-\pi_{\mathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\mAthfrak{g}}]),\ \ U_{k}:=z_{k}-[\pi_{\mAthFraK{g}},x_k].$$ UsIng LEmma \[Lemma\_domaiN\_Of\_flOwfixed\] and inequality (\[EQ\_x\_n\_nfiXeD\]), ThesE TeRms caN be bounDEd BY $$\BegIn{AlIgnEd}
\LAbEl{EQ_VFixEd}
\|V_k\|_{s,r_{k+1}}&\Leq C\|\pi_{\mathFraK{g}}\|_{S+2,r_k}\|X_k\|_{0,r_k}\|X_k\|_{s+2,r_k}\LEq ct_k^{s+2-2\alpha},\\
\lAbeL{EQ_Ufixed}\|\varPhi_{x_k}^{*}(U_{k})\|_{s,r_{K+1}}&\lEq C(\|u_K\|_{s,r_k}+\|U_K\|_{0,r_k}\|x_k\|_{s+1,r_{k}})\Leq\\
\nonUMber&\leQ C(\|U_k\|_{s,r_k}+t_k^{S+1-\aLPha}\|U_k\|_{0,r_K}).\End{aligNED}$$ TO comPute the $C^s$-norm for $U_K$, We REwrite it as $$\begiN{alignEd}
u_K&=Z_K-[\PI_{\maThfRak{g}},X_k]=[\pi_{\maThFrak{g}},H_1^{R_k}(Z_k)]+h_2^{r_k}([\PI_{\mATHFraK{G}},Z_k])-[\pi_{\mathfrak{G}},X_k]=\\
&=[\pi_{\mathfrAK{g}},(I-s_k)h_1^{r_k}(Z_K)]-\fRac{1}{2}H_2^{R_k}([Z_k,Z_k]).\End{alIgNEd}$$ BY tameness of The LIe bracket, The firST term caN Be boundEd by $$\beGin{AliGned}
\|[\PI_{\mAtHfrAk{G}},(i-S_k)& H_1^{R_k}(z_k)]\|_{s,R_K}\leQ C \|(I-S_k)h_1^{r_k}(z_k)\|_{S+1,r | g}})-\pi_{\mathfrak{g}}-\v arphi_{X_k }^{*} ([X _k, \p i_{\ math frak{g}}]),\ \ U_{k }:=Z_{k}-[\pi_{\mathfr ak{g} }, X _k]. $ $Using Lemma\ [L e m ma\ _d om ain \_ o f\ _flow fix ed\] an d inequali ty(\ [EQ\_X\_n\_n f ix ed\]), the seterms can be bo undedby $$ \ begin {al igned }
\lab e l{EQ_V fixed}
\| V_ k \|_{s, r _{k+1}} & \ le q C\ |\pi_{\mathfrak{g } }\ | _{s+2,r_k}\|X_ k\|_{0 ,r _ k} \ | X_k \|_ {s+2,r_k}\ le q Ct _ k^{s+2- 2 \a l p h a}, \ \
\label{EQ_U fixed}\|\va r phi _{X_k} ^{ *}( U _{k})\ |_{s, r_ { k+1 }}&\leq C(\ |U_k \|_{s,r_k }+\|U_ k \|_{0,r _ k}\|X_k \|_{s+ 1,r _{k }})\ l eq \\
\n on u mbe r &\ leq C(\ |U_k\|_{ s, r_ k}+t_ k^{s + 1 - \ alph a}\ |U_k \|_{0 ,r_k}).\end{a lig ned} $ $ T o com putethe$C ^s$-n orm fo r $U_ k$ , we rewrite it as$$\begin{ ali gn ed}
U _k&=Z _ k-[\pi _{\ mat hfrak{g }},X_k] = [\p i_ { \ m at hfrak{g}},h_1^{r_k }( Z _ k) ]+h_2^{r _k}([\ p i_ {\ m athfrak{ g} },Z _k]) - [ \pi_{ \mat h fr ak{g}},X _k]=\\ &= [\ pi_{\ma th frak{g }} ,(I -S_ k)h_1 ^ {r_k }(Z_k) ]-\frac{ 1}{2} h _2^{r_k}([Z_k, Z _k]).\end{ali g ne d } $$ By t ame ness of the Lie brac ket, th e f i rst t erm c an be bounded by $$\begin {a ligned }
\|[ \pi_{\mathfra k{g}},(I-S _ k ) & h_1^{r _k}( Z _k ) ]\|_{s,r_k}\le q C \ |(I-S_k)h_ 1 ^{r_k}(Z _k)\| _{s+1,r | g}})-\pi_{\mathfrak{g}}-\varphi_{X_k}^{*}([X_k,\pi_{\mathfrak{g}}]),\ \_U_{k}:=Z_{k}-[\pi_{\mathfrak{g}},X_k].$$ Using_Lemma \[Lemma\_domain\_of\_flowfixed\] and inequality_(\[EQ\_X\_n\_nfixed\]), these_terms_can be_bounded_by $$\begin{aligned}
\label{EQ_Vfixed}
\|V_k\|_{s,r_{k+1}}&\leq C\|\pi_{\mathfrak{g}}\|_{s+2,r_k}\|X_k\|_{0,r_k}\|X_k\|_{s+2,r_k}\leq_ Ct_k^{s+2-2\alpha},\\
\label{EQ_Ufixed}\|\varphi_{X_k}^{*}(U_{k})\|_{s,r_{k+1}}&\leq C(\|U_k\|_{s,r_k}+\|U_k\|_{0,r_k}\|X_k\|_{s+1,r_{k}})\leq\\
\nonumber&\leq_C(\|U_k\|_{s,r_k}+t_k^{s+1-\alpha}\|U_k\|_{0,r_k}).\end{aligned}$$ To compute the_$C^s$-norm for $U_k$,_we_rewrite it as $$\begin{aligned}
U_k&=Z_k-[\pi_{\mathfrak{g}},X_k]=[\pi_{\mathfrak{g}},h_1^{r_k}(Z_k)]+h_2^{r_k}([\pi_{\mathfrak{g}},Z_k])-[\pi_{\mathfrak{g}},X_k]=\\
&=[\pi_{\mathfrak{g}},(I-S_k)h_1^{r_k}(Z_k)]-\frac{1}{2}h_2^{r_k}([Z_k,Z_k]).\end{aligned}$$ By tameness of the Lie bracket, the first term can_be_bounded by_$$\begin{aligned}
\|[\pi_{\mathfrak{g}},(I-S_k)&_h_1^{r_k}(Z_k)]\|_{s,r_k}\leq_C \|(I-S_k)h_1^{r_k}(Z_k)\|_{s+1,r |
_u,
\label{small_K_kick}\end{aligned}$$ where $N_u$ is the number of undulator periods. Intuitively Eq. \[small\_K\_kick\] is just equal to the total energy loss as the electron travels through one undulator. When $K_u$ is increased (thereby resulting in an increased angular deflection) and the finite angular acceptance of the lens, $\theta_m$, taken into account, the on-axis electric field $E_x(x=y=0)$ in the KU is reduced by a factor $F_h(K_u,\gamma \theta_m)\leq1 $. The expression of $F_h(K_u,\gamma \theta_m)$ is derived in [@OSC_val] and its dependence on $K_u$ appears in Fig. \[fig1\] for three cases of $\gamma \theta_m$. There is an additional efficiency factor, $F_u(\kappa_u)=J_0(\kappa_u)-J_1(\kappa_u)$, which accounts the effect of the longitudinal oscillation \[given by $\frac{K_u^2}{8\gamma^2k_u}\sin(2k_u z)$\] of the particle in the KU where $\kappa_u\equiv K_u^2/4(1+K_u^2/2)$. The kick amplitude from Eq. \[small\_K\_kick\] is thus reduced by the factor of $F_h(K_u,\theta_m\gamma)\times F_u(\kappa_u)$.\
The simulation in SRW are performed in the frequency domain: the field frequency components within the PU-radiation bandwidth are propagated and the field amplitude in the time domain inside the KU is computed [@light_optics_ipac16]. This is first done for the case of a single focusing lens using $L_u$ and $\lambda_o$ from Tab. \[tab:TLT\], but varying $N_u$ and other parameters appropriately. For this benchmarking simulation, the distance between the PU and KU centers is taken to be $L_t=19.5$ m (i.e. $L_t\gg L_u$) in order to suppress the depth-of-field effect and the focal length of the lens is $f=L_t/2$. | _ u,
\label{small_K_kick}\end{aligned}$$ where $ N_u$ is the number of undulator periods. Intuitively Eq. \[small\_K\_kick\ ] is just equal to the entire department of energy loss as the electron travels through one undulator. When $ K_u$ is increase (thereby leave in an increased angular deflection) and the finite angular credence of the lens, $ \theta_m$, taken into account, the on - bloc electric sphere $ E_x(x = y=0)$ in the KU is reduced by a gene $ F_h(K_u,\gamma \theta_m)\leq1 $. The expression of $ F_h(K_u,\gamma \theta_m)$ is derived in [ @OSC_val ] and its addiction on $ K_u$ appears in Fig. \[fig1\ ] for three cases of $ \gamma \theta_m$. There is an additional efficiency factor, $ F_u(\kappa_u)=J_0(\kappa_u)-J_1(\kappa_u)$, which accounts the effect of the longitudinal cycle \[given by $ \frac{K_u^2}{8\gamma^2k_u}\sin(2k_u z)$\ ] of the particle in the KU where $ \kappa_u\equiv K_u^2/4(1+K_u^2/2)$. The kick amplitude from Eq. \[small\_K\_kick\ ] is thus reduce by the factor of $ F_h(K_u,\theta_m\gamma)\times F_u(\kappa_u)$.\
The simulation in SRW are perform in the frequency domain: the battlefield frequency components within the PU - radiation bandwidth are propagated and the field amplitude in the time knowledge domain inside the KU is computed [ @light_optics_ipac16 ]. This is first done for the case of a single focusing lens using $ L_u$ and $ \lambda_o$ from Tab. \[tab: TLT\ ], but varying $ N_u$ and other parameters appropriately. For this benchmarking simulation, the distance between the PU and KU centers is taken to be $ L_t=19.5 $ m (i.e. $ L_t\gg L_u$) in order to suppress the depth - of - field consequence and the focal length of the lens is $ f = L_t/2$. | _u,
\lwbel{small_K_kick}\end{aligned}$$ where $N_u$ is thg bumber of unsulator oeriods. Intuitively Eq. \[small\_N\_kuck\] iw just equal to the togal energj loss as the wlectron tczvels tmxough lne bnvulator. When $K_u$ is increaved (thereby revuutnng in an increased angular deflectijn) and yhf finite angulwr abcqptahbe of the lens, $\theta_m$, taken intk accouit, the on-axis ekectric field $E_x(x=y=0)$ in the NU id reduced by a fachor $F_h(K_u,\gamna \trwta_m)\leq1 $. The expression of $F_h(K_u,\gamja \theta_m)$ is derived in [@OSC_val] xnd ics dependenxe on $N_u$ appears mn Fig. \[sig1\] for threc cases of $\gamka \theta_m$. Therc is en aeditional efficiency hactor, $F_u(\kappa_u)=J_0(\kappw_u)-J_1(\kappa_u)$, wkich accounts the effwcr of jhe lmngigydival oxcjllatiln \[jiven by $\frzc{K_u^2}{8\gamma^2k_u}\win(2k_u z)$\] of the partocjv in the KU wgere $\kwp[a_u\equiv K_u^2/4(1+K_u^2/2)$. The kick amplitude from Ex. \[sjall\_K\_kick\] is thus reducwd by the factor of $F_j(K_u,\theta_m\damma)\times F_u(\kappa_u)$.\
The simulation in SRW are perfmrmed kn uhc weeeuency domain: the field frequency components rjtnik the PU-radiatiok bandwidth are prppwgsjed and the figld amplithde in the time dolain infide rhe KU is comluted [@light_optics_ipac16]. This us first donv foe the case of a siugle focusiny lens usinb $L_u$ and $\lambda_o$ from Tcb. \[tab:TMT\], but varylng $N_u$ ans other parameterr aipro[riately. For this benchmarhing simuoatipn, the aistsnce bqtween the PU akg KU centers is tanen tl te $L_t=19.5$ m (i.e. $P_t\gg L_u$) in order to suppress thx depth-of-fielc affvct and tke focsl length of ehe lens is $f=L_j/2$. | _u, \label{small_K_kick}\end{aligned}$$ where $N_u$ is the number periods. Eq. \[small\_K\_kick\] just equal to the travels through one When $K_u$ is (thereby resulting in an increased angular and the finite angular acceptance of the lens, $\theta_m$, taken into account, the electric field $E_x(x=y=0)$ in the KU is reduced by a factor $F_h(K_u,\gamma \theta_m)\leq1 The of \theta_m)$ derived in [@OSC_val] and its dependence on $K_u$ appears in Fig. \[fig1\] for three cases of \theta_m$. There is an additional efficiency factor, $F_u(\kappa_u)=J_0(\kappa_u)-J_1(\kappa_u)$, accounts the effect of longitudinal oscillation \[given by $\frac{K_u^2}{8\gamma^2k_u}\sin(2k_u of particle in KU $\kappa_u\equiv The kick amplitude Eq. \[small\_K\_kick\] is thus reduced by the factor of $F_h(K_u,\theta_m\gamma)\times F_u(\kappa_u)$.\ The simulation in SRW are performed the frequency field frequency within PU-radiation are propagated and amplitude in the time domain inside computed [@light_optics_ipac16]. This is first done for the of a focusing lens using $L_u$ and $\lambda_o$ Tab. \[tab:TLT\], but varying $N_u$ and other parameters For this benchmarking simulation, the distance between the PU and KU centers is taken to m (i.e. $L_t\gg L_u$) order to suppress depth-of-field and focal of the is $f=L_t/2$. | _u,
\label{small_K_kick}\end{aligneD}$$ where $N_u$ is The nuMbeR of UnDulaTor pEriods. IntuitivELy Eq. \[Small\_K\_kick\] is just equal tO the tOtAL eneRGy Loss aS the eleCTrON TraVeLs ThrOuGH oNe undUlaTor. When $k_u$ is increaSed (ThEreby resultiNG iN an increasEd aNgular deflecTioN) and thE fIniTE anguLar AccepTance oF The lenS, $\theta_m$, taKeN Into acCOunt, the ON-AxIs elEctric field $E_x(x=y=0)$ in THe ku is reduced by a fActor $F_H(K_U,\GaMMA \thEta_M)\leq1 $. The expReSsion OF $F_h(K_u,\gaMMa \THETa_m)$ IS derived in [@OSC_Val] and its dePEndEnce on $k_u$ AppEArs in FIg. \[fig1\] FoR ThrEe cases of $\gaMma \tHeta_m$. TherE is an aDDitionaL EfficieNcy facTor, $f_u(\kAppa_U)=j_0(\kApPa_u)-j_1(\kAPpa_U)$, WhIch ACcoUnts the eFfEcT of thE lonGITUDinaL osCillAtion \[Given by $\frac{K_u^2}{8\GamMa^2k_u}\SIn(2k_U z)$\] of tHe parTiclE iN the Ku where $\Kappa_U\eQuiv K_u^2/4(1+K_u^2/2)$. The kick AmplItude from eq. \[sMaLl\_K\_KiCk\] is tHUs reduCed By tHe factoR of $F_h(K_u,\THetA_m\GAMMa)\Times F_u(\kappa_u)$.\
The siMuLATiOn in SRW aRe perfORmEd IN the freqUeNcy DomaIN: The fiEld fREqUency comPonentS WiThIn the PU-RaDiatioN bAndWidTh are PRopaGated aNd the fieLd ampLItude in the time DOmain inside thE kU IS CoMPuteD [@liGht_optics_ipAc16]. ThIS is fIrst DOnE foR The caSe of a SiNGlE Focusing lens using $L_u$ AnD $\lambdA_o$ froM Tab. \[tab:TLT\], but Varying $N_u$ aND OTher paraMeteRS aPPropriately. For This bEnchmarkinG SimulatiOn, the Distance Between thE pu and KU ceNteRs iS taKen TO Be $l_t=19.5$ m (i.e. $L_t\gg L_u$) in ORDer tO sUppress The Depth-of-FieLd eFfeCt aNd The focal lEngth of tHe LeNs Is $F=L_t/2$. | _u,
\label{small_K_kick}\ end{aligne d}$$whe re$N _u$is t he number of u n dula tor periods. Intuitive ly Eq .\ [sma l l\ _K\_k ick\] i s j u s t e qu al to t h etotal en ergy lo ss as theele ct ron travelst hr ough one u ndu lator. When$K_ u$ isin cre a sed ( the rebyresult i ng inan increa se d angul a r defle c t io n) a nd the finite ang u la r acceptance of the l en s ,$ \ the ta_ m$, takenin to ac c ount, t h eo n - axi s electric fie ld $E_x(x=y = 0)$ in th eKUi s redu ced b ya fa ctor $F_h(K _u,\ gamma \th eta_m) \ leq1 $. The exp ressio n o f $ F_h( K _u ,\ gam ma \th e ta _m) $ is derived i n[@OSC _val ] a n d it s d epen dence on $K_u$ app ear s in Fig . \[f ig1\] for t hreecasesof $\ ga mma \theta_m$.Ther e is an a ddi ti ona leffic i ency f act or, $F_u(\ kappa_u ) =J_ 0( \ k a pp a_u)-J_1(\kappa_u) $, w hi ch accou nts th e e ff e ct of th elon gitu d i nal o scil l at ion \[gi ven by $\ fr ac{K_u^ 2} {8\gam ma ^2k _u} \sin( 2 k_uz)$\]of the p artic l e in the KU wh e re $\kappa_u\ e qu i v K _ u^2/ 4(1 +K_u^2/2)$. The kick amp l it ude fromEq. \ [s m al l \_K\_kick\] is thus r educed by t he factor of$F_h(K_u,\ t h e ta_m\gam ma)\ t im e s F_u(\kappa_u )$.\The simula t ion in S RW ar e perfor med in th e frequenc y d oma in: th e fi eld frequency c ompo ne nts wit hin the PU -ra dia tio n b an dwidth ar e propag at ed a nd th e fie l d amplit ud e i nthe time domain insi de t he K U is comput e d[ @ ligh t_ op tics _ip ac 16].This isfirst d one for t hec aseof a single focusing len susing $L_u $and $\lam b d a_o$ fro m Tab. \[tab:TLT\], but varying $N _u$ a nd o ther para met ers ap pro p riatel y. For this b enc h m arkin g si mul at ion, the d i s tan ce be tw eenthe PUand KU centers ist ake n to be $L_t= 19. 5$ m ( i. e.$ L_ t \gg L _ u$) i n order to supp ress the d ep t h- of-field e f fec tand the focallengt h of the lens is$f=L_t/2$ . | _u,
\label{small_K_kick}\end{aligned}$$_where $N_u$_is the number of_undulator periods._Intuitively_Eq. \[small\_K\_kick\]_is_just equal to_the total energy_loss as the electron_travels through one_undulator._When $K_u$ is increased (thereby resulting in an increased angular deflection) and the finite_angular_acceptance of_the_lens,_$\theta_m$, taken into account, the_on-axis electric field $E_x(x=y=0)$ in_the KU_is reduced by a factor $F_h(K_u,\gamma \theta_m)\leq1 $._The_expression of $F_h(K_u,\gamma_\theta_m)$ is derived in [@OSC_val] and its dependence on_$K_u$ appears in Fig. \[fig1\] for three_cases of $\gamma_\theta_m$._There_is an additional efficiency_factor, $F_u(\kappa_u)=J_0(\kappa_u)-J_1(\kappa_u)$, which accounts the effect_of the longitudinal oscillation \[given by_$\frac{K_u^2}{8\gamma^2k_u}\sin(2k_u z)$\] of the particle in the_KU where $\kappa_u\equiv K_u^2/4(1+K_u^2/2)$. The kick_amplitude from Eq. \[small\_K\_kick\] is_thus reduced_by the factor of $F_h(K_u,\theta_m\gamma)\times_F_u(\kappa_u)$.\
The simulation in_SRW are_performed in the_frequency domain: the field frequency components_within the PU-radiation_bandwidth are propagated and the field_amplitude_in the time_domain_inside_the KU_is computed [@light_optics_ipac16]._This_is first_done_for the case of a single_focusing_lens using $L_u$ and $\lambda_o$ from Tab. \[tab:TLT\],_but varying $N_u$ and_other_parameters appropriately. For this_benchmarking simulation, the distance between_the PU and KU centers is_taken to_be $L_t=19.5$ m_(i.e. $L_t\gg L_u$) in order to suppress the depth-of-field effect and_the focal length of the lens_is $f=L_t/2$. |
onso99] relations, calibrated on the infrared flux method, are used to derive effective temperatures based on $b-y$, $v-y$, $B-V$, and $V-I$. Note that, at this metallicity and stellar-parameter space, the corresponding relations of @Ramirez05 produce very similar effective temperatures (see Paper I). There are two Alonso et al. calibrations for each index, one suitable for main sequence stars of spectral types F0-K5 [@Alonso96] and one for giant stars in the same spectral range [@Alonso99][^5]. Since our targets range from TO stars at the very end of the MS to bRGB stars, both the dwarf and giant calibrations are used to separately derive effective temperatures. Note that by doing so, we also apply the $T_{\rm eff}-\rm colour$ relations to stars that do not fall in the range in colour covered by the calibrations at this metallicity, which may affect the results.\
Before applying the calibrations, we eliminate star-to-star scatter caused by errors in the observed colours. We clean the photometric sample based on data quality, as measured by the DAPHOT $SHARP$ parameter, and construct fiducial relations between each colour index and the $V$ magnitude, which is the best observed and calibrated quantity. The relations are obtained by averaging the colours for stars in $V$ bins of 0.22 magnitudes. We then shift the observed colour for each star onto the sequence and, finally, we correct all colours for the reddening of NGC 6397. This value has been measured in several studies and typical estimates lie in the range $E(B-V)=0.17-0.20$ (see e.g. @Reid98 1998). Here a value of $0.179$ is adopted, following @AnthonyTwarog00. The reddening in the other colour indices are derived from $E(B-V)$, using the relation coefficients given in @Ramirez05 and $E(v-y)=1.7\times E(b-y)$.\
Fig. \[fig:temp\] shows a comparison between our derived spectroscopic and photometric effective temperatures for 122 stars (all stars for which the H$\alpha$ line is observed). The difference between the H$\alpha$-based values and the effective temperatures obtained from the dwarf and giant calibrations of the narrow-band index $v-y$ and the broad-band index $ | onso99 ] relations, calibrated on the infrared flux method, are use to deduce effective temperatures based on $ boron - y$, $ vanadium - y$, $ B - V$, and $ V - I$. Note that, at this metallicity and leading - parameter space, the corresponding relation of @Ramirez05 produce very similar effective temperature (see Paper I). There are two Alonso et al. calibration for each index, one suitable for main sequence stars of spectral types F0 - K5 [ @Alonso96 ] and one for elephantine stars in the same spectral image [ @Alonso99][^5 ]. Since our targets range from TO star at the very goal of the MS to bRGB stars, both the dwarf and giant calibrations are used to individually derive effective temperatures. Note that by doing so, we also give the $ T_{\rm eff}-\rm colour$ relation back to stars that do not fall in the range in colour covered by the calibrations at this metallicity, which may involve the results.\
Before apply the calibrations, we eliminate star topology - to - asterisk scatter caused by errors in the observed color. We clean the photometric sample based on data timbre, as measured by the DAPHOT $ SHARP$ parameter, and construct fiducial relations between each colour index and the $ V$ magnitude, which is the best observed and calibrated quantity. The sexual intercourse are obtain by averaging the colours for stars in $ V$ bins of 0.22 magnitudes. We then lurch the observed colour for each star onto the sequence and, finally, we correct all colours for the reddening of NGC 6397. This value has been measured in respective studies and typical estimates lie in the range $ E(B - V)=0.17 - 0.20 $ (attend e.g. @Reid98 1998). Here a value of $ 0.179 $ is adopt, following @AnthonyTwarog00. The reddening in the early colour indices are derive from $ E(B - V)$, using the relation coefficient given in @Ramirez05 and $ E(v - y)=1.7\times E(b - y)$.\
Fig. \[fig: temp\ ] shows a comparison between our derived spectroscopic and photometric effective temperatures for 122 stars (all stars for which the H$\alpha$ line is observed). The difference between the H$\alpha$-based values and the effective temperatures obtained from the gnome and giant calibrations of the narrow-minded - band index $ v - y$ and the broad - band index $ | onsl99] relations, calibrated ok the infrared flux metiod, are used to derive effective temperaturxs bqsed in $b-y$, $v-y$, $B-V$, and $V-I$. Notd that, at this merallmcity and stellac-larametcx spadc, the rorresponding rglations of @Samirez05 produca xexy similar effective temperatures (seq Paper I). There are two Alomfo ef al. calibrations for each index, ons suitaule for main seauence stars of spectral tjpes F0-K5 [@Alonso96] and one for giant wtarf in the same spectral gcnge [@Alonso99][^5]. Since our targets range from TU staxs at the vgxt ejg of the MS to bGGB stars, botm the dfarf anc giant calibrstimns are used to separateny derive effectivg temperatgrzs. Note that by doing si, we dlso apput tfe $U_{\rm erf}-\rm cllonr$ relationa to stars rhat do not fall in trv range in comour cjvqred by the calibrations at this metallpcitg, which may affect the eesults.\
Before applyinh the calybrations, we eliminate star-to-star scatter caused ty ercofs nk gye observed colours. We clean the photometric saillt bssed on data qmality, as measured bj yre DAPHOT $SHATP$ parametsr, and construct flducial relarions betreen each colour index and the $C$ magnitude, chixh is the best obszrved and cauibrsted auantity. The relations cre obfained by ageraging ffe colours for sgarx hn $V$ bins of 0.22 magnitudes. Wq then shmft tke obserxed volour for each dtar onto the sequence ajd, fiually, we correch all colours for the reddening of NGC 6397. This fanue has beeu measmred in several studies and tipical escimater lie in tge rangx $E(B-V)=0.17-0.20$ (see e.g. @Weid98 1998). Here a ewlue of $0.179$ is edopted, fjlloqing @AnthonhGwarog00. The redcening in the other colour indices arc derkbed from $E(B-V)$, usnug the relation cpefwicyejtx dhven in @Ramisez05 xnd $R(v-y)=1.7\tioes E(b-y)$.\
Fig. \[flg:tdmp\] xhows a comparison batwesn our derived spevtvoscopic qnd photjmetric effecyive temperatures vor 122 vtacs (all stwrs for which the H$\alpha$ line js observfd). Bhe differencq bebweeg the H$\alpka$-based values and the effective temperavures obtained from thg dwarf and giant calnbvations of tie narwow-band itdex $v-y$ and the broae-band index $ | onso99] relations, calibrated on the infrared flux used derive effective based on $b-y$, that, this metallicity and space, the corresponding of @Ramirez05 produce very similar effective (see Paper I). There are two Alonso et al. calibrations for each index, suitable for main sequence stars of spectral types F0-K5 [@Alonso96] and one for stars the spectral [@Alonso99][^5]. Since our targets range from TO stars at the very end of the MS to stars, both the dwarf and giant calibrations are to separately derive effective Note that by doing so, also the $T_{\rm colour$ to that do not in the range in colour covered by the calibrations at this metallicity, which may affect the results.\ applying the eliminate star-to-star caused errors the observed colours. the photometric sample based on data by the DAPHOT $SHARP$ parameter, and construct fiducial between each index and the $V$ magnitude, which the best observed and calibrated quantity. The relations obtained by averaging the colours for stars in $V$ bins of 0.22 magnitudes. We then observed colour for each onto the sequence finally, correct colours the reddening NGC 6397. This value has been measured in several studies and estimates lie in the range $E(B-V)=0.17-0.20$ (see e.g. @Reid98 1998). value $0.179$ is adopted, @AnthonyTwarog00. The reddening in other indices are derived from the coefficients and E(b-y)$.\ \[fig:temp\] shows a comparison our derived spectroscopic and photometric temperatures for 122 stars H$\alpha$ line is observed). The difference between the values and the effective temperatures obtained from dwarf and giant calibrations of the narrow-band index $v-y$ and the broad-band $ | onso99] relations, calibrated on The infrareD flux MetHod, ArE useD to dErive effective TEmpeRatures based on $b-y$, $v-y$, $B-V$, anD $V-I$. NoTe THat, aT ThIs metAllicitY AnD STelLaR-pAraMeTEr Space, The CorrespOnding relaTioNs Of @Ramirez05 proDUcE very similAr eFfective tempEraTures (sEe papER I). TheRe aRe two alonso ET al. calIbrations FoR Each inDEx, one suITAbLe foR main sequence starS Of SPectral types F0-K5 [@alonso96] AnD OnE FOr gIanT stars in thE sAme spECtral raNGe [@aLONso99][^5]. sInce our targetS range from To StaRs at thE vEry ENd of thE MS to BRgb stArs, both the dWarf And giant cAlibraTIons are USed to seParateLy dEriVe efFEcTiVe tEmPEraTUrEs. NOTe tHat by doiNg So, We alsO appLY THE $T_{\rm Eff}-\Rm coLour$ rElations to staRs tHat dO Not Fall iN the rAnge In ColouR coverEd by tHe Calibrations at tHis mEtallicitY, whIcH maY aFfect THe resuLts.\
befOre applYing the CAliBrATIOnS, we eliminate star-to-StAR ScAtter cauSed by eRRoRs IN the obseRvEd cOlouRS. we cleAn thE PhOtometriC samplE BaSeD on data QuAlity, aS mEasUreD by thE dAPHoT $SHARp$ parametEr, and COnstruct fiduciAL relations betWEeN EAcH ColoUr iNdex and the $V$ MagnITude, WhicH Is The BEst obServeD aND cALibrated quantity. The ReLationS are oBtained by averAging the coLOURs for staRs in $v$ BiNS of 0.22 magnitudes. WE then Shift the obSErved colOur foR each staR onto the sEQUence and, FinAllY, we CorRECt All colours for THE redDeNing of NgC 6397. THis valuE haS beEn mEasUrEd in severAl studieS aNd TyPiCal EstimATes lie in ThE raNgE $E(B-v)=0.17-0.20$ (see e.G. @reid98 1998). HeRe a vaLue oF $0.179$ iS aDOptEd, folloWInG @aNthoNytwArog00. the ReDdeniNg in THe oTher colOur indiceS arE DeriVeD fRom $E(B-V)$, uSing the relatiOn CoefficienTs GivEn in @RaMIRez05 and $E(v-Y)=1.7\times E(b-y)$.\
Fig. \[fig:temp\] showS A comparIsoN betwEen oUr derived SpeCtroscOpiC And phoTometrIc effEcTivE TEmperATUrEs fOr 122 Stars (all stARS foR whicH tHe H$\aLpha$ linE is observed). The diffERenCe between the H$\AlpHa$-baSED vAluES aND thE eFFecTIVe temperatures oBtained froM tHE dWarf and giaNT caLiBrationS of the nArrow-BAnd indeX $v-y$ and the Broad-band InDex $ | onso99] relations, calibra ted on the infr are d f lu x me thod , are used tod eriv e effective temperatur es ba se d on$ b- y$, $ v-y$, $ B -V $ , an d$V -I$ .N ot e tha t,at this metallici tyan d stellar-pa r am eter space , t he correspon din g rela ti ons of @R ami rez05 produ c e very similaref f ective tempera t u re s (s ee Paper I). Ther e a r e two Alonso e t al.ca l ib r a tio nsfor each i nd ex, o n e suita b le f o r m a in sequence s tars of spe c tra l type sF0- K 5 [@Al onso9 6] and one for gi antstars inthe sa m e spect r al rang e [@Al ons o99 ][^5 ] .Si nce o u r t a rg ets ran ge fromTO s tarsat t h e v eryend ofthe M S to bRGB sta rs, bot h th e dwa rf an d gi an t cal ibrati ons a re used to separa tely derive e ffe ct ive t emper a tures. No tethat by doings o,we a l so apply the $T_{\rm e f f }- \rm colo ur$ re l at io n s to sta rs th at d o not f alli nthe rang e in c o lo ur covere dby the c ali bra tions at t his me tallicit y, wh i ch may affectt he results.\B ef o r ea pply ing the calibr atio n s, w e el i mi nat e star -to-s ta r s c atter caused by err or s in t he ob served colour s. We clea n t he photo metr i cs ample based on data quality,a s measur ed by the DAP HOT $SHAR P $ paramet er, an d c ons t r uc t fiducial re l a tion sbetween ea ch colo urind exand t he $V$ ma gnitude, w hi ch i s t he be s t observ ed an dcal ibrat e d quan tity. The r el a tio ns areo bt a i nedby a vera gin gthe c olou r s f or star s in $V$bin s of0. 22 magnit udes. We then s hift the o bs erv ed col o u r for ea ch star onto the sequen c e and,fin ally, wecorrect a llcolour s f o r theredden ing o fNGC 6 397.T h is va lu e has been m eas uredin sev eral st udies and typicale sti mates lie inthe ran g e $ E(B - V) = 0.1 7- 0 .20 $ (see e.g. @Reid 98 1998).He r ea value of $0. 17 9$ is a dopted, foll o wing @A nthonyTwa rog00. Th eredd e n ing in the ot her colo ur indice s ared er ivedfro m $E(B -V )$, usin g ther ela tioncoeffi ci ents g ivenin @Ramire z05 and $E(v-y)=1.7\tim es E(b -y)$. \
F ig. \[fig :te m p\] shows acomp arison bet wee n o ur de riv e d spe ctro s co pic and p hoto m etric eff e ct ive t em peratures f o r 122 star s ( a ll sta rs f or which the H$\a l pha$ line is o bser v e d). Th e dif fe rence betweenthe H $ \ alpha$-b as ed values a nd the e ff e ctive tempe rature s obtai n e df rom th e dw arf and gian t c al i bration sof the na rrow -b and in dex $v - y$ a n d the broad-bandindex $ | onso99] relations,_calibrated on_the infrared flux method,_are used_to_derive effective_temperatures_based on $b-y$,_$v-y$, $B-V$, and_$V-I$. Note that, at_this metallicity and_stellar-parameter_space, the corresponding relations of @Ramirez05 produce very similar effective temperatures (see Paper I)._There_are two_Alonso_et_al. calibrations for each index, one_suitable for main sequence stars_of spectral_types F0-K5 [@Alonso96] and one for giant stars_in_the same spectral_range [@Alonso99][^5]. Since our targets range from TO stars_at the very end of the_MS to bRGB_stars,_both_the dwarf and giant_calibrations are used to separately derive_effective temperatures. Note that by doing_so, we also apply the $T_{\rm eff}-\rm_colour$ relations to stars that do_not fall in the range_in colour_covered by the calibrations at_this metallicity, which_may affect_the results.\
Before applying_the calibrations, we eliminate star-to-star scatter_caused by errors_in the observed colours. We clean_the_photometric sample based_on_data_quality, as_measured by the_DAPHOT_$SHARP$ parameter,_and_construct fiducial relations between each colour_index_and the $V$ magnitude, which is the_best observed and calibrated_quantity._The relations are obtained_by averaging the colours for_stars in $V$ bins of 0.22_magnitudes. We_then shift_the observed colour for each star onto the sequence and, finally,_we correct all colours for the_reddening of NGC 6397._This value_has_been measured in_several_studies and_typical estimates lie in the range $E(B-V)=0.17-0.20$_(see e.g. @Reid98_1998). Here a value of $0.179$_is adopted, following @AnthonyTwarog00._The_reddening in the other colour indices_are derived from $E(B-V)$, using the_relation coefficients given in @Ramirez05_and_$E(v-y)=1.7\times_E(b-y)$.\
Fig. \[fig:temp\] shows a comparison between_our derived spectroscopic and photometric effective_temperatures for 122_stars (all stars for which the H$\alpha$_line_is observed). The difference between the_H$\alpha$-based_values and the effective temperatures obtained_from_the_dwarf and giant calibrations of_the narrow-band index $v-y$ and the_broad-band index $ |
X{\longrightarrow}X\cup_A X$ is the embedding of the first copy of $X$. The Mayer-Vietoris sequence for the pushout square $$\xymatrix{ A \ar[r]^-i\ar[d]_i & X\ar[d]^{i_2} \\ X\ar[r]_-{i_1} & X\cup_A X}$$ has the form $$\begin{aligned}
\cdots\ {\longrightarrow}\
{\mathcal H}^{k-1}(A)\ &\xrightarrow{\ \delta^{k-1}\ } \
{\mathcal H}^k(X\cup_A X)\ \xrightarrow{(i_1^*,i_2^*)}\\
&{\mathcal H}^k(X)\times {\mathcal H}^k(X)\ \xrightarrow{(x,x')\mapsto i^*(x)-i^*(x')}
\ {\mathcal H}^k(A) \ \xrightarrow{\ \delta^k\ } \ \cdots \.\end{aligned}$$ So the connecting homomorphism $\delta^{k-1}:{\mathcal H}^{k-1}(A){\longrightarrow}{\mathcal H}^k(X\cup_A X)$ lands in the subgroup ${\mathcal H}^k(X,A)$, and the exact sequence splits off a long exact sequence $$\label{eq:les proper coho}
\cdots\ {\longrightarrow}\
{\mathcal H}^{k-1}(A)\ \xrightarrow{\ \delta^{k-1}\ } \
{\mathcal H}^k(X,A)\ \xrightarrow{\ r\ }\
{\mathcal H}^k(X)\ \xrightarrow{\ i^*\ }\ {\mathcal H}^k(A) \ {\longrightarrow}\ \cdots\.$$
For every orthogonal $G$-spectrum $E$, the functor $E_G(-)=[-,E]^G\circ \Sigma^\infty_+$ is excisive by Proposition \[prop:represented is excisive\]. For every finite proper $G$-CW-pair $(X,A)$, the relative group $E_G(X,A)$ is defined as in Definition \[def:relative\]. This relative group can in fact be described more directly as a morphism group in $\operatorname{Ho}(\operatorname{Sp}_G)$: we let $q:X\cup_A X{\longrightarrow}X/A$ denote the map that sends the second copy | X{\longrightarrow}X\cup_A X$ is the embedding of the first copy of $ X$. The Mayer - Vietoris sequence for the pushout square $ $ \xymatrix { A \ar[r]^-i\ar[d]_i & X\ar[d]^{i_2 } \\ X\ar[r]_-{i_1 } & X\cup_A X}$$ take the human body $ $ \begin{aligned }
\cdots\ { \longrightarrow}\
{ \mathcal H}^{k-1}(A)\ & \xrightarrow{\ \delta^{k-1}\ } \
{ \mathcal H}^k(X\cup_A X)\ \xrightarrow{(i_1^*,i_2^*)}\\
& { \mathcal H}^k(X)\times { \mathcal H}^k(X)\ \xrightarrow{(x, x')\mapsto i^*(x)-i^*(x') }
\ { \mathcal H}^k(A) \ \xrightarrow{\ \delta^k\ } \ \cdots \.\end{aligned}$$ So the connecting homomorphism $ \delta^{k-1}:{\mathcal H}^{k-1}(A){\longrightarrow}{\mathcal H}^k(X\cup_A X)$ lands in the subgroup $ { \mathcal H}^k(X, A)$, and the accurate sequence splits off a long accurate succession $ $ \label{eq: les proper coho }
\cdots\ { \longrightarrow}\
{ \mathcal H}^{k-1}(A)\ \xrightarrow{\ \delta^{k-1}\ } \
{ \mathcal H}^k(X, A)\ \xrightarrow{\ r\ } \
{ \mathcal H}^k(X)\ \xrightarrow{\ i^*\ } \ { \mathcal H}^k(A) \ { \longrightarrow}\ \cdots\.$$
For every orthogonal $ G$-spectrum $ E$, the functor $ E_G(-)=[-,E]^G\circ \Sigma^\infty_+$ is excisive by Proposition \[prop: represented is excisive\ ]. For every finite proper $ G$-CW - pair $ (X, A)$, the proportional group $ E_G(X, A)$ is defined as in Definition \[def: relative\ ]. This relative group can in fact be described more directly as a morphism group in $ \operatorname{Ho}(\operatorname{Sp}_G)$: we lease $ q: X\cup_A X{\longrightarrow}X / A$ denote the map that sends the second copy | X{\lojgrightarrow}X\cup_A X$ is tme embedding of jhw firsv copy kf $X$. The Mayer-Vietoris sequence for vhe pushoyt square $$\xymatrix{ A \xr[r]^-i\ar[d]_i & X\ar[d]^{i_2} \\ Z\ar[r]_-{m_1} & X\cup_A X}$$ has tis form $$\nzgin{amlgned}
\cdots\ {\longrlghtarrow}\
{\mathcal H}^{k-1}(A)\ &\xfiyhtarrow{\ \delta^{k-1}\ } \
{\mathcal H}^k(X\cup_W X)\ \xrogjtarrow{(i_1^*,i_2^*)}\\
&{\mwthcsj H}^k(S)\nines {\mathcal H}^k(X)\ \xrightarrow{(x,x')\japsto p^*(x)-i^*(x')}
\ {\mathcal H}^k(A) \ \xrightarrow{\ \delta^k\ } \ \cdohs \.\end{aligned}$$ So tje connectibg hjnomorphism $\ddlta^{k-1}:{\mathcal H}^{k-1}(A){\longrifhtarrow}{\mathcal H}^k(X\cup_A X)$ lands in tke subgroup ${\mqthfdl H}^k(X,A)$, and the vxact sequencc splitv off a long exact seaueice $$\oabel{eq:les proper cohm}
\cdots\ {\longrigrtarrow}\
{\mathcal H}^{k-1}(A)\ \xrigytqrrow{\ \delda^{k-1}\ } \
{\mzticam H}^k(X,A)\ \xcightarrow{\ d\ }\
{\mathcql H}^k(X)\ \xrightarrow{\ o^*\ }\ {\mathcal H}^k(A) \ {\longrygrtarrow}\ \cdots\.$$
For every orthogonal $G$-specurum $S$, the functor $E_G(-)=[-,E]^G\circ \Wigma^\infty_+$ is excisivg by Propofition \[prop:represented is excisive\]. For every finide prkoer $G$-GW-pake $(D,A)$, the relative group $E_G(X,A)$ is defined as in Dqriminion \[def:relative\]. Bhis relative groul fam in fact be dgscribeb mkre directly as a lorphisi groyp in $\opewatotname{Ho}(\operatorname{Sp}_G)$: we lwt $q:X\cup_A X{\ljbgrightarrow}X/A$ denlte the map thaj sendx the second copy | X{\longrightarrow}X\cup_A X$ is the embedding of the of The Mayer-Vietoris for the pushout X\ar[d]^{i_2} X\ar[r]_-{i_1} & X\cup_A has the form \cdots\ {\longrightarrow}\ {\mathcal H}^{k-1}(A)\ &\xrightarrow{\ \delta^{k-1}\ \ {\mathcal H}^k(X\cup_A X)\ \xrightarrow{(i_1^*,i_2^*)}\\ &{\mathcal H}^k(X)\times {\mathcal H}^k(X)\ \xrightarrow{(x,x')\mapsto i^*(x)-i^*(x')} \ {\mathcal \ \xrightarrow{\ \delta^k\ } \ \cdots \.\end{aligned}$$ So the connecting homomorphism $\delta^{k-1}:{\mathcal H}^{k-1}(A){\longrightarrow}{\mathcal X)$ in subgroup H}^k(X,A)$, and the exact sequence splits off a long exact sequence $$\label{eq:les proper coho} \cdots\ {\longrightarrow}\ H}^{k-1}(A)\ \xrightarrow{\ \delta^{k-1}\ } \ {\mathcal H}^k(X,A)\ \xrightarrow{\ }\ {\mathcal H}^k(X)\ \xrightarrow{\ }\ {\mathcal H}^k(A) \ {\longrightarrow}\ For orthogonal $G$-spectrum the $E_G(-)=[-,E]^G\circ is excisive by \[prop:represented is excisive\]. For every finite proper $G$-CW-pair $(X,A)$, the relative group $E_G(X,A)$ is defined as in \[def:relative\]. This can in be more as a morphism $\operatorname{Ho}(\operatorname{Sp}_G)$: we let $q:X\cup_A X{\longrightarrow}X/A$ denote sends the second copy | X{\longrightarrow}X\cup_A X$ is thE embedding Of the FirSt cOpY of $X$. the MAyer-Vietoris seQUencE for the pushout square $$\xyMatriX{ A \AR[r]^-i\aR[D]_i & x\ar[d]^{i_2} \\ x\ar[r]_-{i_1} & X\cUP_A x}$$ HAs tHe FoRm $$\bEgIN{aLigneD}
\cdOts\ {\longRightarrow}\
{\MatHcAl H}^{k-1}(A)\ &\xrightaRRoW{\ \delta^{k-1}\ } \
{\matHcaL H}^k(X\cup_A X)\ \xriGhtArrow{(i_1^*,I_2^*)}\\
&{\mAthCAl H}^k(X)\TimEs {\matHcal H}^k(x)\ \XrightArrow{(x,x')\maPsTO i^*(x)-i^*(x')}
\ {\mAThcal H}^k(a) \ \XRiGhtaRrow{\ \delta^k\ } \ \cdots \.\enD{AlIGned}$$ So the conneCting hOmOMoRPHisM $\deLta^{k-1}:{\mathcaL H}^{K-1}(A){\lonGRightarROw}{\MATHcaL h}^k(X\cup_A X)$ lands In the subgroUP ${\maThcal H}^K(X,a)$, anD The exaCt seqUeNCe sPlits off a loNg exAct sequenCe $$\labeL{Eq:les prOPer coho}
\Cdots\ {\lOngRigHtarROw}\
{\MaThcAl h}^{K-1}(A)\ \xRIgHtaRRow{\ \Delta^{k-1}\ } \
{\maThCaL H}^k(X,A)\ \XrigHTARRow{\ r\ }\
{\MatHcal h}^k(X)\ \xrIghtarrow{\ i^*\ }\ {\matHcaL H}^k(A) \ {\LOngRightArrow}\ \CdotS\.$$
FOr eveRy orthOgonaL $G$-Spectrum $E$, the funCtor $e_G(-)=[-,E]^G\circ \SIgmA^\iNftY_+$ iS exciSIve by PRopOsiTion \[proP:represENteD iS EXCiSive\]. For every finite PrOPEr $g$-CW-pair $(X,a)$, the reLAtIvE Group $E_G(X,a)$ iS deFineD AS in DeFiniTIoN \[def:relaTive\]. ThIS rElAtive grOuP can in FaCt bE deScribED morE direcTly as a moRphisM Group in $\operatoRName{Ho}(\operatoRNaME{sp}_g)$: We leT $q:X\Cup_A X{\longriGhtaRRow}X/a$ denOTe The MAp thaT sendS tHE sECond copy | X{\longrightarrow}X\cup_AX$ is theembed din g o fthefirs t copy of $X$. TheMayer-Vietoris sequenc e for t h e pu s ho ut sq uare $$ \ xy m a tri x{ A \ ar [ r] ^-i\a r[d ]_i & X \ar[d]^{i_ 2}\\ X\ar[r]_-{i _ 1} & X\cup_A X} $$ has the f orm $$\be gi n{a l igned }
\c dots\{ \longr ightarrow }\
{ \ mathcal H }^ {k-1 }(A)\ &\xrightarr o w{ \ \delta^{k-1}\ } \ { \ m ath cal H}^k(X\cu p_ A X)\ \xrigh t ar r o w {(i _ 1^*,i_2^*)}\\
&{\mat h cal H}^k( X) \ti m es {\m athca lH }^k (X)\ \xrigh tarr ow{(x,x') \mapst o i^*(x) - i^*(x') }
\ {\m athc a lH} ^k( A) \ \ x ri ght a rro w{\ \del ta ^k \ } \ \cd o t s \.\e nd{ alig ned}$ $ So the conn ect ingh omo morph ism $ \del ta ^{k-1 }:{\ma thcal H }^{k-1}(A){\lon grig htarrow}{ \ma th cal H }^k(X \ cup_AX)$ la nds inthe sub g rou p$ { \ ma thcal H}^k(X,A)$,an d th e exactsequen c esp l its offalon g ex a c t seq uenc e $ $\label{ eq:les pr op er coho } \c do ts\ {\ longr i ghta rrow}\
{\ mathc a l H}^{k-1}(A)\ \xrightarrow{ \ \ d e lt a ^{k- 1}\ } \
{ \mat h calH}^k ( X, A)\ \xri ghtar ro w {\ r\ }\
{\mathcal H }^k(X) \ \xr ightarrow{\ i ^*\ }\ {\ m a t hcal H}^ k(A) \{ \longrightarro w}\ \ cdots\.$$For ever y ort hogonal$G$-spect r u m $E$, t hefun cto r $ E _ G( -)=[-,E]^G\ci r c \Si gm a^\inft y_+ $ is ex cis ive by Pr op osition \ [prop:re pr es en te d i s exc i sive\].Fo r e ve ryfinit e prope r $G$ -CW- pa ir $(X ,A)$, t h er e lati ve g roup $E _G (X,A) $ is def ined as in Defin iti o n \[ de f: relativ e\]. This rel at ive groupca n i n fact b e descri bed more directly as am orphism gr oup i n $\ operatorn ame {Ho}(\ ope r atorna me{Sp} _G)$: w e l e t $q:X \ c up _AX{ \longright a r row }X/A$ d enot e the m ap that sends thes eco nd copy | X{\longrightarrow}X\cup_A X$_is the_embedding of the first_copy of_$X$._The Mayer-Vietoris_sequence_for the pushout_square $$\xymatrix{ _A \ar[r]^-i\ar[d]_i & X\ar[d]^{i_2}_\\ X\ar[r]_-{i_1} &_X\cup_A_X}$$ has the form $$\begin{aligned}
\cdots\ {\longrightarrow}\
_{\mathcal_H}^{k-1}(A)\ &\xrightarrow{\_\delta^{k-1}\_}_\
_{\mathcal H}^k(X\cup_A X)\ \xrightarrow{(i_1^*,i_2^*)}\\
_ _ &{\mathcal H}^k(X)\times {\mathcal H}^k(X)\ \xrightarrow{(x,x')\mapsto i^*(x)-i^*(x')}
__ \ _{\mathcal H}^k(A) \ \xrightarrow{\ \delta^k\ } \ \cdots \.\end{aligned}$$_So the connecting homomorphism $\delta^{k-1}:{\mathcal H}^{k-1}(A){\longrightarrow}{\mathcal_H}^k(X\cup_A X)$ lands_in_the_subgroup ${\mathcal H}^k(X,A)$, and_the exact sequence splits off a_long exact sequence $$\label{eq:les proper coho}
_ \cdots\ {\longrightarrow}\
_ {\mathcal H}^{k-1}(A)\ \xrightarrow{\ \delta^{k-1}\_} \
_ {\mathcal_H}^k(X,A)\ \xrightarrow{\ r\ }\
_ _{\mathcal H}^k(X)\_\xrightarrow{\ i^*\ }\_ {\mathcal H}^k(A) \ {\longrightarrow}\ \cdots\.$$
For_every orthogonal $G$-spectrum_$E$, the functor $E_G(-)=[-,E]^G\circ \Sigma^\infty_+$ is_excisive_by Proposition \[prop:represented_is_excisive\]._For every_finite proper $G$-CW-pair_$(X,A)$,_the relative_group_$E_G(X,A)$ is defined as in Definition_\[def:relative\]._This relative group can in fact be_described more directly as_a_morphism group in $\operatorname{Ho}(\operatorname{Sp}_G)$:_we let $q:X\cup_A X{\longrightarrow}X/A$ denote_the map that sends the second_copy |
ometric automorphism $f$ of $\mathscr K_1$ such that $\|y-f(x)\|\leq\e$. This clearly follows from Theorem \[th:iso-uniq\] and the fact that all lines are 1-complemented in all Banach spaces, by Hahn-Banach. That cannot we achieved for $\e=0$ since the unit sphere of $\mathscr K_1$ contains (many) points where the norm is smooth and points where it is not (think of an isometric copy of, say, $\ell_\infty^2$) and note that a surjective isometry must preserve both classes.
In contrast, there is no equivalent $p$-norm rendering $\mathscr K_p$ almost isotropic when $p<1$. For if $X$ is almost isotropic and linearly isomorphic with $\mathscr K_p$, then the functional $$|x|=\|x\|+\sup_{\|x^*\|\leq 1}|x^*(x)|$$ is another $p$-norm that is preserved by every isometry for the original $p$-norm of $X$. It quickly follows (cf. [@maximal Theorem 3.3] for the complete argument) that $|x|=2\|x\|$ for all $x\in X$ and so $\|x\|=\sup_{\|x^*\|\leq 1}|x^*(x)|$, that is, $X$ is locally convex, which is not the case.
Wheeling around $\e$
--------------------
It is clear that moving the number $\e$ from here to there in the definitions opening Section 3 one obtains other variants that are more or less equivalent to these appearing in the text.
Actually the version of property $[\Game]$ and the definition of a space of almost universal complemented disposition that we have used here do not agree with those of [@GW] and [@CM-ACUD]. The following simple remark shows that $[\Game]$ is equivalent to Garbulińska’s property (E) of [@GW] and that Definition \[def:aucd\] is equivalent to Definition 3.1 in [@CM-ACUD].
\[defec\] Let $f^\dagger:E{\longrightarrow}F$ and $f^\ddagger:F{\longrightarrow}E$ be operators such that $\|f^\ddagger\, f^\dagger-{\bf 1}_E\|\leq \varepsilon$, where $\varepsilon<1$. Then there is an automorphism $a$ of $ | ometric automorphism $ f$ of $ \mathscr K_1 $ such that $ \|y - f(x)\|\leq\e$. This clearly follows from Theorem \[th: iso - uniq\ ] and the fact that all lines are 1 - complemented in all Banach space, by Hahn - Banach. That cannot we achieve for $ \e=0 $ since the unit sphere of $ \mathscr K_1 $ control (many) point where the norm is smooth and decimal point where it is not (think of an isometric copy of, say, $ \ell_\infty^2 $) and notice that a surjective isometry must save both classes.
In line, there is no equivalent $ p$-norm rendering $ \mathscr K_p$ almost isotropic when $ p<1$. For if $ X$ is almost isotropic and linearly isomorphous with $ \mathscr K_p$, then the functional $ $ |x|=\|x\|+\sup_{\|x^*\|\leq 1}|x^*(x)|$$ is another $ p$-norm that is preserved by every isometry for the original $ p$-norm of $ X$. It promptly follows (cf. [ @maximal Theorem 3.3 ] for the accomplished argument) that $ |x|=2\|x\|$ for all $ x\in X$ and so $ \|x\|=\sup_{\|x^*\|\leq 1}|x^*(x)|$, that is, $ X$ is locally convex, which is not the case.
bicycle around $ \e$
--------------------
It is clear that moving the number $ \e$ from here to there in the definitions open Section 3 one obtains other variants that are more or less equivalent to these appearing in the textbook.
Actually the version of property $ [ \Game]$ and the definition of a space of almost universal complemented disposition that we have use here do not harmonize with those of [ @GW ] and [ @CM - ACUD ]. The following simple remark shows that $ [ \Game]$ is equivalent to Garbulińska ’s property (E) of [ @GW ] and that Definition \[def: aucd\ ] is equivalent to Definition 3.1 in [ @CM - ACUD ].
\[defec\ ] Let $ f^\dagger: E{\longrightarrow}F$ and $ f^\ddagger: F{\longrightarrow}E$ be operators such that $ \|f^\ddagger\, f^\dagger-{\bf 1}_E\|\leq \varepsilon$, where $ \varepsilon<1$. Then there is an automorphism $ a$ of $ | omehric automorphism $f$ of $\mxthscr K_1$ such tkqt $\|y-f(x)\|\neq\e$. Tgis cleafly follows from Theorem \[th:isl-ubiq\] abd the fact that all lknes are 1-bomplemenred mn all Banach spedes, by Mchn-Bahwch. Chet cannot we acmieved for $\a=0$ since the unht s'here of $\mathscr K_1$ contains (many) poigts whete the norm is siootn and iolnts where it is not (think of an isometgic copy of, say, $\ekl_\infty^2$) and note that a sugjechive isometry must preserve bith soasses.
In congrast, thert ns no equivzlent $p$-norm rendering $\mathscr K_o$ almpst isotroknx wjgn $p<1$. For if $E$ is ajmost isotroipc and ninearlu isomorphic wlth $\methsxr K_p$, then the functimnal $$|x|=\|x\|+\sup_{\|x^*\|\leq 1}|x^*(x)|$$ ys anothes $'$-norm that is preservwd by eeery isoowtrh fkr tge orihinel $p$-norm of $X$. It quickoy follows (cf. [@maximsl Nneorem 3.3] for tge com[lqte argument) that $|x|=2\|x\|$ for all $x\in X$ and vo $\|s\|=\sup_{\|x^*\|\leq 1}|x^*(x)|$, that is, $X$ iw locally convex, whicj is not ehe case.
Wheeling around $\e$
--------------------
It is clear that moving dhe nnmcer $\e$ weol here to there in the definitions opening Sesfipn 3 one obtains obher variants that age iore or less gquivalzht to these appearinh in thg text.
Qctually uhe vrrsion of property $[\Game]$ and the definitpon if a space of almodt universau cokplemrnted disposition that ce havs used here do not affee with those ow [@GE] dnd [@CM-ACUD]. The following symple remerk skows thag $[\Gake]$ is qquivalent to Garbulińska’s property (E) of [@GF] and that Definition \[def:aucd\] is equivalenv to Definitipn 3.1 it [@CM-ACUD].
\[befec\] Ket $f^\dagger:E{\ljngrightarrow}F$ and $f^\dbagger:W{\longrightzrrow}E$ ue operators such that $\|f^\dgwgger\, f^\daggec-{\bf 1}_E\|\leq \darepsilin$, wherd $\varepsilon<1$. Thrn there ps an qutomorphism $a$ of $ | ometric automorphism $f$ of $\mathscr K_1$ such This follows from \[th:iso-uniq\] and the 1-complemented all Banach spaces, Hahn-Banach. That cannot achieved for $\e=0$ since the unit of $\mathscr K_1$ contains (many) points where the norm is smooth and points it is not (think of an isometric copy of, say, $\ell_\infty^2$) and note a isometry preserve classes. In contrast, there is no equivalent $p$-norm rendering $\mathscr K_p$ almost isotropic when $p<1$. For $X$ is almost isotropic and linearly isomorphic with K_p$, then the functional 1}|x^*(x)|$$ is another $p$-norm that preserved every isometry the $p$-norm $X$. It quickly (cf. [@maximal Theorem 3.3] for the complete argument) that $|x|=2\|x\|$ for all $x\in X$ and so $\|x\|=\sup_{\|x^*\|\leq that is, locally convex, is the Wheeling around $\e$ is clear that moving the number to there in the definitions opening Section 3 obtains other that are more or less equivalent these appearing in the text. Actually the version property $[\Game]$ and the definition of a space of almost universal complemented disposition that we here do not agree those of [@GW] [@CM-ACUD]. following remark that $[\Game]$ equivalent to Garbulińska’s property (E) of [@GW] and that Definition \[def:aucd\] equivalent to Definition 3.1 in [@CM-ACUD]. \[defec\] Let $f^\dagger:E{\longrightarrow}F$ and operators that $\|f^\ddagger\, f^\dagger-{\bf \varepsilon$, where $\varepsilon<1$. Then is automorphism $a$ of $ | ometric automorphism $f$ of $\matHscr K_1$ such tHat $\|y-f(X)\|\leQ\e$. THiS cleArly Follows from TheORem \[tH:iso-uniq\] and the fact that All liNeS Are 1-cOMpLemenTed in alL baNACh sPaCeS, by haHN-BAnach. thaT cannot We achieved For $\E=0$ sInce the unit sPHeRe of $\mathscR K_1$ cOntains (many) pOinTs wherE tHe nORm is sMooTh and Points WHere it Is not (thinK oF An isomETric copY OF, sAy, $\elL_\infty^2$) and note that A SuRJective isometrY must pReSErVE BotH clAsses.
In conTrAst, thERe is no eQUiVALEnt $P$-Norm rendering $\Mathscr K_p$ alMOst IsotroPiC whEN $p<1$. For iF $X$ is aLmOSt iSotropic and LineArly isomoRphic wITh $\mathsCR K_p$, then The funCtiOnaL $$|x|=\|x\|+\sUP_{\|x^*\|\LeQ 1}|x^*(x)|$$ Is ANotHEr $P$-noRM thAt is presErVeD by evEry iSOMETry fOr tHe orIginaL $p$-norm of $X$. It quIckLy foLLowS (cf. [@maXimal theoReM 3.3] for tHe compLete aRgUment) that $|x|=2\|x\|$ for aLl $x\iN X$ and so $\|x\|=\sUp_{\|x^*\|\LeQ 1}|x^*(x)|$, ThAt is, $X$ IS localLy cOnvEx, which Is not thE CasE.
WHEELiNg around $\e$
--------------------
It is clear ThAT MoVing the nUmber $\e$ FRoM hERe to therE iN thE defINItionS opeNInG Section 3 One obtAInS oTher varIaNts thaT aRe mOre Or lesS EquiValent To these aPpearINg in the text.
ActUAlly the versioN Of PROpERty $[\GAme]$ And the definItioN Of a sPace OF aLmoST univErsal CoMPlEMented disposition thAt We have Used hEre do not agree With those oF [@gw] And [@CM-ACUd]. The FOlLOwing simple remArk shOws that $[\GamE]$ Is equivaLent tO GarbuliŃska’s propERTy (E) of [@GW] aNd tHat defIniTIOn \[Def:aucd\] is equiVALent To definitIon 3.1 In [@CM-ACUd].
\[deFec\] let $F^\daGgEr:E{\longriGhtarrow}f$ aNd $F^\dDaGgeR:F{\lonGRightarrOw}e$ be OpEraTors sUCh that $\|F^\ddagGer\, f^\DaGgER-{\bf 1}_e\|\leq \varEPsILOn$, whErE $\vArepSilOn<1$. then tHere IS an AutomorPhism $a$ of $ | ometric automorphism $f$ o f $\mathsc r K_1 $ s uch t hat$\|y -f(x)\|\leq\e$ . Thi s clearly follows from Theo re m \[t h :i so-un iq\] an d t h e fa ct t hat a l llines ar e 1-com plementedinal l Banach spa c es , by Hahn- Ban ach. That ca nno t we a ch iev e d for $\ e=0$sincet he uni t sphereof $\math s cr K_1$ c on tain s (many) points w h er e the norm is s moothan d p o i nts wh ere it isno t (th i nk of a n i s o m etr i c copy of, sa y, $\ell_\i n fty ^2$) a nd no t e that a su rj e cti ve isometry mus t preserv e both classes .
In co ntrast , t her e is no e qui va l ent $p $-n o rmrenderin g$\ maths cr K _ p $ almo stisot ropic when $p<1$.For if$ X$is al mostisot ro pic a nd lin early i somorphic with$\ma thscr K_p $,th enth e fun c tional $$ |x| =\|x\|+ \sup_{\ | x^* \| \ l e q1}|x^*(x)|$$ is an ot h e r$p$-norm thati spr e served b yeve ry i s o metry for th e origin al $p$ - no rm of $X$ .It qui ck lyfol lows( cf.[@maxi mal Theo rem 3 . 3] for the com p lete argument ) t h a t$ |x|= 2\| x\|$ for al l $x \ in X $ an d s o $ \ |x\|= \sup_ {\ | x^ * \|\leq 1}|x^*(x)|$, t hat is , $X$ is locally c onvex, whi c h is not t he c a se .
Wheeling aro und $ \e$
------ - -------- -----
It isclear tha t moving t henum ber $\ e $ f rom here to t h e re i nthe def ini tions o pen ing Se cti on 3 one ob tains ot he rva ri ant s tha t are mor eorle ssequiv a lent t o the se a pp ea r ing in the te x t .
A ct ua llythe v ersio n of pro perty $ [\Game]$and thede fi nitionof a space of a lmost univ er sal compl e m ented di sposition that we haveu sed her e d o not agr ee with t hos e of [ @GW ] and [ @CM-AC UD].Th e f o l lowin g si mpl eremark sho w s th at $[ \G ame] $ is eq uivalent to Garbul i ńsk a’s property(E) of[ @ GW ] a n dt hat D e fin i t ion \[def:aucd\ ] is equiv al e nt to Defini t ion 3 .1 in [ @CM-ACU D].
\ [defec\ ] Let $f^ \dagger:E {\ long r i ght arrow}F$ a nd $f^\d dagger:F{ \ longr i gh tarro w}E $ be o pe rat ors s uch th a t $ \|f^\ ddagge r\ , f^\d agger -{ \bf 1}_E \|\leq \varepsilon$, wh ere $\ varep sil on<1$. Th ent her e is an a utom orphism $a $ o f $ | ometric automorphism_$f$ of_$\mathscr K_1$ such that_$\|y-f(x)\|\leq\e$. This_clearly_follows from_Theorem \[th:iso-uniq\]_and the fact_that all lines_are 1-complemented in all_Banach spaces, by_Hahn-Banach._That cannot we achieved for $\e=0$ since the unit sphere of $\mathscr K_1$ contains_(many)_points where_the_norm_is smooth and points where_it is not (think of_an isometric_copy of, say, $\ell_\infty^2$) and note that a_surjective_isometry must preserve_both classes.
In contrast, there is no equivalent $p$-norm rendering_$\mathscr K_p$ almost isotropic when $p<1$._For if $X$_is_almost_isotropic and linearly isomorphic_with $\mathscr K_p$, then the functional_$$|x|=\|x\|+\sup_{\|x^*\|\leq 1}|x^*(x)|$$ is another $p$-norm that_is preserved by every isometry for the_original $p$-norm of $X$. It quickly_follows (cf. [@maximal Theorem 3.3] for_the complete_argument) that $|x|=2\|x\|$ for all_$x\in X$ and_so $\|x\|=\sup_{\|x^*\|\leq_1}|x^*(x)|$, that is,_$X$ is locally convex, which is_not the case.
Wheeling_around $\e$
--------------------
It is clear that moving_the_number $\e$ from_here_to_there in_the definitions opening_Section 3_one obtains_other_variants that are more or less_equivalent_to these appearing in the text.
Actually the_version of property $[\Game]$_and_the definition of a_space of almost universal complemented_disposition that we have used here_do not_agree with_those of [@GW] and [@CM-ACUD]. The following simple remark shows that_$[\Game]$ is equivalent to Garbulińska’s property_(E) of [@GW] and_that Definition \[def:aucd\]_is_equivalent to Definition_3.1_in [@CM-ACUD].
\[defec\]_Let $f^\dagger:E{\longrightarrow}F$ and $f^\ddagger:F{\longrightarrow}E$ be operators such_that $\|f^\ddagger\,_f^\dagger-{\bf 1}_E\|\leq \varepsilon$, where $\varepsilon<1$. Then_there is an automorphism_$a$_of $ |
mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $\overline{\mathcal{U}}_{0}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map, unitary loops with symmetry $({\rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ and $({\rm AII},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ in dimension $(d,d_{\parallel},D,D_{\parallel})$ are mapped to static Hamiltonians with symmetry $({\rm CI},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ and $({\rm DIII},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ in dimension $(d,d_{\parallel},D+1,D_{\parallel})$, respectively.
#### $\overline{\mathcal{O}}=\overline{\mathcal{U}}_{T/2}$
Due to the order-two antisymmetry realized by $\overline{\mathcal{U}}_{T/2}$, we haves $$\overline{\mathcal{U}}_{T/2}'\mathcal{H}(\boldsymbol{k},\boldsymbol{r},t)\overline{\mathcal{U}}_{T/2}'^{-1}=\mathcal{H}(-\boldsymbol{k}_{\parallel},\boldsymbol{k}_{\perp},-\boldsymbol{r}_{\parallel},\boldsymbol{r}_{\perp},T-t),$$ with $\overline{\mathcal{U}}_{T/2}'=\rho_{0}\otimes\overline{\mathcal{U}}_{T/2}$, which satisfies $$\begin{gathered}
\hat{\mathcal{U}}_{0}'\hat{\mathcal{T}'}=\overline{\eta}_{T}\hat{\mathcal{T}}'\hat{\mathcal{U}}_{0}'\\
\hat{\mathcal{U}}_{0}'\hat{\mathcal{C}}'=\overline{\eta}_{T}\hat{\mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $\overline{\mathcal{U}}_{T/2}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map, unitary loops with symmetry $({\rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{T/2,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ | mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $ \overline{\mathcal{U}}_{0}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map, unitary loops with symmetry $ ({ \rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ and $ ({ \rm AII},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ in property $ (d, d_{\parallel},D, D_{\parallel})$ are map to static Hamiltonians with symmetry $ ({ \rm CI},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ and $ ({ \rm DIII},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ in dimension $ (d, d_{\parallel},D+1,D_{\parallel})$, respectively.
# # # # $ \overline{\mathcal{O}}=\overline{\mathcal{U}}_{T/2}$
Due to the order - two antisymmetry understand by $ \overline{\mathcal{U}}_{T/2}$, we haves $ $ \overline{\mathcal{U}}_{T/2}'\mathcal{H}(\boldsymbol{k},\boldsymbol{r},t)\overline{\mathcal{U}}_{T/2}'^{-1}=\mathcal{H}(-\boldsymbol{k}_{\parallel},\boldsymbol{k}_{\perp},-\boldsymbol{r}_{\parallel},\boldsymbol{r}_{\perp},T - t),$$ with $ \overline{\mathcal{U}}_{T/2}'=\rho_{0}\otimes\overline{\mathcal{U}}_{T/2}$, which satisfies $ $ \begin{gathered }
\hat{\mathcal{U}}_{0}'\hat{\mathcal{T}'}=\overline{\eta}_{T}\hat{\mathcal{T}}'\hat{\mathcal{U}}_{0}'\\
\hat{\mathcal{U}}_{0}'\hat{\mathcal{C}}'=\overline{\eta}_{T}\hat{\mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $ \overline{\mathcal{U}}_{T/2}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian function, unitary loops with symmetry $ ({ \rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{T/2,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ | matjcal{C}}'\hat{\mathcal{U}}_{0}',\end{gatheved}$$ and $\overline{\mathcal{N}}_{0}'^{2}=\overlihe{\epsilov}_{U}$. Under the hermitian map, uiitaey loips with symmetry $({\rm AK},\text{\ensugemath{\oveelint{\mathcal{U}}_{0,\overline{\xfa}_{T}}^{\overline{\epallon}_{U}}}})$ end $({\rm AII},\text{\eksuremath{\ovarline{\mathcal{U}}_{0,\mvdrpine{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ in dimensijn $(d,d_{\pataplel},D,D_{\parallel})$ are iappsd to static Hamiltonians with symjetry $({\rk CI},\hat{\mathcak{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\lverpine{\epsilon}_{U}})$ and $({\rl DIII},\hat{\majgcaj{Y}}_{\overline{\eta}_{G},-\overline{\eua}_{C}}^{\overline{\epailon}_{U}})$ in dimension $(d,d_{\parallel},D+1,A_{\paraklel})$, respextuvepi.
#### $\overline{\mavhcal{O}}=\jverline{\mathgsl{U}}_{T/2}$
Dua to thr order-two antlsymmxtry realized by $\overline{\kathcal{U}}_{T/2}$, we haves $$\overline{\kachcal{U}}_{T/2}'\mathcal{H}(\boldsynbil{k},\bondsykbol{f},r)\ovdrljnx{\mafhcal{U}}_{H/2}'^{-1}=\mavhcal{H}(-\boldsgmbol{k}_{\paraloel},\boldsymbol{k}_{\perp},-\bplqwymbol{r}_{\parallsl},\boldfyibol{r}_{\perp},T-t),$$ with $\overline{\mathcal{U}}_{T/2}'=\rho_{0}\otpmes\kverline{\mathcal{U}}_{T/2}$, which satisfies $$\begin{gatheted}
\hat{\mathsal{U}}_{0}'\hat{\mathcal{T}'}=\overline{\eta}_{T}\hat{\mathcal{T}}'\hat{\mathcal{U}}_{0}'\\
\vat{\mavhzal{B}}_{0}'\mwg{\nahhcal{C}}'=\overline{\eta}_{T}\hat{\mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gaegetec}$$ and $\overline{\iathcal{U}}_{T/2}'^{2}=\ovrrpimg{\epsilon}_{U}$. Undet the hzdmjtian map, unitary poops wyth stmmetry $({\ri AI},\yext{\ensuremath{\overline{\mathcql{U}}_{T/2,\overline{\vta}_{T}}^{\iverline{\epsilon}_{U}}}})$ | mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $\overline{\mathcal{U}}_{0}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map, with $({\rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ $({\rm AII},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ in static with symmetry $({\rm and $({\rm DIII},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ dimension $(d,d_{\parallel},D+1,D_{\parallel})$, respectively. #### $\overline{\mathcal{O}}=\overline{\mathcal{U}}_{T/2}$ Due the order-two antisymmetry realized by $\overline{\mathcal{U}}_{T/2}$, we haves $$\overline{\mathcal{U}}_{T/2}'\mathcal{H}(\boldsymbol{k},\boldsymbol{r},t)\overline{\mathcal{U}}_{T/2}'^{-1}=\mathcal{H}(-\boldsymbol{k}_{\parallel},\boldsymbol{k}_{\perp},-\boldsymbol{r}_{\parallel},\boldsymbol{r}_{\perp},T-t),$$ with $\overline{\mathcal{U}}_{T/2}'=\rho_{0}\otimes\overline{\mathcal{U}}_{T/2}$, which satisfies \hat{\mathcal{U}}_{0}'\hat{\mathcal{T}'}=\overline{\eta}_{T}\hat{\mathcal{T}}'\hat{\mathcal{U}}_{0}'\\ \hat{\mathcal{U}}_{0}'\hat{\mathcal{C}}'=\overline{\eta}_{T}\hat{\mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and $\overline{\mathcal{U}}_{T/2}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map, unitary loops with symmetry $({\rm | mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gatHered}$$ and $\ovErlinE{\maThcAl{u}}_{0}'^{2}=\oveRlinE{\epsilon}_{U}$. Under THe heRmitian map, unitary loops With sYmMEtry $({\RM Ai},\text{\EnsuremATh{\OVErlInE{\mAthCaL{u}}_{0,\oVerliNe{\eTa}_{T}}^{\overLine{\epsiloN}_{U}}}})$ aNd $({\Rm AII},\text{\ensUReMath{\overliNe{\mAthcal{U}}_{0,\overlIne{\Eta}_{T}}^{\ovErLinE{\EpsilOn}_{U}}}})$ In dimEnsion $(D,D_{\paralLel},D,D_{\paraLlEL})$ are maPPed to stATIc hamiLtonians with symmeTRy $({\RM CI},\hat{\mathcal{U}}_{\OverliNe{\ETa}_{t},-\OVerLinE{\eta}_{T}}^{\overlInE{\epsiLOn}_{U}})$ and $({\rM dIii},\HAt{\mAThcal{U}}_{\overlinE{\eta}_{T},-\overliNE{\etA}_{T}}^{\overLiNe{\ePSilon}_{U}})$ In dimEnSIon $(D,d_{\parallel},D+1,d_{\parAllel})$, respEctiveLY.
#### $\overliNE{\mathcaL{O}}=\overLinE{\maThcaL{u}}_{T/2}$
duE to ThE OrdER-tWo aNTisYmmetry rEaLiZed by $\OverLINE{\MathCal{u}}_{T/2}$, we Haves $$\Overline{\mathcAl{U}}_{t/2}'\matHCal{h}(\boldSymboL{k},\boLdSymboL{r},t)\oveRline{\MaThcal{U}}_{T/2}'^{-1}=\mathcal{H}(-\BoldSymbol{k}_{\paRalLeL},\boLdSymboL{K}_{\perp},-\bOldSymBol{r}_{\parAllel},\boLDsyMbOL{R}_{\PeRp},T-t),$$ with $\overline{\maThCAL{U}}_{t/2}'=\rho_{0}\otimEs\overLInE{\mAThcal{U}}_{T/2}$, wHiCh sAtisFIEs $$\begIn{gaTHeRed}
\hat{\maThcal{U}}_{0}'\HAt{\MaThcal{T}'}=\oVeRline{\eTa}_{t}\haT{\maThcal{t}}'\Hat{\mAthcal{u}}_{0}'\\
\hat{\mathCal{U}}_{0}'\hAT{\mathcal{C}}'=\overlINe{\eta}_{T}\hat{\mathCAl{c}}'\HAt{\MAthcAl{U}}_{0}',\End{gathered}$$ And $\oVErliNe{\maTHcAl{U}}_{t/2}'^{2}=\OverlIne{\epSiLOn}_{u}$. under the hermitian maP, uNitary Loops With symmetry $({\rM AI},\text{\ensUREMath{\overLine{\MAtHCal{U}}_{T/2,\overline{\eTa}_{T}}^{\ovErline{\epsiLOn}_{U}}}})$ | mathcal{C}}'\hat{\mathcal{ U}}_{0}',\ end{g ath ere d} $$ a nd $ \overline{\mat h cal{ U}}_{0}'^{2}=\overline {\eps il o n}_{ U }$ . Und er theh er m i tia nma p,un i ta ry lo ops with s ymmetry $( {\r mAI},\text{\e n su remath{\ov erl ine{\mathcal {U} }_{0,\ ov erl i ne{\e ta} _{T}} ^{\ove r line{\ epsilon}_ {U } }}})$a nd $({\ r m A II}, \text{\ensuremath { \o v erline{\mathca l{U}}_ {0 , \o v e rli ne{ \eta}_{T}} ^{ \over l ine{\ep s il o n } _{U } }}})$ in dime nsion $(d,d _ {\p aralle l} ,D, D _{\par allel }) $ ar e mapped to sta tic Hamil tonian s with s y mmetry$({\rm CI },\ hat{ \ ma th cal {U } }_{ \ ov erl i ne{ \eta}_{T }, -\ overl ine{ \ e t a }_{T }}^ {\ov erlin e{\epsilon}_{ U}} )$ a n d $ ({\rm DIII },\h at {\mat hcal{U }}_{\ ov erline{\eta}_{T },-\ overline{ \et a} _{T }} ^{\ov e rline{ \ep sil on}_{U} })$ ind ime ns i o n $ (d,d_{\parallel},D +1 , D _{ \paralle l})$,r es pe c tively.
# ### $\o v e rline {\ma t hc al{O}}=\ overli n e{ \m athcal{ U} }_{T/2 }$
D ueto th e ord er-two antisym metry realized by $\ o verline{\math c al { U }} _ {T/2 }$, we haves $ $\ov e rlin e{\m a th cal { U}}_{ T/2}' \m a th c al{H}(\boldsymbol{k }, \bolds ymbol {r},t)\overli ne{\mathca l { U }}_{T/2} '^{- 1 }= \ mathcal{H}(-\b oldsy mbol{k}_{\ p arallel} ,\bol dsymbol{ k}_{\perp } , -\boldsy mbo l{r }_{ \pa r a ll el},\boldsymb o l {r}_ {\ perp},T -t) ,$$ wit h $ \ov erl ine {\ mathcal{U }}_{T/2} '= \r ho _{ 0}\ otime s \overlin e{ \ma th cal {U}}_ { T/2}$, whic h sa ti sf i es$$\begi n {g a t here d}
\ hat{ \ma th cal{U }}_{ 0 }'\ hat{\ma thcal{T}' }=\ o verl in e{ \eta}_{ T}\hat{\mathc al {T}}'\hat{ \m ath cal{U} } _ {0}'\\
\ hat{\mathcal{U}}_{0}'\h a t{\math cal {C}}' =\ov erline{\e ta} _{T}\h at{ \ mathca l{C}}' \hat{ \m ath c a l{U}} _ { 0} ',\ en d{gathered } $ $ a nd $\ ov erli ne{\mat hcal{U}}_{T/2}'^{2 } =\o verline{\epsi lon }_{U } $ .Und e rt hehe r mit i a n map, unitaryloops with s y mm etry $({\r m AI }, \text{\ ensurem ath{\ o verline {\mathcal {U}}_{T/2 ,\ over l i ne{ \eta}_{T}} ^{\overl ine{\epsi l on}_{ U }} }})$ | mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and_$\overline{\mathcal{U}}_{0}'^{2}=\overline{\epsilon}_{U}$. Under_the hermitian map, unitary_loops with_symmetry_$({\rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$_and_$({\rm AII},\text{\ensuremath{\overline{\mathcal{U}}_{0,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ in_dimension $(d,d_{\parallel},D,D_{\parallel})$ are_mapped to static Hamiltonians_with symmetry $({\rm_CI},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$_and $({\rm DIII},\hat{\mathcal{U}}_{\overline{\eta}_{T},-\overline{\eta}_{T}}^{\overline{\epsilon}_{U}})$ in dimension $(d,d_{\parallel},D+1,D_{\parallel})$, respectively.
#### $\overline{\mathcal{O}}=\overline{\mathcal{U}}_{T/2}$
Due to the order-two antisymmetry realized by_$\overline{\mathcal{U}}_{T/2}$,_we haves_$$\overline{\mathcal{U}}_{T/2}'\mathcal{H}(\boldsymbol{k},\boldsymbol{r},t)\overline{\mathcal{U}}_{T/2}'^{-1}=\mathcal{H}(-\boldsymbol{k}_{\parallel},\boldsymbol{k}_{\perp},-\boldsymbol{r}_{\parallel},\boldsymbol{r}_{\perp},T-t),$$_with_$\overline{\mathcal{U}}_{T/2}'=\rho_{0}\otimes\overline{\mathcal{U}}_{T/2}$, which satisfies $$\begin{gathered}
\hat{\mathcal{U}}_{0}'\hat{\mathcal{T}'}=\overline{\eta}_{T}\hat{\mathcal{T}}'\hat{\mathcal{U}}_{0}'\\
\hat{\mathcal{U}}_{0}'\hat{\mathcal{C}}'=\overline{\eta}_{T}\hat{\mathcal{C}}'\hat{\mathcal{U}}_{0}',\end{gathered}$$ and_$\overline{\mathcal{U}}_{T/2}'^{2}=\overline{\epsilon}_{U}$. Under the hermitian map,_unitary loops_with symmetry $({\rm AI},\text{\ensuremath{\overline{\mathcal{U}}_{T/2,\overline{\eta}_{T}}^{\overline{\epsilon}_{U}}}})$ |
main focus topics: dataset improvement, methodology improvement, simulation/performance analysis. The limited dataset and class space used for this study is known. Future efforts will include a more complete class space, as well as more data to support under-represented class types. Specifically datasets such as the Catalina Real Time Transient Survey [@Drake2009], will provide greater depth and completeness as a prelude to the data sets that will be available from the Panoramic Survey Telescope & Rapid Response System and the Large Synoptic Survey Telescope (LSST).
In addition to improving the underlying training data used, the methodology outline will also be researched to determine if more optimal methods are available. Exploring the effects of variable size state space for the translation could potentially yield performance improvements, as could a comparison of slotting methods (e.g. box slots vs. Gaussian slots vs. other kernels or weighting schemes). Likewise, implementations beyond supervised classification (e.g., unsupervised classification) were not explored as part of this analysis. How the feature space outlined in this analysis would lend itself to clustering or expectation-maximization algorithms is yet to be determined.
In a future paper, how sampling rates and photometric errors affect the ability to represent the underlying time-domain functionality using synthetic time-domain signals will be explored. Simulation of the expected time domain signals will allow for an estimation of performance of other spectral methods (DWT/DFT for irregular sampling), which will intern allow for and understanding of the benefits and drawbacks of each methodology, relative to both class type and observational conditions. This type of analysis would require the modeling and development of synthetic stellar variable functions to produce reasonable (and varied) time domain signature.
Acknowledgments
---------------
The authors are grateful for valuable discussion with Stephen Wiechecki-Vergara and Hakeem Oluseyi. Research was partially supported by Vencore, Inc. The LINEAR program is sponsored by the National Aeronautics and Space Administration (NRA Nos. NNH09ZDA001N, 09-NEOO09-0010) and the United States Air Force under Air Force Contract FA8721-05-C-0002
References
==========
---
abstract: 'We show that the biomolecular exciton dynamics under the influence of slow polarization fluctuations in the solvent cannot be described by approaches which are perturbative in the system-bath coupling. For this, we compare results for the decoherence rate of the exciton dynamics of a resumed perturbation theory with | main focus topics: dataset improvement, methodology improvement, simulation / operation psychoanalysis. The circumscribed dataset and class space used for this report is known. Future efforts will admit a more complete class space, equally well as more datum to support under - represented course types. Specifically datasets such as the Catalina Real Time Transient Survey [ @Drake2009 ], will provide big astuteness and completeness as a prelude to the data set that will be available from the Panoramic Survey Telescope & Rapid Response System and the Large Synoptic Survey Telescope (LSST).
In addition to improving the underlie training data used, the methodology outline will besides be researched to determine if more optimum methods are available. explore the effects of varying size state space for the translation could potentially yield operation improvements, as could a comparison of slotting methods (e.g. box slots vs. Gaussian slots vs. other kernels or weighting schemes). Likewise, implementations beyond supervised classification (e.g., unsupervised classification) were not explored as part of this psychoanalysis. How the feature of speech space outlined in this analysis would lend itself to bunch or arithmetic mean - maximization algorithms is yet to be settle.
In a future paper, how sampling rates and photometric errors involve the ability to represent the underlying time - domain functionality using synthetic fourth dimension - domain signals will be explored. Simulation of the ask clock time domain signals will allow for an estimation of performance of early spectral methods (DWT / DFT for irregular sampling), which will intern allow for and understanding of the benefits and drawbacks of each methodology, relative to both course type and observational conditions. This type of analysis would necessitate the modeling and growth of synthetic leading variable functions to grow reasonable (and varied) time knowledge domain key signature.
Acknowledgments
---------------
The authors are grateful for valuable discussion with Stephen Wiechecki - Vergara and Hakeem Oluseyi. Research was partially supported by Vencore, Inc. The LINEAR program is sponsored by the National Aeronautics and Space Administration (NRA Nos. NNH09ZDA001N, 09 - NEOO09 - 0010) and the United States Air Force under Air Force Contract FA8721 - 05 - C-0002
References
= = = = = = = = = =
---
abstract:' We show that the biomolecular exciton dynamics under the influence of slow polarization fluctuation in the solvent cannot be described by approach which are perturbative in the system - bath coupling. For this, we compare results for the decoherence rate of the exciton dynamics of a sum up perturbation theory with | maln focus topics: dataset lmprovement, methodology improbement, skmulation/performance analysid. Rhe lumited dataset and clars space lsed for rhis wtudy is kikwn. Futmxe efrlrts xill include a kore complate class spaca, xs well as more data to support under-wepresemtfd class types. Spebisicamly datasets such as the Catalina Deal Tike Transient Xurvey [@Drake2009], will provide hreaher depth and comppeteness as a pwwlude to the data sets that will bg available from the Panoramic Sjrvey Telescope & Eaplg Response Wystei and the Lavbe Synmptic Sirvey Telescopc (LSSV).
In qddition to improving the underlying trwining dada used, the methodoooty oujline wilu alro ge rssearcjed to determjne if more optimal methods art adqilable. Exploding tre effects of variable size state space fmr fhe translation could pitentially yield perflrmance iiprovements, as could a comparison of slotting metvods (x.g. bor slotr vd. Gaussian slots vs. other kernels or weightind svhvmes). Likewise, impjementationx heujnd supervisea classifjcation (e.g., unsupergised cjassidication) rere not explored as part of thus analysis. Kow the feature space outlined iu this analusis would lend itself co cluatering or fxpectatikv-maximization aleorpthmv is yet to be determined.
Ig a futurx papzr, how sxmplong raees and phltomebsic errors affect hhe ayilitf to repredent the underlying time-domain hnnctionality ositg vynthetie time-comain signalf will be explpred. Siiulatkon of the expectxd time domayn signals winp allow for en estimaeion of performxvce of other slectral mvtkods (DWT/DDT for irregular semplkhg), which will iuuerb allow for and unaerftwnving jx the benefids avd atawbazks of tcch oethpdology, relative to toth class type and obxevvational conditijns. This type of analysis would requmre thx modekind and development of synthetic stellar garlable functiogs ti produce recsonable (and varied) time domain signaturx.
Acknowledgments
---------------
The aujhors are grateful fot vsluable disrussiog with Staphen Wiechecki-Vergaea and Hakeem Oluxeyi. Research was parfially supplrted by Vencore, Inc. The LINEAR program is sponsored by the National Aerobautics and Space Zdmimistrdtnon (NRA Njs. NIH09VDA001N, 09-NEOO09-0010) and the United States Air Force under Eir Force Woutract FA8721-05-C-0002
References
==========
---
abstract: 'Wd show that tfe biomolecular exciton synamics under the influence of slow pokarization fluctuations ln the soovent cannon bx described by approsches chich qre perrurbwtive in the rysyel-bath couolinj. Fkr this, we compare results for tye decihfrence rate of tie exciton dnnxmics of a tesumed perturbatjon fjeory with | main focus topics: dataset improvement, methodology improvement, The dataset and space used for efforts include a more class space, as as more data to support under-represented types. Specifically datasets such as the Catalina Real Time Transient Survey [@Drake2009], will greater depth and completeness as a prelude to the data sets that will available the Survey & Rapid Response System and the Large Synoptic Survey Telescope (LSST). In addition to improving the training data used, the methodology outline will also researched to determine if optimal methods are available. Exploring effects variable size space the could potentially yield improvements, as could a comparison of slotting methods (e.g. box slots vs. Gaussian slots vs. other kernels weighting schemes). beyond supervised (e.g., classification) not explored as this analysis. How the feature space analysis would lend itself to clustering or expectation-maximization is yet be determined. In a future paper, sampling rates and photometric errors affect the ability represent the underlying time-domain functionality using synthetic time-domain signals will be explored. Simulation of the domain signals will allow an estimation of of spectral (DWT/DFT irregular sampling), will intern allow for and understanding of the benefits and drawbacks each methodology, relative to both class type and observational conditions. of would require the and development of synthetic variable to produce reasonable (and domain Acknowledgments are for discussion with Stephen Wiechecki-Vergara Hakeem Oluseyi. Research was partially by Vencore, Inc. The the National Aeronautics and Space Administration (NRA Nos. 09-NEOO09-0010) and the United States Air Force Air Force Contract FA8721-05-C-0002 References ========== --- abstract: 'We show that the exciton dynamics influence of slow polarization fluctuations in the solvent be described by approaches are perturbative in the system-bath coupling. For this, we results the decoherence of the exciton of a resumed theory with | main focus topics: dataset impRovement, meThodoLogY imPrOvemEnt, sImulation/perfoRMancE analysis. The limited datAset aNd CLass SPaCe useD for thiS StUDY is KnOwN. FuTuRE eFfortS wiLl incluDe a more comPleTe Class space, as WElL as more datA to Support under-RepResentEd ClaSS typeS. SpEcifiCally dATasets Such as the caTAlina REAl Time TRANsIent survey [@Drake2009], will prOViDE greater depth aNd compLeTEnESS as A prElude to the DaTa setS That wilL Be AVAIlaBLe from the PanoRamic Survey tEleScope & RApId RESponse systeM aND thE Large SynopTic SUrvey TeleScope (LssT).
In addITion to iMproviNg tHe uNderLYiNg TraInINg dATa UseD, The MethodolOgY oUtlinE wilL ALSO be rEseArchEd to dEtermine if morE opTimaL MetHods aRe avaIlabLe. exploRing thE effeCtS of variable size StatE space for The TrAnsLaTion cOUld potEntIalLy yield PerformANce ImPROVeMents, as could a compaRiSON oF slottinG methoDS (e.G. bOX slots vs. gaUssIan sLOTs vs. oTher KErNels or weIghtinG ScHeMes). LikeWiSe, implEmEntAtiOns beYOnd sUperviSed classIficaTIon (e.g., unsuperviSEd classificatIOn) WERe NOt exPloRed as part of This ANalySis. HOW tHe fEAture Space OuTLiNEd in this analysis wouLd Lend itSelf tO clustering or ExpectatioN-MAXimizatiOn alGOrIThms is yet to be dEtermIned.
In a futURe paper, hOw samPling ratEs and photOMEtric errOrs AffEct The ABIlIty to represenT THe unDeRlying tIme-Domain fUncTioNalIty UsIng syntheTic time-dOmAiN sIgNalS will BE exploreD. SImuLaTioN of thE ExpectEd timE domAiN sIGnaLs will aLLoW FOr an EsTiMatiOn oF pErforMancE Of oTher speCtral methOds (dwT/DFt fOr IrregulAr sampling), whiCh Will intern AlLow For and UNDerstandIng of the benefits and drawBAcks of eAch MethoDoloGy, relativE to Both clAss TYpe and ObservAtionAl ConDITions. tHIs TypE oF analysis wOULd rEquirE tHe moDeling aNd development of synTHetIc stellar variAblE funCTIoNs tO PrODucE rEAsoNABle (and varied) timE domain sigNaTUrE.
AcknowledGMenTs
---------------
the authOrs are gRatefUL for valUable discUssion witH STephEN wieChecki-VergAra and HaKeem OluseYI. ReseARcH was pArtIally sUpPorTed by vencorE, inc. the LInEAR prOgRam is sPonsoReD by the NaTional Aeronautics and SpaCe AdmiNistrAtiOn (NRA Nos. NnH09Zda001N, 09-NeOO09-0010) and the unitEd States AiR FoRce Under air fOrce COntrACt fA8721-05-C-0002
rEfereNces
==========
---
ABstract: 'We SHoW thAT ThE biomoleculAR EXciTon dyNamICs undeR the Influence of slow poLArization fluctUatiONS in The SOlveNt Cannot be descriBed By APProaches WhIch are pertuRbative iN tHE systEm-bath CoupliNg. For thIS, We COmpare ResuLts For the decOheReNCe rate oF tHe EXciton DynaMiCs of a rEsumed PErtuRBAtion theory with | main focus topics: datase t improvem ent,met hod ol ogyimpr ovement, simul a tion /performance analysis. Theli m ited da taset and cl a ss s pac eus edfo r t his s tud y is kn own. Futur e e ff orts will in c lu de a morecom plete classspa ce, as w ell as mo redatato sup p ort un der-repre se n ted cl a ss type s . S peci fically datasetss uc h as the Catali na Rea lT im e Tra nsi ent Survey [ @Drak e 2009],w il l p rov i de greater de pth and com p let enessas ap relude to t he dat a sets that wil l be avai lablef rom the Panoram ic Sur vey Te lesc o pe & Ra pi d Re s po nse Sys tem andth eLarge Syn o p t i c Su rve y Te lesco pe (LSST).
I n a ddit i onto im provi ng t he unde rlying trai ni ng data used, t he m ethodolog y o ut lin ewilla lso be re sea rched t o deter m ine i f m or e optimal methodsar e av ailable. Explo r in gt he effec ts of var i a ble s izes ta te space for t h etr anslati on could p ote nti allyy ield perfo rmance i mprov e ments, as coul d a comparison of s lo t ting me thods (e.g. box slot s vs . G aus s ian s lotsvs . o t her kernels or weig ht ing sc hemes ). Likewise,implementa t i o ns beyon d su p er v ised classific ation (e.g., un s upervise d cla ssificat ion) were n ot explo red as pa rto f t his analysis. H ow t he featur e s pace ou tli ned in th is analysis would l en dit se lfto cl u steringor ex pe cta tion- m aximiz ation alg or it h msis yett ob e det er mi ned.
I na fut urep ape r, howsamplingrat e s an dph otometr ic errors aff ec t the abil it y t o repr e s ent theunderlying time-domainf unction ali ty us ingsynthetic ti me-dom ain signal s will be e xp lor e d . Sim u l at ion o f the expe c t edtimedo main signal s will allow for a n es timation of p erf orma n c eofo th e r s pe c tra l methods (DWT/DF T for irre gu l ar sampling) , wh ic h willinternallow for and understa nding ofth e be n e fit s and draw backs of each met h odolo g y, rela tiv e to b ot h c lasstype a n d o bserv ationa lcondit ions. T his type of analysis would requ ire th e mod eli ng and de vel o pme nt of syn thet ic stellar va ria ble f unc t ionsto p r od uce reaso nabl e (and var i ed ) t i m edomain sign a t u re.
Ack now l edgmen ts
- --------------
T h e authors aregrat e f ulfor valu ab le discussionwit hS t ephen Wi ec hecki-Verga ra and H ak e em Ol useyi. Resea rch was p ar t iallysupp ort ed by Ven cor e, Inc. Th eLI N EAR pr ogra mis spo nsored by t h e National Aerona utics a nd Sp a ceAdmin is tration (NRA Nos. NNH0 9ZDA001N, 0 9-NEOO 09-0 010)and the U nitedSta te s Air Forc e under Ai r For ce Cont ra ct F A87 21-05- C-00 0 2
Ref eren ce s
= ========= -- -
a bs t rac t: ' We sh ow tha t the bio m olecular ex c iton dy na mic s undert he i nfluence o f s low p o l arizationf luct u at i ons i n thesolven t canno t be d escribe d b y approache s which a r e p er turb ative in t he s ys tem -b ath coupling. F or this, we comp a r e resu lts fort he dec oh erence rateof th ee xcitond y nam ics ofa re sum edpert urb atio n theory with | main_focus topics:_dataset improvement, methodology improvement,_simulation/performance analysis._The_limited dataset_and_class space used_for this study_is known. Future efforts_will include a_more_complete class space, as well as more data to support under-represented class types. Specifically_datasets_such as_the_Catalina_Real Time Transient Survey [@Drake2009],_will provide greater depth and_completeness as_a prelude to the data sets that will_be_available from the_Panoramic Survey Telescope & Rapid Response System and the_Large Synoptic Survey Telescope (LSST).
In addition_to improving the_underlying_training_data used, the methodology_outline will also be researched to_determine if more optimal methods are_available. Exploring the effects of variable size_state space for the translation could_potentially yield performance improvements, as_could a_comparison of slotting methods (e.g._box slots vs._Gaussian slots_vs. other kernels_or weighting schemes). Likewise, implementations beyond_supervised classification (e.g.,_unsupervised classification) were not explored as_part_of this analysis._How_the_feature space_outlined in this_analysis_would lend_itself_to clustering or expectation-maximization algorithms is_yet_to be determined.
In a future paper, how_sampling rates and photometric_errors_affect the ability to_represent the underlying time-domain functionality_using synthetic time-domain signals will be_explored. Simulation_of the_expected time domain signals will allow for an estimation of performance_of other spectral methods (DWT/DFT for_irregular sampling), which will_intern allow_for_and understanding of_the_benefits and_drawbacks of each methodology, relative to both_class type_and observational conditions. This type of_analysis would require the_modeling_and development of synthetic stellar variable_functions to produce reasonable (and varied)_time domain signature.
Acknowledgments
---------------
The authors are_grateful_for_valuable discussion with Stephen Wiechecki-Vergara_and Hakeem Oluseyi. Research was partially_supported by Vencore,_Inc. The LINEAR program is sponsored by_the_National Aeronautics and Space Administration (NRA_Nos._NNH09ZDA001N, 09-NEOO09-0010) and the United States_Air_Force_under Air Force Contract FA8721-05-C-0002
References
==========
_---
abstract: 'We show that the biomolecular_exciton dynamics under the influence of slow polarization fluctuations_in the solvent_cannot be described by approaches_which_are_perturbative in the system-bath coupling. For this, we compare results_for the_decoherence rate of_the exciton dynamics of a resumed perturbation theory with |
12). Throwing in equation (14) would bring us even closer to equation (12), which dominates because of its small error. Combining the estimates would be unwise, because they do not agree well.
This small value from the IRAS survey is not entirely new (cf. Fisher, Scharf, & Lahav 1994). It has received little attention because the larger value, $\Gamma_{\mathrm{o}} \approx 0.25$, was already seen as a major challenge to the popular SCDM model ($\Omega_{\mathrm{M}} = 1$). The smaller value poses an even more severe challenge to the SCDM model. But it gives more scope to low-density models, which are popular now. Until the discrepant values of $\Gamma_{\mathrm{o}}$ are understood, we think it wise to show joint CRs using separately the larger and the smaller values of $\Gamma_{\mathrm{o}}$.
Figures \[Fig:H-Omega\_M\_L0\_G0.15\] and \[Fig:H-eta\_L0\_G0.15\], analogs of Figures \[Fig:H-Omega\_M\_L0\] and \[Fig:H-eta\_L0\], show CRs for our four standard constraints, but with $\Gamma_{\mathrm{o}} =
0.255\pm 0.017$ replaced by $\Gamma_{\mathrm{o}} = 0.15 \pm 0.04$. The $\chi^2_{\mathrm{min}}$ is 0.63 for 1 DOF (good) and is located at ($h$, $\Omega_{\mathrm{M}}$, $\eta_{10}$) = (0.60, 0.30, 4.6).
The CRs now exclude the SCDM model strongly and favor low density. The value $\eta_{10} \approx 5$, favored by the Burles and Tytler (1997a,b,c) deuterium abundance determination (see §1), is now near the point of optimum fit. The CRs clearly would accept the added cluster constraint $\Omega_{\mathrm{M}} \approx 0.2$ if we were to apply it. But note that in our three-dimensional CRs, low $\Omega_{\mathrm{M}}$ goes with low $\eta_{10}$, because of the $f_{\mathrm{G}}$ constraint. For example, the combinations (0.7, 0.2, 3) and (0.7 | 12). Throwing in equation (14) would bring us even closer to equality (12), which dominate because of its small error. compound the estimate would be unwise, because they do not agree well.
This little value from the IRAS view is not entirely fresh (cf. Fisher, Scharf, & Lahav 1994). It has received short attention because the larger value, $ \Gamma_{\mathrm{o } } \approx 0.25 $, was already see as a major challenge to the popular SCDM model ($ \Omega_{\mathrm{M } } = 1 $). The smaller value poses an even more severe challenge to the SCDM exemplar. But it gives more scope to low - concentration models, which are popular nowadays. Until the discrepant value of $ \Gamma_{\mathrm{o}}$ are understood, we think it wise to show joint CRs use separately the larger and the smaller values of $ \Gamma_{\mathrm{o}}$.
Figures \[Fig: H - Omega\_M\_L0\_G0.15\ ] and \[Fig: planck's constant - eta\_L0\_G0.15\ ], analogs of Figures \[Fig: H - Omega\_M\_L0\ ] and \[Fig: H - eta\_L0\ ], show CRs for our four standard constraints, but with $ \Gamma_{\mathrm{o } } =
0.255\pm 0.017 $ replaced by $ \Gamma_{\mathrm{o } } = 0.15 \pm 0.04$. The $ \chi^2_{\mathrm{min}}$ is 0.63 for 1 DOF (good) and is located at ($ h$, $ \Omega_{\mathrm{M}}$, $ \eta_{10}$) = (0.60, 0.30, 4.6).
The CRs now bar the SCDM model strongly and privilege depleted density. The value $ \eta_{10 } \approx 5 $, favored by the Burles and Tytler (1997a, b, c) deuterium abundance determination (determine § 1), is now near the point of optimum fit. The CRs intelligibly would accept the added cluster constraint $ \Omega_{\mathrm{M } } \approx 0.2 $ if we were to apply it. But note that in our three - dimensional chromium, low $ \Omega_{\mathrm{M}}$ goes with low $ \eta_{10}$, because of the $ f_{\mathrm{G}}$ constraint. For example, the combinations (0.7, 0.2, 3) and (0.7 | 12). Thgowing in equation (14) woula bring us even closer to eqhation (12), dhich dominates because of ivs snall tgror. Combining the esgimates wluld be ynwiww, because vgey do kjt afvee wzlo.
This small vakue from tve IRAS survey ir uot entirely new (cf. Fisher, Scharf, & Lwhav 1994). Iy jas received lyttlt aetenfpok because the larger value, $\Gamma_{\jathrm{o}} \approx 0.25$, was slready seen as a major chwllejge to the popular SCDM model ($\Omedq_{\mathrm{M}} = 1$). Tfe smaller value poses an even more severe challenge go thz SCDM modeo. Vut ht gives moce scoie to low-denslny modens, whicn are popular kow. Uitil the discrepant valuev of $\Gamma_{\mathrm{o}}$ wre undervtkod, we think it wusw to vhow joivr CFs hsmng separwtemy the larfer and the smaller values of $\Baina_{\mathrm{o}}$.
Figudes \[Fig:R-Oiega\_M\_L0\_G0.15\] and \[Fig:H-eta\_L0\_G0.15\], analogs of Figurev \[Fif:H-Omega\_M\_L0\] and \[Fig:H-eta\_L0\], whow CRs for our four standard constraints, but with $\Gamma_{\mathrm{o}} =
0.255\pm 0.017$ replaced ty $\Gajoa_{\mcbmrm{o}} = 0.15 \om 0.04$. The $\chi^2_{\mathrm{min}}$ is 0.63 for 1 DOF (good) and is movaned at ($h$, $\Omega_{\matmrm{M}}$, $\eta_{10}$) = (0.60, 0.30, 4.6).
The CRx joe exclude the RCDM modem strongly and favlr low qensiry. The vajue $\rta_{10} \approx 5$, favored by the Vurles and Tjtlee (1997a,b,c) deuterium abbndance detexminatoon (sre §1), is now near the poiut of kptimum fit. The CRs duearly would accdpt tve added cluster constraine $\Omega_{\mavhrm{M}} \approx 0.2$ if we wewe to applj it. Ngt note that in oug thrge-dimetsional CRd, low $\Omega_{\mathrm{M}}$ goes with lox $\eta_{10}$, because ox tve $f_{\mathxm{G}}$ cokstraint. For exwmple, the combnnations (0.7, 0.2, 3) avd (0.7 | 12). Throwing in equation (14) would bring closer equation (12), dominates because of estimates be unwise, because do not agree This small value from the IRAS is not entirely new (cf. Fisher, Scharf, & Lahav 1994). It has received attention because the larger value, $\Gamma_{\mathrm{o}} \approx 0.25$, was already seen as a challenge the SCDM ($\Omega_{\mathrm{M}} = 1$). The smaller value poses an even more severe challenge to the SCDM model. it gives more scope to low-density models, which popular now. Until the values of $\Gamma_{\mathrm{o}}$ are understood, think wise to joint using the larger and smaller values of $\Gamma_{\mathrm{o}}$. Figures \[Fig:H-Omega\_M\_L0\_G0.15\] and \[Fig:H-eta\_L0\_G0.15\], analogs of Figures \[Fig:H-Omega\_M\_L0\] and \[Fig:H-eta\_L0\], show CRs for four standard with $\Gamma_{\mathrm{o}} 0.255\pm replaced $\Gamma_{\mathrm{o}} = 0.15 The $\chi^2_{\mathrm{min}}$ is 0.63 for 1 is located at ($h$, $\Omega_{\mathrm{M}}$, $\eta_{10}$) = (0.60, 4.6). The now exclude the SCDM model strongly favor low density. The value $\eta_{10} \approx 5$, by the Burles and Tytler (1997a,b,c) deuterium abundance determination (see §1), is now near the optimum fit. The CRs would accept the cluster $\Omega_{\mathrm{M}} 0.2$ we were apply it. But note that in our three-dimensional CRs, low $\Omega_{\mathrm{M}}$ with low $\eta_{10}$, because of the $f_{\mathrm{G}}$ constraint. For example, (0.7, 3) and (0.7 | 12). Throwing in equation (14) would brIng us even cLoser To eQuaTiOn (12), whIch dOminates becausE Of itS small error. Combining thE estiMaTEs woULd Be unwIse, becaUSe THEy dO nOt AgrEe WElL.
This SmaLl value From the IRAs suRvEy is not entirELy New (cf. FisheR, ScHarf, & Lahav 1994). It hAs rEceiveD lIttLE atteNtiOn becAuse thE Larger Value, $\GammA_{\mAThrm{o}} \aPProx 0.25$, was ALReAdy sEen as a major challeNGe TO the popular SCDm model ($\omEGa_{\MAThrM{M}} = 1$). THe smaller vAlUe posES an even MOrE SEVerE Challenge to thE SCDM model. BUT it Gives mOrE scOPe to loW-densItY ModEls, which are PopuLar now. UntIl the dIScrepanT Values oF $\Gamma_{\MatHrm{O}}$ are UNdErStoOd, WE thINk It wISe tO show joiNt cRS usinG sepARATEly tHe lArgeR and tHe smaller valuEs oF $\GamMA_{\maThrm{o}}$.
figurEs \[FiG:H-omega\_m\_L0\_G0.15\] and \[fig:H-eTa\_l0\_G0.15\], analogs of FiguRes \[FIg:H-Omega\_M\_l0\] anD \[FIg:H-EtA\_L0\], shoW cRs for Our FouR standaRd constRAinTs, BUT WiTh $\Gamma_{\mathrm{o}} =
0.255\pm 0.017$ rePlACEd By $\Gamma_{\mAthrm{o}} = 0.15 \PM 0.04$. THe $\CHi^2_{\mathrm{MiN}}$ is 0.63 For 1 Dof (Good) aNd is LOcAted at ($h$, $\OMega_{\maTHrM{M}}$, $\Eta_{10}$) = (0.60, 0.30, 4.6).
The Crs Now excLuDe tHe ScDM moDEl stRongly And favor Low deNSity. The value $\etA_{10} \Approx 5$, favored BY tHE buRLes aNd TYtler (1997a,b,c) deuTeriUM abuNdanCE dEteRMinatIon (seE §1), iS NoW Near the point of optimUm Fit. The cRs clEarly would accEpt the addeD CLUster conStraINt $\oMega_{\mathrm{M}} \appRox 0.2$ if We were to apPLy it. But nOte thAt in our tHree-dimenSIOnal CRs, lOw $\OMegA_{\maThrM{m}}$ GoEs with low $\eta_{10}$, bECAuse Of The $f_{\matHrm{g}}$ constrAinT. FoR exAmpLe, The combinAtions (0.7, 0.2, 3) anD (0.7 | 12). Throwing in equation(14) would brin g u s e ve n cl oser to equation ( 1 2),which dominates becaus e ofit s sma l lerror . Combi n in g the e st ima te s w ouldbeunwise, because t hey d o not agreew el l.
This s mal l value from th e IRAS s urv e y isnot enti rely n e w (cf. Fisher,Sc h arf, & Lahav 1 9 9 4) . It has received lit t le attention beca use th el ar g e r v alu e, $\Gamma _{ \math r m{o}} \ a pp r o x 0. 2 5$, was alrea dy seen asa ma jor ch al len g e to t he po pu l arSCDM model($\O mega_{\ma thrm{M } } = 1$) . The sm allerval uepose s a neve nm ore se ver e ch allengeto t he SC DM m o d e l . Bu t i t gi ves m ore scope tolow -den s ity mode ls, w hich a re po pularnow.Un til the discrep antvalues of $\ Ga mma _{ \math r m{o}}$ ar e u ndersto od, wet hin ki t wi se to show joint C Rs u si ng separ atelyt he l a rger and t hesmal l e r val ueso f$\Gamma_ {\math r m{ o} }$.
Fi gu res \[ Fi g:H -Om ega\_ M \_L0 \_G0.1 5\] and\[Fig : H-eta\_L0\_G0. 1 5\], analogso fF i gu r es \ [Fi g:H-Omega\_ M\_L 0 \] a nd \ [ Fi g:H - eta\_ L0\], s h ow CRs for our four st an dard c onstr aints, but wi th $\Gamma _ { \ mathrm{o }} =
0 . 255\pm 0.017$repla ced by $\G a mma_{\ma thrm{ o}} = 0. 15 \pm 0. 0 4 $. The $ \ch i^2 _{\ mat h r m{ min}}$ is 0.6 3 for1DOF (go od) and is lo cat edat($ h$, $\Ome ga_{\mat hr m{ M} }$ , $ \eta_ { 10}$) =(0 .60 ,0.3 0, 4. 6 ).
Th e CRs now e xc l ude the SC D Mm o delst ro ngly an dfavor low den sity. T he value$\e t a_{1 0} \ approx5$, favored b ythe Burles a ndTytler ( 1997a,b, c) deuterium abundanced etermin ati on (s ee § 1), is no w n ear th e p o int of optim um fi t. Th e CRs c l e ar lywo uld accept t headded c lust er cons traint $\Omega_{\m a thr m{M}} \approx 0. 2$ i f we we r et o a pp l y i t . But note thatin our thr ee - di mensionalC Rs, l ow $\Om ega_{\m athrm { M}}$ go es with l ow $\eta_ {1 0}$, b eca use of the $f_{\ma thrm{G}}$ const r ai nt. F orexampl e, th e com binati o ns(0.7, 0.2,3) and ( 0.7 | 12). Throwing_in equation_(14) would bring us_even closer_to_equation (12),_which_dominates because of_its small error._Combining the estimates would_be unwise, because_they_do not agree well.
This small value from the IRAS survey is not entirely new_(cf._Fisher, Scharf,_&_Lahav_1994). It has received little_attention because the larger value,_$\Gamma_{\mathrm{o}} \approx_0.25$, was already seen as a major challenge_to_the popular SCDM_model ($\Omega_{\mathrm{M}} = 1$). The smaller value poses an_even more severe challenge to the_SCDM model. But_it_gives_more scope to low-density_models, which are popular now. Until_the discrepant values of $\Gamma_{\mathrm{o}}$ are_understood, we think it wise to show_joint CRs using separately the larger_and the smaller values of_$\Gamma_{\mathrm{o}}$.
Figures \[Fig:H-Omega\_M\_L0\_G0.15\] and_\[Fig:H-eta\_L0\_G0.15\], analogs of Figures \[Fig:H-Omega\_M\_L0\] and_\[Fig:H-eta\_L0\], show CRs_for our_four standard constraints,_but with $\Gamma_{\mathrm{o}} =
0.255\pm 0.017$_replaced by $\Gamma_{\mathrm{o}}_= 0.15 \pm 0.04$. The $\chi^2_{\mathrm{min}}$_is_0.63 for 1_DOF_(good)_and is_located at ($h$,_$\Omega_{\mathrm{M}}$,_$\eta_{10}$) =_(0.60,_0.30, 4.6).
The CRs now exclude the_SCDM_model strongly and favor low density. The_value $\eta_{10} \approx 5$,_favored_by the Burles and_Tytler (1997a,b,c) deuterium abundance determination_(see §1), is now near the_point of_optimum fit._The CRs clearly would accept the added cluster constraint $\Omega_{\mathrm{M}} \approx_0.2$ if we were to apply_it. But note that_in our_three-dimensional_CRs, low $\Omega_{\mathrm{M}}$_goes_with low_$\eta_{10}$, because of the $f_{\mathrm{G}}$ constraint. For_example, the_combinations (0.7, 0.2, 3) and (0.7 |
$\mathcal A$. Namely, by the previous argument, from every vertex of the first type departs exactly one edge to the only vertex of the second type on the positive real axis, whose color is $(+,+)$ and whose operator-valued part corresponds to derivation and contraction with respect to $\mathfrak p$. The Cartan relation $[\mathfrak k,\mathfrak p]\subseteq \mathfrak p$ implies that at each vertex of the first type in a wheel-like graph $\Gamma$, the edge arriving at such a vertex must have color either $(+,+)$ or $(+,-)$, while the edge departing from it on th wheel must have opposite color. In other words, the edges of the cycle in a wheel-like graph $\Gamma$ must have alternating colours $(+,+)$ and $(+,-)$: this, in turn, excludes immediately $n$-wheels with $n$ odd.
Summarizing all previous arguments, the restriction of the operator $\mathcal A$ to $K$, which we denote (improperly) by the same symbol, defines an invertible, translation-invariant differential operator on $K$. Its symbol, regarded as an element of the completed symmetric algebra $\widehat{\mathrm S}(\mathfrak p)$ and defined through $j_\mathcal A(x)=e^{-x}\mathcal A(e^x)$, has the explicit form $$j_\mathcal A(x)=\exp\left(\sum_{n\geq 1}W_{2n}^\mathcal A\mathrm{tr}_\mathfrak p(\mathrm{ad}^{2n}(x))\right),\ x\in\mathfrak p,$$ where $W_{2n}^\mathcal A$, $n\geq 1$, denotes the integral weight of following wheel-like graph:
\
We observe that $j_\mathcal A$ is analytic in a neighborhood of $0$ in $\mathfrak p$. We observe that the Cartan relations for the symmetric pair $(\mathfrak g,\sigma)$ imply immediately that, for a general element $x$ of $\mathfrak p$, $\mathrm{ad}(x)^2$ is a well-defined endomorphism of $\mathfrak p$, thus all even powers of the adjoint representation restricted to $\mathfrak p$: hence the above expression is well-defined. Finally, all previous computations imply also the direct sum decomposition of $(A,\star_A)$: $$A=K\oplus (A\star_A\mathfrak k^{-\delta+\frac{1}4\mathrm{tr}_\mathfrak g\circ\ | $ \mathcal A$. Namely, by the previous argument, from every vertex of the first type departs precisely one boundary to the only vertex of the second type on the incontrovertible substantial axis, whose color is $ (+, +) $ and whose hustler - valued part corresponds to derivation and compression with obedience to $ \mathfrak p$. The Cartan relation $ [ \mathfrak k,\mathfrak p]\subseteq \mathfrak p$ incriminate that at each vertex of the first type in a roulette wheel - like graph $ \Gamma$, the boundary arriving at such a vertex must have semblance either $ (+, +) $ or $ (+, -)$, while the edge departing from it on th steering wheel must have opposite color. In early words, the edges of the hertz in a wheel - like graph $ \Gamma$ must experience alternating colours $ (+, +) $ and $ (+, -)$: this, in turn, excludes immediately $ n$-wheels with $ n$ odd.
sum up all previous arguments, the restriction of the operator $ \mathcal A$ to $ K$, which we denote (improperly) by the same symbol, defines an invertible, translation - invariant differential operator on $ K$. Its symbol, regarded as an element of the completed symmetric algebra $ \widehat{\mathrm S}(\mathfrak p)$ and defined through $ j_\mathcal A(x)=e^{-x}\mathcal A(e^x)$, has the explicit form $ $ j_\mathcal A(x)=\exp\left(\sum_{n\geq 1}W_{2n}^\mathcal A\mathrm{tr}_\mathfrak p(\mathrm{ad}^{2n}(x))\right),\ x\in\mathfrak p,$$ where $ W_{2n}^\mathcal A$, $ n\geq 1 $, denote the integral weight of stick to rack - like graph:
\
We observe that $ j_\mathcal A$ is analytic in a neighborhood of $ 0 $ in $ \mathfrak p$. We note that the Cartan relation for the symmetric pair $ (\mathfrak g,\sigma)$ entail immediately that, for a general element $ x$ of $ \mathfrak p$, $ \mathrm{ad}(x)^2 $ is a well - defined endomorphism of $ \mathfrak p$, thus all even powers of the adjoint representation restricted to $ \mathfrak p$: hence the above expression is well - defined. Finally, all former computations incriminate besides the lineal sum decomposition of $ (A,\star_A)$: $ $ A = K\oplus (A\star_A\mathfrak k^{-\delta+\frac{1}4\mathrm{tr}_\mathfrak g\circ\ | $\mahhcal A$. Namely, by the prtvious argument, fton everb vertes of the first type departs exactly lnw edgt to the only verteb of the decond ttpe ib the posivjve real axis, ahosz rolor is $(+,+)$ and wmose operatmr-valued part wofrzsponds to derivation and contractiog with tedpect to $\mathftak p$. Ehe Dartan relation $[\mathfrak k,\mathfrak p]\subseueq \mathfrak p$ impkies that at each vertex ov thf first type in a aheel-like gtzph $\Tamma$, the edee arriving at such a bertex must have color either $(+,+)$ ur $(+,-)$, wkile the edte deodrting from it og th wheel mmxt hava opposote color. In obher xordw, the edges of the cyrle in a wheel-like gtaph $\Gamma$ mbst have alternating xooours $(+,+)$ ang $(+,-)$: tfus, kn uuri, escluded ijmediately $n$-wheels wirh $n$ odd.
Summarizing ajo previous arfumentf, ehe restriction of the operator $\mathcal A$ fo $K$, which we denote (improperly) by the same dymbol, desines an invertible, translation-invariant differendial koercbjf oj $K$. Its symbol, regarded as an element of the skmklened symmetric algcbra $\widehat{\mathrm S}(\layrfrak p)$ and dgfined cgrkugh $j_\mathcal A(x)=e^{-x}\lathcal A(e^x)$, yas the evplivit form $$j_\mathcal A(x)=\exp\left(\wum_{n\geq 1}W_{2n}^\manhcao A\mathrm{tr}_\mathfran p(\mathrm{ad}^{2u}(x))\righj),\ x\in\msthfrak p,$$ where $W_{2n}^\mathccl A$, $n\feq 1$, denoted the intseral weight of fullpwhng wheel-like graph:
\
We obsewve that $o_\matheal A$ is anakytic yn a neighhorhood of $0$ in $\mathfrak o$. We lbverve that the Cartan relations for the sbkmetric pair $(\mdthxrak g,\siyma)$ imily immediately that, for a geueral elzment $b$ of $\mathfgak p$, $\matirm{ad}(x)^2$ is a rell-defined etfomorphism oh $\mathfrah p$, rhus all evdv powers of thr adjoint representqtion restricted tp $\mxfhfrak p$: hence cke above expressipn ks repl-vefinqg. Finally, aln prdviuis cooputations lmpuy akso the direct sum dacomlosition of $(A,\star_A)$: $$A=H\oplus (A\wtar_A\matrfrak k^{-\delta+\ftac{1}4\mathrm{tr}_\mathfran g\cicc\ | $\mathcal A$. Namely, by the previous argument, vertex the first departs exactly one of second type on positive real axis, color is $(+,+)$ and whose operator-valued corresponds to derivation and contraction with respect to $\mathfrak p$. The Cartan relation k,\mathfrak p]\subseteq \mathfrak p$ implies that at each vertex of the first type a graph the arriving at such a vertex must have color either $(+,+)$ or $(+,-)$, while the edge departing it on th wheel must have opposite color. other words, the edges the cycle in a wheel-like $\Gamma$ have alternating $(+,+)$ $(+,-)$: in turn, excludes $n$-wheels with $n$ odd. Summarizing all previous arguments, the restriction of the operator $\mathcal A$ to $K$, we denote the same defines invertible, differential operator on symbol, regarded as an element of algebra $\widehat{\mathrm S}(\mathfrak p)$ and defined through $j_\mathcal A(e^x)$, has explicit form $$j_\mathcal A(x)=\exp\left(\sum_{n\geq 1}W_{2n}^\mathcal A\mathrm{tr}_\mathfrak x\in\mathfrak p,$$ where $W_{2n}^\mathcal A$, $n\geq 1$, denotes integral weight of following wheel-like graph: \ We observe that $j_\mathcal A$ is analytic in of $0$ in $\mathfrak We observe that Cartan for symmetric $(\mathfrak g,\sigma)$ immediately that, for a general element $x$ of $\mathfrak p$, $\mathrm{ad}(x)^2$ a well-defined endomorphism of $\mathfrak p$, thus all even powers adjoint restricted to $\mathfrak hence the above expression well-defined. all previous computations imply direct decomposition (A\star_A\mathfrak g\circ\ | $\mathcal A$. Namely, by the previoUs argument, From eVerY veRtEx of The fIrst type departS ExacTly one edge to the only verTex of ThE SecoND tYpe on The posiTIvE REal AxIs, WhoSe COlOr is $(+,+)$ aNd wHose opeRator-valueD paRt Corresponds tO DeRivation anD coNtraction witH reSpect tO $\mAthFRak p$. THe CArtan RelatiON $[\mathfRak k,\mathfRaK P]\subseTEq \mathfRAK p$ ImplIes that at each vertEX oF The first type in A wheel-LiKE gRAPh $\GAmmA$, the edge arRiVing aT Such a veRTeX MUSt hAVe color either $(+,+)$ Or $(+,-)$, while the eDGe dEpartiNg FroM It on th Wheel MuST haVe opposite cOlor. in other woRds, the EDges of tHE cycle iN a wheeL-liKe gRaph $\gAmMa$ MusT hAVe aLTeRnaTIng Colours $(+,+)$ aNd $(+,-)$: ThIs, in tUrn, eXCLUDes iMmeDiatEly $n$-wHeels with $n$ odd.
sumMariZIng All prEviouS argUmEnts, tHe restRictiOn Of the operator $\maThcaL A$ to $K$, whicH we DeNotE (iMpropERly) by tHe sAme Symbol, dEfines aN InvErTIBLe, Translation-invariaNt DIFfErential OperatOR oN $K$. iTs symbol, ReGarDed aS AN elemEnt oF ThE completEd symmETrIc Algebra $\WiDehat{\mAtHrm s}(\maThfraK P)$ and DefineD through $J_\mathCAl A(x)=e^{-x}\mathcal A(E^X)$, has the explicIT fORM $$j_\MAthcAl A(X)=\exp\left(\sum_{N\geq 1}w_{2N}^\matHcal a\MaThrM{Tr}_\matHfrak P(\mAThRM{ad}^{2n}(x))\right),\ x\in\mathfrAk P,$$ where $w_{2n}^\matHcal A$, $n\geq 1$, denoTes the inteGRAL weight oF folLOwINg wheel-like graPh:
\
We oBserve that $J_\Mathcal A$ Is anaLytic in a NeighborhOOD of $0$ in $\matHfrAk p$. we oBseRVE tHat the Cartan rELAtioNs For the sYmmEtric paIr $(\mAthFraK g,\sIgMa)$ imply imMediatelY tHaT, fOr A geNeral ELement $x$ oF $\mAthFrAk p$, $\MathrM{Ad}(x)^2$ is a Well-dEfinEd EnDOmoRphism oF $\MaTHFrak P$, tHuS all EveN pOwers Of thE AdjOint repResentatiOn rEStriCtEd To $\mathfRak p$: hence the aBoVe expressiOn Is wEll-defINEd. FinallY, all previous computationS Imply alSo tHe dirEct sUm decompoSitIon of $(A,\StaR_a)$: $$A=K\oplUs (A\staR_A\matHfRak K^{-\DElta+\fRAC{1}4\mAthRm{Tr}_\mathfrak G\CIrc\ | $\mathcal A$. Namely, bythe previo us ar gum ent ,from eve ry vertex of t h e fi rst type departs exact ly on ee dget othe o nly ver t ex o f t he s eco nd ty pe on th e posit ive real a xis ,whose colori s$(+,+)$ an d w hose operato r-v aluedpa rtc orres pon ds to deriv a tion a nd contra ct i on wit h respec t to $\m athfrak p$. The C a rt a n relation $[\ mathfr ak k, \ m ath fra k p]\subse te q \ma t hfrak p $ i m p l ies that at eachvertex of t h e f irst t yp e i n a whe el-li ke gra ph $\Gamma$ , th e edge ar riving at such a verte x must ha vecolo r e it her $ ( +,+ ) $or$ (+, -)$, whi le t he ed ge d e p a r ting fr om i t onth wheel must ha ve o p pos ite c olor. Inot her w ords,the e dg es of the cycle ina wheel-l ike g rap h$\Gam m a$ mus t h ave altern ating c o lou rs $ ( +, +)$ and $(+,-)$: t hi s , i n turn,exclud e sim m ediately $ n$- whee l s with $n$ od d.
Summ arizin g a ll previo us argum en ts, th e res t rict ion of the ope rator $\mathcal A$ t o $K$, which w e d e n ot e (im pro perly) by t he s a me s ymbo l ,def i nes a n inv er t ib l e, translation-inva ri ant di ffere ntial operato r on $K$.I t s symbol, reg a rd e d as an elemen t ofthe comple t ed symme tricalgebra$\widehat { \ mathrm S }(\ mat hfr akp ) $and defined t h r ough $ j_\math cal A(x)=e ^{- x}\ mat hca lA(e^x)$,has theex pl ic it fo rm $$ j _\mathca lA(x )= \ex p\lef t (\sum_ {n\ge q 1} W_ {2 n }^\ mathcal A\ m a thrm {t r} _\ma thf ra k p(\ math r m{a d}^{2n} (x))\righ t), \ x\i n\ ma thfrakp,$$ where $W _{ 2n}^\mathc al A$ , $n\g e q 1$, den otes the integral weigh t of fol low ing w heel -like gra ph:
\
W e o b servethat $ j_\ma th cal A $ isa n al yti cin a neigh b o rho od of $ 0$ i n $\mat hfrak p$. We obser v e t hat the Carta n r elat i o ns fo r t h e s ym m etr i c pair $(\mathfr ak g,\sigm a) $ i mply immed i ate ly that,for a g enera l elemen t $x$ of$\mathfra kp$,$ \ mat hrm{ad}(x) ^2$ is a well-def i ned e n do morph ism of $\ ma thf rak p $, thu s al l eve n powe rs of th e adj oi nt repre sentation restricted to $\mat hfrak p$ : hence t hea bov e express ionis well-de fin ed. Fina lly , allprev i ou s c o mputa tion s imply al s othe d ir ect sum dec o m p osi tionof$ (A,\st ar_A )$: $$A=K\oplus ( A \star_A\mathfr ak k ^ { -\d elt a +\fr ac {1}4\mathrm{tr }_\ ma t h frak g\c ir c\ | $\mathcal_A$. Namely,_by the previous argument,_from every_vertex_of the_first_type departs exactly_one edge to_the only vertex of_the second type_on_the positive real axis, whose color is $(+,+)$ and whose operator-valued part corresponds to_derivation_and contraction_with_respect_to $\mathfrak p$. The Cartan_relation $[\mathfrak k,\mathfrak p]\subseteq \mathfrak_p$ implies_that at each vertex of the first type_in_a wheel-like graph_$\Gamma$, the edge arriving at such a vertex must_have color either $(+,+)$ or $(+,-)$,_while the edge_departing_from_it on th wheel_must have opposite color. In other_words, the edges of the cycle_in a wheel-like graph $\Gamma$ must have_alternating colours $(+,+)$ and $(+,-)$: this,_in turn, excludes immediately $n$-wheels_with $n$_odd.
Summarizing all previous arguments, the_restriction of the_operator $\mathcal_A$ to $K$,_which we denote (improperly) by the_same symbol, defines_an invertible, translation-invariant differential operator on_$K$._Its symbol, regarded_as_an_element of_the completed symmetric_algebra_$\widehat{\mathrm S}(\mathfrak_p)$_and defined through $j_\mathcal A(x)=e^{-x}\mathcal A(e^x)$,_has_the explicit form $$j_\mathcal A(x)=\exp\left(\sum_{n\geq 1}W_{2n}^\mathcal A\mathrm{tr}_\mathfrak_p(\mathrm{ad}^{2n}(x))\right),\ x\in\mathfrak p,$$ where_$W_{2n}^\mathcal_A$, $n\geq 1$, denotes_the integral weight of following_wheel-like graph:
\
We observe that $j_\mathcal A$_is analytic_in a_neighborhood of $0$ in $\mathfrak p$. We observe that the Cartan_relations for the symmetric pair $(\mathfrak_g,\sigma)$ imply immediately that,_for a_general_element $x$ of_$\mathfrak_p$, $\mathrm{ad}(x)^2$_is a well-defined endomorphism of $\mathfrak p$,_thus all_even powers of the adjoint representation_restricted to $\mathfrak p$:_hence_the above expression is well-defined. Finally,_all previous computations imply also the_direct sum decomposition of $(A,\star_A)$:_$$A=K\oplus_(A\star_A\mathfrak_k^{-\delta+\frac{1}4\mathrm{tr}_\mathfrak g\circ\ |
0$ in $\overline{B(x_0, 0, r)}$, then we have $|\nabla^2 u| \leq C_2\frac{a}{r^2}$ on $PB_\frac{r}{4}(x_0, T)$, because in this case we can choose $\beta \Psi_2^2$ instead of $\Phi_2$ as the (space-time) comparison function.
Similar to Theorems 1.1 and 1.2 we have
\[thm 2.1\] Fix $T >0$. Let $M$ be a manifold (without boundary) of dimension $n$. Fix $x_0 \in M$ and $r>0$. Suppose $g(t)$ is a solution (not necessarily complete) to the Ricci flow on $M \times [0, T]$. Assume that for any $t\in [0, T]$ the closure of the open metric ball $B(x_0, t, r)$ is compact, and $|Rm|\leq \frac{1}{r^2}$ on the parabolic ball $PB_r(x_0, T)$. Let $ u $ be a smooth solution to the heat equation $(\partial_t-\Delta_{g(t)})u=0$ coupled to the Ricci flow on $M \times [0, T]$. Suppose $|u|\leq a$ on $PB_r(x_0, T)$, where $a$ is a positive constant. Then for any $k\geq 2$, $$|\nabla^k u| \leq C_ka(\frac{1}{r^{k}}+\frac{1}{t^{k/2}}) \hspace{2mm} \text{on} \hspace{2mm} PB_\frac{r}{2^k}(x_0, T)\setminus \{(x, 0) | x \in M\},$$ where the constant $C_k$ depends only on $k$ and the dimension.
[**Proof**]{} The proof is by induction. On $PB_\frac{r}{2}(x_0, T)\setminus \{(x, 0) | x \in M\}$ we have $$|\nabla u| \leq C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ by Theorem 1.1. For $k=2$, the result is exactly Theorem 1.2. | 0 $ in $ \overline{B(x_0, 0, r)}$, then we have $ |\nabla^2 u| \leq C_2\frac{a}{r^2}$ on $ PB_\frac{r}{4}(x_0, T)$, because in this case we can choose $ \beta \Psi_2 ^ 2 $ instead of $ \Phi_2 $ as the (space - meter) comparison affair.
Similar to Theorems 1.1 and 1.2 we have
\[thm 2.1\ ] Fix $ T > 0$. get $ M$ be a manifold (without boundary) of dimension $ n$. Fix $ x_0 \in M$ and $ r>0$. Suppose $ g(t)$ is a solution (not inevitably complete) to the Ricci flow on $ M \times [ 0, T]$. Assume that for any $ t\in [ 0, T]$ the closing of the open metric testis $ B(x_0, t, r)$ is compact, and $ |Rm|\leq \frac{1}{r^2}$ on the parabolic ball $ PB_r(x_0, T)$. get $ uracil $ be a smooth solution to the heat equation $ (\partial_t-\Delta_{g(t)})u=0 $ coupled to the Ricci flow on $ M \times [ 0, T]$. think $ |u|\leq a$ on $ PB_r(x_0, T)$, where $ a$ is a positive constant. Then for any $ k\geq 2 $, $ $ |\nabla^k u| \leq C_ka(\frac{1}{r^{k}}+\frac{1}{t^{k/2 } }) \hspace{2 mm } \text{on } \hspace{2 mm } PB_\frac{r}{2^k}(x_0, T)\setminus \{(x, 0) | x \in M\},$$ where the constant $ C_k$ depends entirely on $ k$ and the dimension.
[ * * Proof * * ] { } The proof is by induction. On $ PB_\frac{r}{2}(x_0, T)\setminus \{(x, 0) | x \in M\}$ we experience $ $ |\nabla u| \leq C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ by Theorem 1.1. For $ k=2 $, the result is exactly Theorem 1.2. | 0$ in $\overline{B(x_0, 0, r)}$, then we hxve $|\nabla^2 u| \leq C_2\frac{a}{c^2}$ on $PB_\rrac{r}{4}(x_0, T)$, because in this case we can cyoose $\beta \Psi_2^2$ instead of $\Pfi_2$ as the (space-tine) cinparison fnhction.
Sliilad to Chxorems 1.1 and 1.2 we have
\[thm 2.1\] Xix $T >0$. Let $M$ ba x lanifold (without boundary) of dimensyon $n$. Fox $x_0 \in M$ and $r>0$. Fupppfe $g(f)$ is a solution (not necessarily cojplete) uo the Ricci flow pn $M \times [0, T]$. Assume that vor wny $t\in [0, T]$ the clodure of the opeg metric ball $B(x_0, t, r)$ is compact, and $|Rm|\leq \frac{1}{r^2}$ on the parabolic bxll $PY_r(x_0, T)$. Let $ u $ bf a smooth silutijn to the heat equathon $(\paryial_t-\Delta_{g(t)})u=0$ gouplxd ti the Ricci flow on $M \times [0, T]$. Suppose $|o|\leq a$ on $[B_x(x_0, T)$, where $a$ is a posutuve cmnstdnt. Gyen fod eny $k\geq 2$, $$|\naula^k u| \leq D_ka(\frac{1}{r^{k}}+\frqc{1}{t^{k/2}}) \hspace{2mm} \ttxt{jb} \hspace{2mm} PB_\fwas{r}{2^k}(x_0, T)\setminus \{(x, 0) | x \in M\},$$ where the cotstznt $C_k$ depends only on $j$ and the dimension.
[**Prlof**]{} The pwoof is by induction. On $PB_\frac{r}{2}(x_0, T)\setminus \{(x, 0) | x \in M\}$ de kqve $$|\nxvlw u| \leq C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ by Theorem 1.1. For $k=2$, thq rtsukt is exactly Bheorem 1.2. | 0$ in $\overline{B(x_0, 0, r)}$, then we u| C_2\frac{a}{r^2}$ on T)$, because in $\beta instead of $\Phi_2$ the (space-time) comparison Similar to Theorems 1.1 and 1.2 have \[thm 2.1\] Fix $T >0$. Let $M$ be a manifold (without boundary) dimension $n$. Fix $x_0 \in M$ and $r>0$. Suppose $g(t)$ is a solution necessarily to Ricci on $M \times [0, T]$. Assume that for any $t\in [0, T]$ the closure of the metric ball $B(x_0, t, r)$ is compact, and \frac{1}{r^2}$ on the parabolic $PB_r(x_0, T)$. Let $ u be smooth solution the equation coupled to the flow on $M \times [0, T]$. Suppose $|u|\leq a$ on $PB_r(x_0, T)$, where $a$ is a positive Then for 2$, $$|\nabla^k \leq \hspace{2mm} \hspace{2mm} PB_\frac{r}{2^k}(x_0, T)\setminus | x \in M\},$$ where the only on $k$ and the dimension. [**Proof**]{} The is by On $PB_\frac{r}{2}(x_0, T)\setminus \{(x, 0) | \in M\}$ we have $$|\nabla u| \leq C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ Theorem 1.1. For $k=2$, the result is exactly Theorem 1.2. | 0$ in $\overline{B(x_0, 0, r)}$, then we have $|\naBla^2 u| \leq C_2\frAc{a}{r^2}$ oN $PB_\FraC{r}{4}(X_0, T)$, beCausE in this case we cAN choOse $\beta \Psi_2^2$ instead of $\Phi_2$ As the (SpACe-tiME) cOmparIson funCTiON.
simIlAr To THeOReMs 1.1 and 1.2 We hAve
\[thm 2.1\] FIx $T >0$. Let $M$ be a ManIfOld (without boUNdAry) of dimenSioN $n$. Fix $x_0 \in M$ and $R>0$. SuPpose $g(T)$ iS a sOLutioN (noT neceSsarilY CompleTe) to the RiCcI Flow on $m \Times [0, T]$. ASSUmE thaT for any $t\in [0, T]$ the cloSUrE Of the open metriC ball $B(X_0, t, R)$ Is COMpaCt, aNd $|Rm|\leq \fraC{1}{r^2}$ On the PAraboliC BaLL $pb_r(x_0, t)$. let $ u $ be a smooth Solution to tHE heAt equaTiOn $(\pARtial_t-\delta_{G(t)})U=0$ CouPled to the RiCci fLow on $M \timEs [0, T]$. SupPOse $|u|\leq A$ On $PB_r(x_0, T)$, Where $a$ Is a PosItivE CoNsTanT. THEn fOR aNy $k\GEq 2$, $$|\nAbla^k u| \leQ C_Ka(\Frac{1}{r^{K}}+\fraC{1}{T^{K/2}}) \HSpacE{2mm} \Text{On} \hspAce{2mm} PB_\frac{r}{2^k}(X_0, T)\sEtmiNUs \{(x, 0) | X \in M\},$$ wHere tHe coNsTant $C_K$ depenDs onlY oN $k$ and the dimensiOn.
[**PrOof**]{} The proOf iS bY inDuCtion. oN $PB_\fraC{r}{2}(x_0, t)\seTminus \{(x, 0) | X \in M\}$ we hAVe $$|\nAbLA U| \LeQ C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ bY THEOrEm 1.1. For $k=2$, thE resulT Is ExACtly TheoReM 1.2. | 0$ in $\overline{B(x_0, 0, r)}$, the n wehav e $ |\ nabl a^2u| \leq C_2\fr a c{a} {r^2}$ on $PB_\frac{r} {4}(x _0 , T)$ , b ecaus e in th i sc a sewe c anch o os e $\b eta \Psi_2 ^2$ instea d o f$\Phi_2$ ast he (space-ti me) comparisonfun ction.
Sim i lar t o T heore ms 1.1 and 1. 2 we have
\ [thm 2 . 1\] Fix $ T>0$. Let $M$ be a man i fo l d (without bou ndary) o f d i m ens ion $n$. Fix$x _0 \i n M$ and $r > 0 $ . S u ppose $g(t)$is a soluti o n ( not ne ce ssa r ily co mplet e) tothe Ricci f lowon $M \ti mes [0 , T]$. A s sume th at for an y $ t\in [0 ,T]$ t h e c l os ure ofthe open m et ric b all$ B ( x _0,t,r)$is co mpact, and $| Rm| \leq \fr ac{1} {r^2} $ on t he pa raboli c bal l$PB_r(x_0, T)$ . Le t $ u $ b e a s moo th solu t ion to th e h eat equ ation $ ( \pa rt i a l _t -\Delta_{g(t)})u=0 $c o up led to t he Ric c ifl o w on $M\t ime s [0 , T]$.Supp o se $|u|\le q a$ o n $ PB _r(x_0, T )$, wh er e $ a$is ap osit ive co nstant.Thenf or any $k\geq2 $, $$|\nabla^ k u | \l e q C_ ka( \frac{1}{r^ {k}} + \fra c{1} { t^ {k/ 2 }}) \hs pa c e{ 2 mm} \text{on} \h sp ace{2m m} PB_\frac{r}{ 2^k}(x_0,T ) \setminu s \{ ( x, 0) | x \in M\} ,$$ w here the c o nstant $ C_k$dependsonly on $ k $ and the di men sio n.[ ** Proof**]{} Th e proo fis by i ndu ction.On$PB _\f rac {r }{2}(x_0, T)\set mi nu s\{ (x, 0) | x \in M\ }$ we h ave $$|\ n abla u | \le q C_ 1a (\ f rac {1}{r}+ \ fr a c {1}{ \s qr t{t} })$ $by Th eore m 1. 1. For$k=2$, th e r e sult i sexactly Theorem 1.2. | 0$ in_$\overline{B(x_0, 0,_r)}$, then we have_$|\nabla^2 u|_\leq_C_2\frac{a}{r^2}$ on_$PB_\frac{r}{4}(x_0,_T)$, because in_this case we_can choose $\beta \Psi_2^2$_instead of $\Phi_2$_as_the (space-time) comparison function.
Similar to Theorems 1.1 and 1.2 we have
\[thm 2.1\] Fix $T_>0$._Let $M$_be_a_manifold (without boundary) of dimension_$n$. Fix $x_0 \in M$_and $r>0$._Suppose $g(t)$ is a solution (not necessarily complete)_to_the Ricci flow_on $M \times [0, T]$. Assume that for any_$t\in [0, T]$ the closure of_the open metric_ball_$B(x_0,_t, r)$ is compact,_and $|Rm|\leq \frac{1}{r^2}$ on the parabolic_ball $PB_r(x_0, T)$. Let $_u $ be a smooth solution to_the heat equation $(\partial_t-\Delta_{g(t)})u=0$ coupled to_the Ricci flow on $M_\times [0,_T]$. Suppose $|u|\leq a$ on_$PB_r(x_0, T)$, where_$a$ is_a positive constant._Then for any $k\geq 2$, $$|\nabla^k_u| \leq C_ka(\frac{1}{r^{k}}+\frac{1}{t^{k/2}})_ \hspace{2mm} \text{on}__ \hspace{2mm} ___PB_\frac{r}{2^k}(x_0, _T)\setminus \{(x, 0)_|_x \in_M\},$$_where the constant $C_k$ depends only_on_$k$ and the dimension.
[**Proof**]{} The proof is_by induction. On $PB_\frac{r}{2}(x_0,__T)\setminus \{(x, 0) |_x \in M\}$ we have_$$|\nabla u| \leq C_1a(\frac{1}{r}+\frac{1}{\sqrt{t}})$$ by Theorem_1.1. For_$k=2$, the_result is exactly Theorem 1.2. |
locally, which leads to their decoherence when the photons are transmitted over a quantum channel with environment noise or stored in practical quantum devices. Quantum repeater is a necessary technique for long-distance quantum communication and it is used to overcome the influence from this decoherence [@repeater]. In fact, the optimal way to overcome the influence on photon systems from channel noise in quantum communication is the self-error-rejecting qubit transmission [@LIXHAPL] with linear optics as it is an active way to decrease the influence from channel noise and it is very efficient and simple to be implemented in experiment with current feasible techniques. However, this scheme [@LIXHAPL] can only depress most of the influence from the channel noise in the process of photon distribution, as the same as the other active methods for overcoming the influence from noise [@DFSa1; @DFSa2; @DFSa3]. It does not work in depressing the influence of noise from both a long-distance channel and the storage process for quantum states. Moreover, quantum repeaters for long-distance quantum communication require the entangled photons with higher fidelity (usually $\sim$ 99%) beyond that from faithful qubit transmission schemes (about 90% $\sim$ 96% for a polarization quantum state of photons over an optical-fiber channel with several kilometers). That is, entanglement concentration and entanglement purification are not only useful but also absolutely necessary in long-distance quantum communication.
Entanglement concentration is used to distill some nonlocal entangled systems in a maximally entangled state from a set of nonlocal entangled systems in a partially entangled pure state [@ECP1]. Before 2013, entanglement concentration is focused on the nonlocal quantum states in one DOF, such as the polarization states of photons, the two-level quantum states of atom systems, or the spins of electron systems. The first entanglement concentration protocol (ECP) was proposed by Bennett et al. [@ECP1] in 1996, which is based on the Schmidt projection [@ECP1]. Also, it is just a mathematic method for entanglement concentration. In 2001, two ECPs were proposed [@ECP5; @ECP4] with PBSs for two ideal entangled photon sources. In 2008, Sheng et al. [@ECP6] proposed a repeatable ECP to concentrate both bipartite and multipartite quantum systems, and it has an advantage of far higher efficiency and yield than those in Bennett’s ECP [@ECP1] and the PBS | locally, which leads to their decoherence when the photons are transmitted over a quantum channel with environment randomness or store in practical quantum devices. Quantum recidivist is a necessary proficiency for long - distance quantum communication and it is used to overcome the influence from this decoherence [ @repeater ]. In fact, the optimum room to overcome the influence on photon organization from channel noise in quantum communication is the self - erroneousness - rejecting qubit transmission [ @LIXHAPL ] with analogue optics as it is an active way to decrease the influence from channel noise and it is very efficient and simple to be implemented in experiment with current feasible technique. However, this scheme [ @LIXHAPL ] can only depress most of the influence from the distribution channel noise in the process of photon distribution, as the same as the early active methods for get the best the influence from noise [ @DFSa1; @DFSa2; @DFSa3 ]. It does not work in depressing the influence of randomness from both a long - distance channel and the storage process for quantum states. Moreover, quantum repeaters for retentive - distance quantum communication require the entangled photons with higher fidelity (usually $ \sim$ 99 %) beyond that from faithful qubit transmission schemes (about 90% $ \sim$ 96% for a polarization quantum state of photon over an ocular - fiber distribution channel with several kilometers). That is, entanglement concentration and web purification are not merely useful but also absolutely necessary in retentive - distance quantum communication.
Entanglement concentration is used to distill some nonlocal entangled systems in a maximally entangled country from a set of nonlocal entangled systems in a partially entangled arrant state [ @ECP1 ]. Before 2013, web assiduity is concentrate on the nonlocal quantum states in one DOF, such as the polarization states of photons, the two - level quantum states of atom systems, or the spins of electron system. The beginning entanglement concentration protocol (ECP) was nominate by Bennett et al. [ @ECP1 ] in 1996, which is based on the Schmidt protrusion [ @ECP1 ]. Also, it is just a mathematic method acting for entanglement concentration. In 2001, two ECPs were proposed [ @ECP5; @ECP4 ] with PBSs for two ideal embroiled photon sources. In 2008, Sheng et al. [ @ECP6 ] proposed a repeatable ECP to condense both bipartite and multipartite quantum systems, and it has an advantage of far high efficiency and yield than those in Bennett ’s ECP [ @ECP1 ] and the PBS | lofally, which leads to thelr decoherence wkwn the photohs are tfansmitted over a quantum chenneo wity environment noise or stored ij practixal wyantum devmdes. Quakcum rsieatex ms a necessary jechnique fos long-distance qjautum communication and it is used to overcoke the influence frok thia decoherence [@repeater]. In fact, the optiman way to overvome the influence on photln sjstems from channep noise in wuaneym communicagion is tht velf-error-rgjecting qubit transmission [@LIXHXPL] wnth linear ipricd as it is ai actine way to decvvase tha influrnce from chankel nmisw and it is very effirient and simple to fe implemanced in experiment wity xurrett faasicoe gecgnmquss. Howfvec, this scheje [@LIXHAPL] xan only depress moxt if the influehce frjm the channel noise in the process of phmtoh distribution, as the sqme as the other actige methodf for overcoming the influence from noise [@DFSa1; @DFVa2; @DFAx3]. Iu eoes vit work in depressing the influence of noise frjj nonh a long-distance channel anc hhr storage procgss for quzntum states. Moreoger, quagtum eepeaters for long-distance quantum commubication reqlire the entangled phocons with hiyher fodeliyy (usually $\sim$ 99%) beyond chat fdom faithfup qubit tdxnsmission schemds (sbmut 90% $\sim$ 96% for a polarizatijn quantun stcte of pfotoms ovew an opticwl-fibcs channel with sevfral ninometers). Tjat is, entanglement concentration and entanglgmett iurificatnon arc not only useftl but also abxolutelv necersary in lkng-distence quantum communicatiot.
Gntanglement roncentraeion is ysed to aistill some npnlocal euuangled sywtems in a maximalky gnfangled state fxun a set of nonlpcau egtwnjled ffstems in a [artkaluu entxngled kbre rtatr [@ECP1]. Before 2013, entangnemeht concentration ix nocused ob the noglocal quantuk states in one DOV, suci as tie polsrieation states of photons, the twk-level quwntmm states of wtom systems, or che spins of electron systems. The first xntanglement concentrajion protocol (ECP) was kroposed by Bxnnett et al. [@EC[1] in 1996, which is based on the Schmidt pvojection [@ECP1]. Also, it js jusd a mwthematic method for entanglement concentration. In 2001, two ECPs were proposee [@ECP5; @XC[4] with PBSs for two pdecl zntanglqd piovon sources. In 2008, Sneng et al. [@ECP6] proposed a repeavable ECP do concentrate both bipartite sna multipartitg quantum systems, and it hzs an adfantage of far higher efficiencu and yield than those ik Bennett’w ECP [@ECP1] and tie PBS | locally, which leads to their decoherence when are over a channel with environment quantum Quantum repeater is necessary technique for quantum communication and it is used overcome the influence from this decoherence [@repeater]. In fact, the optimal way to the influence on photon systems from channel noise in quantum communication is the qubit [@LIXHAPL] linear as it is an active way to decrease the influence from channel noise and it is efficient and simple to be implemented in experiment current feasible techniques. However, scheme [@LIXHAPL] can only depress of influence from channel in process of photon as the same as the other active methods for overcoming the influence from noise [@DFSa1; @DFSa2; @DFSa3]. does not depressing the of from a long-distance channel storage process for quantum states. Moreover, long-distance quantum communication require the entangled photons with fidelity (usually 99%) beyond that from faithful qubit schemes (about 90% $\sim$ 96% for a polarization state of photons over an optical-fiber channel with several kilometers). That is, entanglement concentration and are not only useful also absolutely necessary long-distance communication. concentration used to some nonlocal entangled systems in a maximally entangled state from a of nonlocal entangled systems in a partially entangled pure state 2013, concentration is focused the nonlocal quantum states one such as the polarization photons, two-level atom or spins of electron systems. first entanglement concentration protocol (ECP) proposed by Bennett et is based on the Schmidt projection [@ECP1]. Also, is just a mathematic method for entanglement In 2001, two ECPs were proposed [@ECP5; @ECP4] with PBSs for two entangled photon 2008, Sheng et al. [@ECP6] proposed a repeatable to concentrate both bipartite multipartite quantum systems, and it has an advantage of higher and yield those in Bennett’s [@ECP1] and the | locally, which leads to their dEcoherence When tHe pHotOnS are TranSmitted over a quANtum Channel with environment Noise Or SToreD In PractIcal quaNTuM DEviCeS. QUanTuM RePeateR is A necessAry techniqUe fOr Long-distance QUaNtum communIcaTion and it is uSed To overCoMe tHE inflUenCe froM this dECohereNce [@repeatEr]. iN fact, tHE optimaL WAy To ovErcome the influencE On PHoton systems frOm chanNeL NoISE in QuaNtum communIcAtion IS the selF-ErROR-RejECting qubit traNsmission [@LIxhAPl] with lInEar OPtics aS it is An ACtiVe way to decrEase The influeNce froM Channel NOise and It is veRy eFfiCienT AnD sImpLe TO be IMpLemENteD in experImEnT with CurrENT FEasiBle TechNiqueS. However, this sCheMe [@LIxhAPl] can oNly dePresS mOst of The infLuencE fRom the channel noIse iN the proceSs oF pHotOn DistrIBution, As tHe sAme as thE other aCTivE mETHOdS for overcoming the iNfLUEnCe from noIse [@DFSA1; @dFsa2; @dfSa3]. It doeS nOt wOrk iN DEpresSing THe InfluencE of noiSE fRoM both a lOnG-distaNcE chAnnEl and THe stOrage pRocess foR quanTUm states. MoreovER, quantum repeaTErS FOr LOng-dIstAnce quantum CommUNicaTion REqUirE The enTanglEd PHoTOns with higher fideliTy (UsuallY $\sim$ 99%) bEyond that from Faithful quBIT TransmisSion SChEMes (about 90% $\sim$ 96% for A polaRization quANtum statE of phOtons oveR an opticaL-FIber chanNel WitH seVerAL KiLometers). That iS, ENtanGlEment coNceNtratioN anD enTanGleMeNt purificAtion are NoT oNlY uSefUl but ALso absolUtEly NeCesSary iN Long-diStancE quaNtUm COmmUnicatiON.
ENTAnglEmEnT conCenTrAtion Is usED to Distill Some nonloCal ENtanGlEd Systems In a maximally eNtAngled statE fRom A set of NONlocal enTangled systems in a partiaLLy entanGleD pure StatE [@ECP1]. BeforE 2013, enTangleMenT ConcenTratioN is foCuSed ON The noNLOcAl qUaNtum states IN One dOF, suCh As thE polariZation states of photONs, tHe two-level quaNtuM staTES oF atOM sYSteMs, OR thE SPins of electron sYstems. The fIrST eNtanglemenT ConCeNtratioN protocOl (ECP) WAs propoSed by BennEtt et al. [@ECp1] iN 1996, whiCH Is bAsed on the SChmidt prOjection [@Ecp1]. Also, IT iS just A maThematIc MetHod foR entanGLemEnt coNcentrAtIon. In 2001, tWo ECPS wEre propoSed [@ECP5; @ECP4] with PBSs for two Ideal eNtangLed Photon souRceS. in 2008, SHeng et al. [@EcP6] prOposed a repEatAblE ECP tO coNCentrAte bOTh BipARtite And mULtipartitE QuAntUM SyStems, and it hAS AN adVantaGe oF Far higHer eFficiency and yield THan those in BennEtt’s ecp [@ECp1] anD The PbS | locally, which leads to t heir decoh erenc e w hen t he p hoto ns are transmi t tedover a quantum channel with e n viro n me nt no ise ors to r e d i npr act ic a lquant umdevices . Quantumrep ea ter is a nec e ss ary techni que for long-di sta nce qu an tum commu nic ation and i t is us ed to ove rc o me the influen c e f romthis decoherence[ @r e peater]. In fa ct, th eo pt i m alway to overco me thei nfluenc e o n p hot o n systems fro m channel n o ise in qu an tum commun icati on isthe self-er ror- rejecting qubit transmi s sion [@ LIXHAP L]wit h li n ea ropt ic s as it is anactive w ay t o dec reas e t h e in flu ence from channel nois e a nd i t is very effi cien tand s impleto be i mplemented in e xper iment wit h c ur ren tfeasi b le tec hni que s. Howe ver, th i s s ch e m e [ @LIXHAPL] can only d e p re ss mostof the in fl u ence fro mthe cha n n el no isei nthe proc ess of ph ot on dist ri bution ,asthe same as t he oth er activ e met h ods for overco m ing the influ e nc e fr o m no ise [@DFSa1; @ DFSa 2 ; @D FSa3 ] .Itd oes n ot wo rk in depressing the infl ue nce of nois e from both a long-dist a n c e channe l an d t h e storage proc ess f or quantum states.Moreo ver, qua ntum repe a t ers forlon g-d ist anc e qu antum communi c a tion r equirethe entang led ph oto nswi th higher fidelit y(u su al ly$\sim $ 99%) be yo ndth atfromf aithfu l qub it t ra ns m iss ion sch e me s (abo ut 9 0% $ \si m$ 96%fora po larizat ion quant ums tate o fphotons over an opti ca l-fiber ch an nel withs e veral ki lometers). That is, ent a nglemen t c oncen trat ion and e nta ngleme ntp urific ationare n ot on l y usef u l b utal so absolut e l y n ecess ar y in long-d istance quantum co m mun ication.
Ent ang leme n t c onc e nt r ati on isu s ed to distill s ome nonloc al en tangled sy s tem sin a ma ximally enta n gled st ate froma set ofno nloc a l en tangled sy stems in a partia l ly en t an gledpur e stat e[@E CP1]. Befor e 20 13, e ntangl em ent co ncent ra tion isfocused on the nonlocal quant um st ate s in oneDOF , su ch as the pol arizationsta tes of p hot o ns, t he t w o- lev e l qua ntum states of at oms y st ems, or the s p ins of e lec t ron sy stem s. The first enta n glement concen trat i o n p rot o col(E CP) was propos edby B ennett e tal. [@ECP1] in 1996 ,w hichis bas ed onthe Sch m i dt projec tion [@ ECP1]. Al so, i t is jus tam athema ticme thod f or ent a ngle m e nt concentration . In2 0 01, t w o E CPs w er e propo s ed [ @ECP5; @EC P4] with PB Ss for two idea l entan gl ed pho ton s ources. In 2008, She ng et al. [@ EC P6]pro poseda re p e atabl e EC Ptoconcentra t e b o th b i par tite andmu ltip artite qu a ntum sys tem s , and i thas a n adva n ta g e of far hi ghe r eff i c iency andy ield th a n tho se inBennet t’s ECP [@E CP 1] andthe P BS | locally,_which leads_to their decoherence when_the photons_are_transmitted over_a_quantum channel with_environment noise or_stored in practical quantum_devices. Quantum repeater_is_a necessary technique for long-distance quantum communication and it is used to overcome the_influence_from this_decoherence_[@repeater]._In fact, the optimal way_to overcome the influence on_photon systems_from channel noise in quantum communication is the_self-error-rejecting_qubit transmission [@LIXHAPL]_with linear optics as it is an active way_to decrease the influence from channel_noise and it_is_very_efficient and simple to_be implemented in experiment with current_feasible techniques. However, this scheme [@LIXHAPL]_can only depress most of the influence_from the channel noise in the_process of photon distribution, as_the same_as the other active methods_for overcoming the_influence from_noise [@DFSa1; @DFSa2;_@DFSa3]. It does not work in_depressing the influence_of noise from both a long-distance_channel_and the storage_process_for_quantum states._Moreover, quantum repeaters_for_long-distance quantum_communication_require the entangled photons with higher_fidelity_(usually $\sim$ 99%) beyond that from faithful_qubit transmission schemes (about_90%_$\sim$ 96% for a_polarization quantum state of photons_over an optical-fiber channel with several_kilometers). That_is, entanglement_concentration and entanglement purification are not only useful but also absolutely_necessary in long-distance quantum communication.
Entanglement concentration_is used to distill_some nonlocal_entangled_systems in a_maximally_entangled state_from a set of nonlocal entangled systems_in a_partially entangled pure state [@ECP1]. Before_2013, entanglement concentration is_focused_on the nonlocal quantum states in_one DOF, such as the polarization_states of photons, the two-level_quantum_states_of atom systems, or the_spins of electron systems. The first_entanglement concentration protocol_(ECP) was proposed by Bennett et al._[@ECP1]_in 1996, which is based on_the_Schmidt projection [@ECP1]. Also, it is_just_a_mathematic method for entanglement concentration._In 2001, two ECPs were proposed_[@ECP5; @ECP4] with PBSs for two ideal entangled photon_sources. In 2008,_Sheng et al. [@ECP6] proposed_a_repeatable_ECP to concentrate both bipartite and multipartite quantum systems, and_it has_an advantage of_far higher efficiency and yield than those in Bennett’s ECP_[@ECP1] and the PBS |
our approach on a formal level insofar, that we have not addressed the question how an infinite dimensional space (of fields) can be treated in a braided category. This is certainly necessary if we want to do quantum field theory, i.e., deal with infinitely many degrees of freedom. An obvious problem is the definition of the coevaluation. It seems that we need at least a completed tensor product for this. However, instead of introducing heavy functional analytic machinery, we can stick with our algebraic approach given a further assumption.
Let us assume that the space of (regular) fields $X$ decomposes into a direct sum $\bigoplus_i X_i$ of countably many finite dimensional comodules under the symmetry quantum group $A$. This corresponds roughly to the classical case of the space-time manifold being compact. In particular, it is the case if the symmetry quantum group $A$ is cosemisimple (or classically the Lie group of symmetries is compact, see Section \[sec:PeterWeyl\] below). Denote the projection $X\to X_i$ by $\tau_i$.
We now allow arbitrary sums of elements in $X$ given that any projection $\tau_i$ annihilates all but finitely many summands. Similarly, we allow infinite sums in the $n$-fold tensor product $X^n$ with the restriction that any projection $\tau_{i_1}\tens\cdots\tens\tau_{i_n}$ yields a finite sum. To define the dual of $X$, we take the dual of each $X_i$ and set $X^*=\bigoplus_i X_i^*$. For each component $X_i$ we have an evaluation map $\ev_i:X_i\tens
X_i^*\to\k$ and a coevaluation map $\coev_i:\k\to X_i\tens X_i^*$ in the usual way. We then formally define $\ev=\sum_i
\ev_i\circ(\tau_i\tens\tau_i^*)$ and $\coev=\sum_i \coev_i$.
Our definition is invariant under coactions of $A$ as it should be, since the projections $\tau_i$ commute with the coaction of $A$. In particular, it is invariant under braidings.
Cosemisimplicity and Peter-Weyl De | our approach on a formal level insofar, that we have not addressed the doubt how an countless dimensional space (of fields) can be treat in a braid category. This is certainly necessary if we desire to do quantum field theory, i.e., cover with infinitely many degrees of exemption. An obvious problem is the definition of the coevaluation. It seems that we need at least a completed tensor merchandise for this. However, instead of introducing heavy functional analytic machinery, we can stick with our algebraic access given a further assumption.
Let us wear that the space of (regular) fields $ X$ decomposes into a lineal sum $ \bigoplus_i X_i$ of countably many finite dimensional comodules under the isotropy quantum group $ A$. This corresponds roughly to the classical case of the distance - time manifold being compact. In particular, it is the case if the isotropy quantum group $ A$ is cosemisimple (or classically the Lie group of symmetry is compact, see Section \[sec: PeterWeyl\ ] below). Denote the projection $ X\to X_i$ by $ \tau_i$.
We now allow arbitrary sums of elements in $ X$ given that any project $ \tau_i$ annihilates all but finitely many summands. Similarly, we allow infinite sum in the $ n$-fold tensor intersection $ X^n$ with the restriction that any projection $ \tau_{i_1}\tens\cdots\tens\tau_{i_n}$ give a finite sum. To define the dual of $ X$, we take the dual of each $ X_i$ and set $ X^*=\bigoplus_i X_i^*$. For each part $ X_i$ we have an evaluation map $ \ev_i: X_i\tens
X_i^*\to\k$ and a coevaluation map $ \coev_i:\k\to X_i\tens X_i^*$ in the usual way. We then formally define $ \ev=\sum_i
\ev_i\circ(\tau_i\tens\tau_i^*)$ and $ \coev=\sum_i \coev_i$.
Our definition is invariant under coactions of $ A$ as it should be, since the projections $ \tau_i$ commute with the coaction of $ A$. In particular, it is invariant under braid.
Cosemisimplicity and Peter - Weyl De | oug approach on a formal ltvel insofar, that we havx not asdressed the question how an infinitx dinensiinal space (of fields) cxn be trewted in q breided category. Tijs is ccxtainmn neczswary if we wanj to do quandum field theosy, i.z., deal with infinitely many degrees jf freecol. An obvious ptoblek is fhe definition of the coevaluation. It seeks that we nerd at least a completed tejsor product for this. Jowever, insjsad if introducivg heavy fluctional anzlytic machinery, we can stick wkth obr algebraix qppgmach given e furtrer assumption.
Let us assume that the spacc of (ceguoar) fields $X$ decomposxs into a direct sum $\bigoplus_h R_i$ of countably many dibite gimetsiovql zomkdnlea undeg tie symmetry quantum griup $A$. This correspomdf roughly to tge clafsycal case of the space-time manifold beitg dompact. In particular, ir is the case if the dymmetry zuantum group $A$ is cosemisimple (or classically tha Lie erobi ud dymmetries is compact, see Section \[sec:PeterWeyl\] gekoe). Denote the pvojection $X\to X_i$ bu $\hai_y$.
We now allow arbitxzrg sums of elements in $X$ gyven rhat any krojevtion $\tau_i$ annihilates all vut finitely nany summands. Simiparly, we aluow onfinote sums in the $n$-fold tznsor lroduct $X^n$ aith the ddstriction that xny psojection $\tau_{i_1}\tens\cdots\tenf\tau_{i_n}$ yixlds c finite sum. To desine the dkal on $X$, we take the duap of gach $X_h$ and set $D^*=\bigoplus_i X_i^*$. For each componenv $X_i$ we have sn evdluation map $\cv_i:X_i\tens
X_i^*\to\k$ wnd a coevaluajion map $\eoev_i:\k\go X_i\tens Q_i^*$ in the usual way. Re then formanpy define $\ev=\vum_i
\ev_i\cyrc(\tqu_i\twns\tau_i^*)$ xnd $\coev=\sum_i \cpev_i$.
Our dvfnnition iw invariant under goactkkns of $A$ as it whoood be, since thr pfojqcniois $\tat_h$ commute widh tfe zpactiun of $A$. In iargicukar, it is invariant gnded braidings.
Cosemisomilicity abd Peter-Reyl De | our approach on a formal level insofar, have addressed the how an infinite be in a braided This is certainly if we want to do quantum theory, i.e., deal with infinitely many degrees of freedom. An obvious problem is definition of the coevaluation. It seems that we need at least a completed product this. instead introducing heavy functional analytic machinery, we can stick with our algebraic approach given a further assumption. us assume that the space of (regular) fields decomposes into a direct $\bigoplus_i X_i$ of countably many dimensional under the quantum $A$. corresponds roughly to classical case of the space-time manifold being compact. In particular, it is the case if the symmetry group $A$ (or classically Lie of is compact, see below). Denote the projection $X\to X_i$ now allow arbitrary sums of elements in $X$ that any $\tau_i$ annihilates all but finitely many Similarly, we allow infinite sums in the $n$-fold product $X^n$ with the restriction that any projection $\tau_{i_1}\tens\cdots\tens\tau_{i_n}$ yields a finite sum. To define of $X$, we take dual of each and $X^*=\bigoplus_i For component $X_i$ have an evaluation map $\ev_i:X_i\tens X_i^*\to\k$ and a coevaluation map $\coev_i:\k\to X_i^*$ in the usual way. We then formally define $\ev=\sum_i $\coev=\sum_i Our definition is under coactions of $A$ it be, since the projections with coaction particular, is under braidings. Cosemisimplicity and De | our approach on a formal level Insofar, thaT we haVe nOt aDdRessEd thE question how an INfinIte dimensional space (of fIelds) CaN Be trEAtEd in a Braided CAtEGOry. thIs Is cErTAiNly neCesSary if wE want to do qUanTuM field theory, I.E., dEal with infIniTely many degrEes Of freeDoM. An OBviouS prOblem Is the dEFinitiOn of the coEvALuatioN. it seems THAt We neEd at least a completED tENsor product for This. HoWeVEr, INSteAd oF introduciNg Heavy FUnctionAL aNALYtiC Machinery, we caN stick with oUR alGebraiC aPprOAch givEn a fuRtHEr aSsumption.
LeT us aSsume that The spaCE of (reguLAr) fieldS $X$ decoMpoSes Into A DiReCt sUm $\BIgoPLuS_i X_I$ Of cOuntably MaNy FinitE dimENSIOnal ComOdulEs undEr the symmetry QuaNtum GRouP $A$. ThiS corrEspoNdS rougHly to tHe claSsIcal case of the spAce-tIme manifoLd bEiNg cOmPact. IN ParticUlaR, it Is the caSe if the SYmmEtRY QUaNtum group $A$ is cosemiSiMPLe (Or classiCally tHE LIe GRoup of syMmEtrIes iS COmpacT, see sEcTion \[sec:PEterWeYL\] bElOw). DenotE tHe projEcTioN $X\tO X_i$ by $\TAu_i$.
WE now alLow arbitRary sUMs of elements in $x$ Given that any pROjECTiON $\tau_I$ anNihilates alL but FInitEly mANy SumMAnds. SImilaRlY, We ALlow infinite sums in tHe $N$-fold tEnsor Product $X^n$ with The restricTION that any ProjECtIOn $\tau_{i_1}\tens\cdotS\tens\Tau_{i_n}$ yieldS A finite sUm. To dEfine the Dual of $X$, we TAKe the duaL of EacH $X_i$ And SET $X^*=\Bigoplus_i X_i^*$. FoR EAch cOmPonent $X_I$ we Have an eValUatIon Map $\Ev_I:X_i\tens
X_i^*\To\k$ and a cOeVaLuAtIon Map $\coEV_i:\k\to X_i\tEnS X_i^*$ In The Usual WAy. We thEn forMallY dEfINe $\eV=\sum_i
\ev_I\CiRC(\Tau_i\TeNs\Tau_i^*)$ And $\CoEv=\sum_I \coeV_I$.
OuR definiTion is invAriANt unDeR cOactionS of $A$ as it shoulD bE, since the pRoJecTions $\tAU_I$ commute With the coaction of $A$. In parTIcular, iT is InvarIant Under braiDinGs.
CoseMisIMpliciTy and PEter-WEyL De | our approach on a formallevel inso far,tha t w ehave not addressed the ques tion how an infinite d imens io n al s p ac e (of fields ) c a n be t re ate di na bra ide d categ ory. Thisisce rtainly nece s sa ry if we w ant to do quant umfieldth eor y , i.e .,dealwith i n finite ly many d eg r ees of freedom . An obv ious problem is t h ed efinition of t he coe va l ua t i on. It seems tha twe ne e d at le a st a com p leted tensorproduct for thi s. How ev er, instea d ofin t rod ucing heavy fun ctional a nalyti c machin e ry, wecan st ick wi th o u ral geb ra i c a p pr oac h gi ven a fu rt he r ass umpt i o n .
Le t u s as sumethat the spac e o f (r e gul ar) f ields $X$ d ecomp oses i nto a d irect sum $\big oplu s_i X_i$ofco unt ab ly ma n y fini tedim ensiona l comod u les u n d e rthe symmetry quant um g ro up $A$.This c o rr es p onds rou gh lyto t h e clas sica l c ase of t he spa c e- ti me mani fo ld bei ng co mpa ct. I n par ticula r, it is thec ase if the sym m etry quantumg ro u p $ A $ is co semisimple(orc lass ical l ythe Lie g roupof sy m metries is compact, s ee Sec tion\[sec:PeterWe yl\] below ) . Denote t he p r oj e ction $X\to X_ i$ by $\tau_i$.
We nowallow arbitra ry sums o f elements in $X $ g ive n th at any projec t i on $ \t au_i$ a nni hilates al l b utfin it ely manysummands .Si mi la rly , wea llow inf in ite s ums in t h e $n$- foldtens or p r odu ct $X^n $ w i t h th ere stri cti on that any pro jection $\tau_{i _1} \ tens \c do ts\tens \tau_{i_n}$ y ie lds a fini te su m. Tod e fine the dual of $X$, we take t h e dualofeach$X_i $ and set $X ^*=\bi gop l us_i X _i^*$. Forea chc o mpone n t $ X_i $we have an e val uatio nmap$\ev_i: X_i\tens
X_i^*\to\ k $ a nd a coevalua tio n ma p $\ coe v _i : \k\ to X_i \ t ens X_i^*$ in t he usual w ay . W e then for m all ydefine$\ev=\s um_i\ ev_i\ci rc(\tau_i \tens\tau _i ^*)$ a nd$\coev=\su m_i \coe v_i$.
Ou r defi n it ion i s i nvaria nt un der c oactio n s o f $A$ as it s houldbe, s in ce the p rojections $\tau_i$ com mute w ith t hecoactionof$ A$. In parti cula r, it is i nva ria nt un der braid ings .
Cos e misim plic i ty and Pe t er -We y l D e | our_approach on_a formal level insofar,_that we_have_not addressed_the_question how an_infinite dimensional space_(of fields) can be_treated in a_braided_category. This is certainly necessary if we want to do quantum field theory, i.e.,_deal_with infinitely_many_degrees_of freedom. An obvious problem_is the definition of the_coevaluation. It_seems that we need at least a completed_tensor_product for this._However, instead of introducing heavy functional analytic machinery, we_can stick with our algebraic approach_given a further_assumption.
Let_us_assume that the space_of (regular) fields $X$ decomposes into_a direct sum $\bigoplus_i X_i$ of_countably many finite dimensional comodules under the_symmetry quantum group $A$. This corresponds_roughly to the classical case_of the_space-time manifold being compact. In_particular, it is_the case_if the symmetry_quantum group $A$ is cosemisimple (or_classically the Lie_group of symmetries is compact, see_Section \[sec:PeterWeyl\]_below). Denote the_projection_$X\to_X_i$ by_$\tau_i$.
We now allow_arbitrary_sums of_elements_in $X$ given that any projection_$\tau_i$_annihilates all but finitely many summands. Similarly,_we allow infinite sums_in_the $n$-fold tensor product_$X^n$ with the restriction that_any projection $\tau_{i_1}\tens\cdots\tens\tau_{i_n}$ yields a finite_sum. To_define the_dual of $X$, we take the dual of each $X_i$ and_set $X^*=\bigoplus_i X_i^*$. For each component_$X_i$ we have an_evaluation map_$\ev_i:X_i\tens
X_i^*\to\k$_and a coevaluation_map_$\coev_i:\k\to X_i\tens_X_i^*$ in the usual way. We then_formally define_$\ev=\sum_i
\ev_i\circ(\tau_i\tens\tau_i^*)$ and $\coev=\sum_i \coev_i$.
Our definition is_invariant under coactions of_$A$_as it should be, since the_projections $\tau_i$ commute with the coaction_of $A$. In particular, it_is_invariant_under braidings.
Cosemisimplicity and Peter-Weyl De |
\pi)^3}\left[\ln \Big({2 \pi [\lambda_1(\bm k, u_0)\lambda_2(\bm k, u_0)]^{1/2}\over \nu}\Big)
-\ln \Gamma\Big(1 + {\lambda_1(\bm k, u_0)\over \nu}\Big)-\ln \Gamma\Big(1 + {\lambda_2(\bm k, u_0)\over \nu}\Big)
\right],
\label{FreeEnergy_DissipativeT}\end{aligned}$$
where $\nu = 2 \pi T/\hbar$ and $\{\lambda_i(\bm k, u_0)\}$ are the roots of the equation $\lambda ^2(\bm k, u_0) - \gamma \lambda(\bm k, u_0) + \omega_c^2(\bm k, u_0) =0$, satisfying the relations
$$\begin{gathered}
\lambda_1 (\bm k, u_0)+ \lambda_2 (\bm k, u_0)= \gamma, \\
\lambda_1 (\bm k, u_0)\lambda_2(\bm k, u_0) = \omega_c ^2(\bm k, u_0).
\label{}\end{gathered}$$
It is worth mentioning that the first term $\widetilde F_0$ in Eq. now includes the contribution due to the reservoir. In principle, this contribution could exhibit a nontrivial behavior close to the transition point as a result of the coupling between the active atoms (which clearly “feel” the transition) and the degrees of freedom in the reservoir (which also “feel” the transition because of this coupling). Nevertheless, if these degrees of freedom forming the reservoir are low-frequency acoustic phonons, which would be quite natural, it can be shown that their contribution i) is not substantially modified because of damping [@Cano04a] and ii) is not the most important one close to the (low-$T$) phase-transition point as we have seen in Sec. \[S:Phonon-like dynamics\] [@Cano04b].
The above formulas allow us to compute the relevant magnitudes close to the phase transition by following the same procedure as before (see Ref. [@Cano04b]). The coefficient at the quadratic term in $u_0$ in the free energy, for instance, | \pi)^3}\left[\ln \Big({2 \pi [ \lambda_1(\bm k, u_0)\lambda_2(\bm k, u_0)]^{1/2}\over \nu}\Big)
-\ln \Gamma\Big(1 + { \lambda_1(\bm k, u_0)\over \nu}\Big)-\ln \Gamma\Big(1 + { \lambda_2(\bm k, u_0)\over \nu}\Big)
\right ],
\label{FreeEnergy_DissipativeT}\end{aligned}$$
where $ \nu = 2 \pi T/\hbar$ and $ \{\lambda_i(\bm k, u_0)\}$ are the roots of the equality $ \lambda ^2(\bm k, u_0) - \gamma \lambda(\bm k, u_0) + \omega_c^2(\bm k, u_0) = 0 $, meet the relations
$ $ \begin{gathered }
\lambda_1 (\bm k, u_0)+ \lambda_2 (\bm k, u_0)= \gamma, \\
\lambda_1 (\bm k, u_0)\lambda_2(\bm k, u_0) = \omega_c ^2(\bm k, u_0).
\label{}\end{gathered}$$
It is worth mention that the beginning term $ \widetilde F_0 $ in Eq. now admit the contribution due to the reservoir. In rationale, this contribution could exhibit a nontrivial behavior close to the transition point as a resultant role of the coupling between the active atoms (which intelligibly “ feel ” the transition) and the degrees of freedom in the reservoir (which also “ find ” the transition because of this coupling). Nevertheless, if these degree of freedom forming the reservoir are low - frequency acoustic phonons, which would be quite natural, it can be show that their contribution i) is not substantially modified because of damping [ @Cano04a ] and ii) is not the most authoritative one close to the (low-$T$) phase - transition point as we have seen in Sec. \[S: Phonon - like dynamics\ ] [ @Cano04b ].
The above formulas allow us to compute the relevant magnitudes close to the phase transition by following the same procedure as before (see Ref. [ @Cano04b ]). The coefficient at the quadratic term in $ u_0 $ in the free department of energy, for case, | \pi)^3}\lfft[\ln \Big({2 \pi [\lambda_1(\bm k, m_0)\lambda_2(\bm k, u_0)]^{1/2}\ovet \bu}\Big)
-\li \Gamma\Gig(1 + {\lamcda_1(\bm k, u_0)\over \nu}\Big)-\ln \Gamma\Bmg(1 + {\oambdq_2(\bm k, u_0)\over \nu}\Big)
\right],
\uabel{FreeVnergy_DiswipauiveT}\end{aligned}$$
whxde $\nu = 2 \pi T/\hgwr$ aud $\{\lambda_i(\bm k, u_0)\}$ are the rmots of the eqgagiln $\lambda ^2(\bm k, u_0) - \gamma \lambda(\bm k, t_0) + \omegs_c^2(\hm k, u_0) =0$, satisfiing uhe relzniins
$$\begin{gathered}
\lambda_1 (\bm k, u_0)+ \lambda_2 (\bm k, u_0)= \gamma, \\
\lambda_1 (\bm k, u_0)\lambda_2(\bm k, u_0) = \omfga_c ^2(\bm k, u_0).
\label{}\enf{gathered}$$
It is rirth mentionkng that the first terj $\widetilde F_0$ in Eq. now includer the contributuob dkg to the resxrvoir. In principlc, this wontribition could exmibit a bontrivial behavior cnose to the transijion point aa a result of the ciuplitg batwedb tfe zcvivs atomd (wiich clearlg “feel” the rransition) and the cedgres of freedkm in ehq reservoir (which also “feel” the transitpon gecause of this couplint). Nevertheless, if thede degreef of freedom forming the reservoir are low-frequenwy ackjstng oyojons, which would be quite natural, it can be srkwm nhat their contrinution i) is not sunshamjially modifiea becabae of damping [@Cano04a] wnd ii) ys nor the mosu impprtant one close to the (low-$R$) phase-transptiob point as we have seen in See. \[S:Phomon-lile dynamics\] [@Cano04b].
The abuve rormulas alpow us to zompute the relexann macnitudes close to the phasq transitmon bv followkng jhe samq procedurf as nafore (see Ref. [@Cano04h]). The cmefficient at the quadratic term in $u_0$ in vie free energi, fmr pnstance, | \pi)^3}\left[\ln \Big({2 \pi [\lambda_1(\bm k, u_0)\lambda_2(\bm k, -\ln + {\lambda_1(\bm u_0)\over \nu}\Big)-\ln \Gamma\Big(1 \right], where $\nu = \pi T/\hbar$ and k, u_0)\}$ are the roots of equation $\lambda ^2(\bm k, u_0) - \gamma \lambda(\bm k, u_0) + \omega_c^2(\bm k, =0$, satisfying the relations $$\begin{gathered} \lambda_1 (\bm k, u_0)+ \lambda_2 (\bm k, u_0)= \\ (\bm u_0)\lambda_2(\bm u_0) = \omega_c ^2(\bm k, u_0). \label{}\end{gathered}$$ It is worth mentioning that the first term $\widetilde in Eq. now includes the contribution due to reservoir. In principle, this could exhibit a nontrivial behavior to transition point a of coupling between the atoms (which clearly “feel” the transition) and the degrees of freedom in the reservoir (which also “feel” transition because coupling). Nevertheless, these of forming the reservoir acoustic phonons, which would be quite be shown that their contribution i) is not modified because damping [@Cano04a] and ii) is not most important one close to the (low-$T$) phase-transition as we have seen in Sec. \[S:Phonon-like dynamics\] [@Cano04b]. The above formulas allow us to relevant magnitudes close to phase transition by the procedure before Ref. [@Cano04b]). coefficient at the quadratic term in $u_0$ in the free energy, instance, | \pi)^3}\left[\ln \Big({2 \pi [\lambda_1(\bm k, u_0)\laMbda_2(\bm k, u_0)]^{1/2}\ovEr \nu}\BIg)
-\lN \GaMmA\Big(1 + {\LambDa_1(\bm k, u_0)\over \nu}\BiG)-\Ln \GaMma\Big(1 + {\lambda_2(\bm k, u_0)\over \nu}\big)
\riGhT],
\LabeL{frEeEneRgy_DissIPaTIVeT}\EnD{aLigNeD}$$
WhEre $\nu = 2 \Pi T/\Hbar$ and $\{\Lambda_i(\bm k, U_0)\}$ arE tHe roots of the EQuAtion $\lambdA ^2(\bm K, u_0) - \gamma \lambdA(\bm K, u_0) + \omegA_c^2(\Bm k, U_0) =0$, SatisFyiNg the RelatiONs
$$\begiN{gathered}
\LaMBda_1 (\bm k, U_0)+ \Lambda_2 (\bM K, U_0)= \gAmma, \\
\Lambda_1 (\bm k, u_0)\lambda_2(\bM K, u_0) = \OMega_c ^2(\bm k, u_0).
\label{}\End{gatHeREd}$$
iT Is wOrtH mentioninG tHat thE First teRM $\wIDETilDE F_0$ in Eq. now inclUdes the contRIbuTion duE tO thE ReservOir. In PrINciPle, this contRibuTion could ExhibiT A nontriVIal behaVior clOse To tHe trANsItIon PoINt aS A rEsuLT of The couplInG bEtweeN the ACTIVe atOms (WhicH cleaRly “feel” the traNsiTion) ANd tHe degRees oF freEdOm in tHe reseRvoir (WhIch also “feel” the tRansItion becaUse Of ThiS cOupliNG). NeverTheLesS, if thesE degreeS Of fReEDOM fOrming the reservoir ArE LOw-FrequencY acousTIc PhONons, whicH wOulD be qUITe natUral, IT cAn be showN that tHEiR cOntribuTiOn i) is nOt SubStaNtialLY modIfied bEcause of DampiNG [@Cano04a] and ii) is nOT the most imporTAnT ONe CLose To tHe (low-$T$) phase-TranSItioN poiNT aS we HAve seEn in SEc. \[s:phONon-like dynamics\] [@Cano04B].
THe abovE formUlas allow us to Compute the RELEvant magNituDEs CLose to the phase TransItion by folLOwing the Same pRocedure As before (sEE ref. [@Cano04b]). the CoeFfiCieNT At The quadratic tERM in $u_0$ In The free EneRgy, for iNstAncE, | \pi)^3}\left[\ln \Big({2 \ pi [\lambd a_1(\ bmk,u_ 0)\l ambd a_2(\bm k, u_0 ) ]^{1 /2}\over \nu}\Big)
-\l n \Ga mm a \Big ( 1+ {\l ambda_1 ( \b m k,u_ 0) \ov er \n u}\Bi g)- \ln \Ga mma\Big(1+ { \l ambda_2(\bmk ,u_0)\over\nu }\Big)
\righ t],
\labe l{ Fre e Energ y_D issip ativeT } \end{a ligned}$$
w here $ \ nu = 2\ p iT/\h bar$ and $\{\lamb d a_ i (\bm k, u_0)\} $ areth e r o o tsofthe equati on $\la m bda ^2( \ bm k , u_ 0 ) - \gamma \l ambda(\bm k , u_ 0) + \ om ega _ c^2(\b m k,u_ 0 ) = 0$, satisfy ingthe relat ions
$ $\begin { gathere d}
\la mbd a_1 (\b m k ,u_0 )+ \la m bd a_2 (\b m k, u_0 )= \ gamma , \\ \ l a mbda _1(\bm k, u _0)\lambda_2( \bm k,u _0) = \o mega_ c ^2 (\ bm k, u_0).
\lab el {}\end{gathered }$$
It is wo rth m ent io ningt hat th e f irs t term$\widet i lde F _ 0 $ i n Eq. now includes t h e c ontribut ion du e t ot he reser vo ir. Inp r incip le,t hi s contri bution co ul d exhib it a non tr ivi albehav i or c lose t o the tr ansit i on point as ar esult of thec ou p l in g bet wee n the activ e at o ms ( whic h c lea r ly “f eel”th e t r ansition) and the d eg rees o f fre edom in the r eservoir ( w h i ch also“fee l ”t he transitionbecau se of this coupling ). Ne verthele ss, if th e s e degree s o f f ree dom f or ming the rese r v oirar e low-f req uency a cou sti c p hon on s, whichwould be q ui te n atu ral,i t can be s how ntha t the i r cont ribut ioni) i s no t subst a nt i a llymo di fied be ca use o f da m pin g [@Can o04a] and ii ) isno tthe mos t important o ne close toth e ( low-$T $ ) phase-t ransition point as we h a ve seen in Sec. \[S :Phonon-l ike dynam ics \ ] [@Ca no04b] .
Th eabo v e form u l as al lo w us to co m p ute there leva nt magn itudes close to th e ph ase transitio n b y fo l l ow ing th e sa me pro c e dure as before(see Ref.[@ C an o04b]). Th e co ef ficient at the quad r atic te rm in $u_ 0$ in the f reee n erg y, for ins tance, | \pi)^3}\left[\ln \Big({2_\pi [\lambda_1(\bm_k, u_0)\lambda_2(\bm k, u_0)]^{1/2}\over_\nu}\Big)
-\ln \Gamma\Big(1_+_{\lambda_1(\bm k,_u_0)\over_\nu}\Big)-\ln \Gamma\Big(1 +_{\lambda_2(\bm k, u_0)\over_\nu}\Big)
\right],
\label{FreeEnergy_DissipativeT}\end{aligned}$$
where $\nu = 2_\pi T/\hbar$ and_$\{\lambda_i(\bm_k, u_0)\}$ are the roots of the equation $\lambda ^2(\bm k, u_0) - \gamma_\lambda(\bm_k, u_0)_+_\omega_c^2(\bm_k, u_0) =0$, satisfying the_relations
$$\begin{gathered}
\lambda_1 (\bm k, u_0)+ \lambda_2_(\bm k,_u_0)= \gamma, \\
\lambda_1 (\bm k, u_0)\lambda_2(\bm k, u_0)_=_\omega_c ^2(\bm k,_u_0).
\label{}\end{gathered}$$
It is worth mentioning that the first term $\widetilde_F_0$ in Eq. now includes the_contribution due to_the_reservoir._In principle, this contribution_could exhibit a nontrivial behavior close_to the transition point as a_result of the coupling between the active_atoms (which clearly “feel” the transition)_and the degrees of freedom_in the_reservoir (which also “feel” the_transition because of_this coupling)._Nevertheless, if these_degrees of freedom forming the reservoir_are low-frequency acoustic_phonons, which would be quite natural,_it_can be shown_that_their_contribution i)_is not substantially_modified_because of_damping_[@Cano04a] and ii) is not the_most_important one close to the (low-$T$) phase-transition_point as we have_seen_in Sec. \[S:Phonon-like dynamics\]_[@Cano04b].
The above formulas allow us_to compute the relevant magnitudes close_to the_phase transition_by following the same procedure as before (see Ref. [@Cano04b]). The_coefficient at the quadratic term in_$u_0$ in the free_energy, for_instance, |
Theorem 6.4] in two ways. Firstly and primarily, it improves the $(n+1)$ in her result to what should be the optimal result $(n.5)$ (although only for $E$-links). Being an $E$-link ensures an extra rank condition that is hidden in Harvey’s proof since it is implied by $(n+1)$-solvability. Our Theorem then allows for a corresponding sharpening of Harvey’s main application of the above [@Ha2 Theorem 6.8].
\[infgenerated\] In the category of m-component ordered oriented string links ($m>1$), each of the quotients $\mathcal{F}_{(n)}/\mathcal{F}^{\mathbb{Q}}_{(n.5)}$ contains a subgroup, consisting entirely of boundary links, whose abelianization has infinite $\mathbb{Q}$-rank.
We remark that Harvey’s additivity result for her $\rho_k$ for boundary string links should hold for any additively closed subset of $E$-links, such as the set of homology boundary links with a fixed “*pattern*” and so the word “boundary links” in the above theorem should be able to be replaced by any such set.
Secondly, our Theorem \[n.5solvable\] improves on Harvey’s version by proving the theorem for so-called “rational” solvability. The definition of the latter is reviewed below. This may seem like a technical advance. However it points the way to certain further improvements in other results in the literature that we will postpone to another paper. Namely, there is a well-studied geometric notion that approximates $n$-solvability of links that has to with generalizing the annuli in the definition of concordance to certain $2$-complexes called *symmetric gropes*. We shall not review these terms here, but in future paper we will show that the torsion-free derived series suggests beautiful generalizations of gropes, that we call *rational homology gropes* and these generalizations are the proper geometric approximation to the algebraic notion of rational n-solvability.
We briefly review the definitions of the Cheeger-Gromov von Neumann $\rho-$invariant (only in the cases that we need here) and $(n)-$solvability. More general definitions are to be found in [@ChGr1] [@Ha2 Section 3] [@COT Section 5].
\[ | Theorem 6.4 ] in two ways. Firstly and primarily, it improves the $ (n+1)$ in her resultant role to what should be the optimum result $ (n.5)$ (although only for $ E$-links). Being an $ E$-link guarantee an excess rank condition that is hidden in Harvey ’s validation since it is incriminate by $ (n+1)$-solvability. Our Theorem then allows for a comparable sharpening of Harvey ’s chief application of the above [ @Ha2 Theorem 6.8 ].
\[infgenerated\ ] In the class of m - component ordered oriented chain links ($ m>1 $), each of the quotients $ \mathcal{F}_{(n)}/\mathcal{F}^{\mathbb{Q}}_{(n.5)}$ contains a subgroup, consisting entirely of boundary link, whose abelianization has infinite $ \mathbb{Q}$-rank.
We remark that Harvey ’s additivity result for her $ \rho_k$ for boundary string connection should hold for any additively closed subset of $ E$-links, such as the set of homology boundary links with a repair “ * pattern * ” and so the word “ boundary links ” in the above theorem should be able to be replaced by any such hardening.
Secondly, our Theorem \[n.5solvable\ ] improves on Harvey ’s version by proving the theorem for so - called “ rational ” solvability. The definition of the latter is reviewed below. This may seem like a technical advance. However it points the way to certain further improvements in other results in the literature that we will postpone to another composition. Namely, there is a well - studied geometric impression that approximate $ n$-solvability of radio link that has to with generalizing the ring in the definition of concordance to certain $ 2$-complexes called * symmetric grope *. We shall not review these terms here, but in future paper we will show that the torsion - free derived series suggests beautiful generalizations of gropes, that we call * rational homology grope * and these generalizations are the proper geometric estimate to the algebraic notion of rational n - solvability.
We briefly review the definitions of the Cheeger - Gromov von Neumann $ \rho-$invariant (only in the subject that we need here) and $ (n)-$solvability. More general definitions are to be find oneself in [ @ChGr1 ] [ @Ha2 department 3 ] [ @COT Section 5 ].
\ [ | Thforem 6.4] in two ways. Firstuy and primarili, ut impcoves tge $(n+1)$ in fer result to what should be tye opupmal result $(n.5)$ (althougf only fog $E$-links). Veinj an $E$-link ensurxa an exbxa rahn coudmtion that is hldden in Hasvey’s proof sitcd nt is implied by $(n+1)$-solvability. Our Thqorem tnej allows for a corgefponspnn sharpening of Harvey’s main appmicatioi of the above [@Na2 Theorem 6.8].
\[infgenerated\] In the category of m-complnent ordergs owuented strine links ($m>1$), taeh of the qootients $\mathcal{F}_{(n)}/\mathcal{F}^{\mathbb{D}}_{(n.5)}$ coutains a suvgeouo, consisting entigely of boundary linkv, whose abelianizatiok has indinite $\mathbb{Q}$-rank.
We cemark that Harvey’s wdditivitf xesult for her $\rho_k$ fir boungary strkbg uinis sgould jolv for any asditively coosed subset of $E$-limkf, such as the aet of hjmology boundary links with a fixed “*patuern*” znd so the word “boundart links” in the above jheorem shjuld be able to be replaced by any such set.
Secondny, ouc Gheircm \[n.5suovwble\] improves on Harvey’s version by proving trs uheprem for so-caljed “rationak” dokdability. The aefinicjoh of the latter is reviewgd beliw. This mwy srem like a technical advancw. However it points the way to eertain furtker imkrovemrnts in other results iu the miterature hhat we wjul postpone to avotmer paper. Namely, there is a wqll-studiev geoketric votipn thae approximwtes $k$-volvability of linns thct hav to with heneralizing the annuli in the vxfinition of votcogdance to certsin $2$-complexes called *symmettic gropef*. We rhall not geview thxse terms hewe, but in futgte paper we wmll show ehat the torsiov-wree derived srries sugytsts beautuful generalizatioks of fropes, that we eclo *rational homokogh gwoies* agg these genesalixatkpns afe the kxopcr eeomrtric approximation do tge algebraic notiom jf ratiobal n-soldability.
We broefly review the dtfinitmons oh the Vhegger-Gromov von Neumann $\rho-$invarjant (only in the cases trat qe need here) and $(n)-$solvability. More general definitiins are to be foune in [@ChGr1] [@Ha2 Section 3] [@GOT Section 5].
\[ | Theorem 6.4] in two ways. Firstly and improves $(n+1)$ in result to what $(n.5)$ only for $E$-links). an $E$-link ensures extra rank condition that is hidden Harvey’s proof since it is implied by $(n+1)$-solvability. Our Theorem then allows for corresponding sharpening of Harvey’s main application of the above [@Ha2 Theorem 6.8]. \[infgenerated\] the of ordered string links ($m>1$), each of the quotients $\mathcal{F}_{(n)}/\mathcal{F}^{\mathbb{Q}}_{(n.5)}$ contains a subgroup, consisting entirely of boundary links, abelianization has infinite $\mathbb{Q}$-rank. We remark that Harvey’s result for her $\rho_k$ boundary string links should hold any closed subset $E$-links, as set of homology links with a fixed “*pattern*” and so the word “boundary links” in the above theorem should be to be any such Secondly, Theorem improves on Harvey’s proving the theorem for so-called “rational” of the latter is reviewed below. This may like a advance. However it points the way certain further improvements in other results in the that we will postpone to another paper. Namely, there is a well-studied geometric notion that of links that has with generalizing the in definition concordance certain $2$-complexes *symmetric gropes*. We shall not review these terms here, but in paper we will show that the torsion-free derived series suggests of that we call homology gropes* and these are proper geometric approximation to notion rational review definitions the Cheeger-Gromov von Neumann (only in the cases that need here) and $(n)-$solvability. be found in [@ChGr1] [@Ha2 Section 3] [@COT 5]. \[ | Theorem 6.4] in two ways. Firstly anD primarily, It impRovEs tHe $(N+1)$ in hEr reSult to what shouLD be tHe optimal result $(n.5)$ (althouGh onlY fOR $E$-liNKs). being An $E$-link ENsUREs aN eXtRa rAnK CoNditiOn tHat is hiDden in HarvEy’s PrOof since it is IMpLied by $(n+1)$-solVabIlity. Our TheoRem Then alLoWs fOR a corResPondiNg sharPEning oF Harvey’s mAiN ApplicATion of tHE AbOve [@HA2 Theorem 6.8].
\[infgeneraTEd\] iN the category of M-compoNeNT oRDEreD orIented striNg Links ($M>1$), Each of tHE qUOTIenTS $\mathcal{F}_{(n)}/\matHcal{F}^{\mathbb{q}}_{(N.5)}$ coNtains A sUbgROup, conSistiNg ENtiRely of boundAry lInks, whose AbeliaNIzation HAs infinIte $\matHbb{q}$-raNk.
We REmArK thAt hArvEY’s AddITivIty resulT fOr Her $\rhO_k$ foR BOUNdarY stRing Links Should hold for Any AddiTIveLy cloSed suBset Of $e$-linkS, such aS the sEt Of homology boundAry lInks with a FixEd “*PatTeRn*” and SO the woRd “bOunDary linKs” in the ABovE tHEOReM should be able to be rEpLACeD by any suCh set.
SECoNdLY, our TheoReM \[n.5sOlvaBLE\] imprOves ON HArvey’s veRsion bY PrOvIng the tHeOrem foR sO-caLleD “ratiONal” sOlvabiLity. The dEfiniTIon of the latter IS reviewed beloW. thIS MaY Seem LikE a technical AdvaNCe. HoWeveR It PoiNTs the Way to CeRTaIN further improvementS iN other ResulTs in the literaTure that we WILL postponE to aNOtHEr paper. Namely, tHere iS a well-studIEd geometRic noTion that ApproximaTES $n$-solvabIliTy oF liNks THAt Has to with geneRALiziNg The annuLi iN the defIniTioN of ConCoRdance to cErtain $2$-coMpLeXeS cAllEd *symMEtric groPeS*. We ShAll Not reVIew theSe terMs heRe, BuT In fUture paPEr WE Will ShOw That The ToRsion-Free DEriVed seriEs suggestS beAUtifUl GeNeralizAtions of gropeS, tHat we call *rAtIonAl homoLOGy gropes* And these generalizations ARe the prOpeR geomEtriC approximAtiOn to thE alGEbraic Notion Of ratIoNal N-SOlvabILItY.
We BrIefly revieW THe dEfiniTiOns oF the CheEger-Gromov von NeumaNN $\rhO-$invariant (onlY in The cASEs ThaT We NEed HeRE) anD $(N)-$Solvability. More General defInITiOns are to be FOunD iN [@ChGr1] [@Ha2 section 3] [@cOT SeCTion 5].
\[ | Theorem 6.4] in two ways. Firstly a nd pr ima ril y, itimpr oves the $(n+1 ) $ in her result to what sh ouldbe theo pt imalresult$ (n . 5 )$(a lt hou gh on ly fo r $ E$-link s). Beingan$E $-link ensur e san extra r ank condition t hat is hi dd eni n Har vey ’s pr oof si n ce itis implie db y $(n+ 1 )$-solv a b il ity. Our Theorem then al l ows for a corr espond in g s h a rpe nin g of Harve y’ s mai n applic a ti o n oft he above [@H a2 Theorem6 .8] .
\[i nf gen e rated\ ] Inth e ca tegory of m -com ponent or deredo riented stringlinks($m >1$ ), e a ch o f t he quo t ie nts $\m athcal{F }_ {( n)}/\ math c a l { F}^{ \ma thbb {Q}}_ {(n.5)}$ cont ain s as ubg roup, cons isti ng enti rely o f bou nd ary links, whos e ab elianizat ion h asin finit e $\mat hbb {Q} $-rank.
We re m ark t h a t H arvey’s additivity r e s ul t for he r $\rh o _k $f or bound ar y s trin g links sho u ld hold fo r anya dd it ively c lo sed su bs etof$E$-l i nks, suchas the s et of homology bound a ry links with af i xe d “*p att ern*” and s o th e wor d “b o un dar y link s” in t h ea bove theorem should b e able to b e replaced by any suchs e t .
Secon dly, ou r Theorem \[n.5 solva ble\] impr o ves on H arvey ’s versi on by pro v i ng the t heo rem fo r s o - ca lled “rationa l ” sol va bility. Th e defin iti onofthe l atter isreviewed b el ow .Thi s may seem lik ea t ec hni cal a d vance. Howe verit p o int s the w a yt o cer ta in fur the rimpro veme n tsin othe r results in theli te raturethat we willpo stpone toan oth er pap e r . Namely , there is a well-studi e d geome tri c not ionthat appr oxi mates$n$ - solvab ilityof li nk s t h a t has t owit hgeneralizi n g th e ann ul i in the de finition of concor d anc e to certain$2$ -com p l ex esc al l ed*s y mme t r ic gropes*. Weshall notre v ie w these te r mshe re, but in fut ure p a per wewill show that the t orsi o n -fr ee derived seriessuggestsb eauti f ul gene ral izatio ns of grop es, th a t w e cal l *rat io nal ho molog ygropes*and these generalizatio ns are thepro per geome tri c ap proximati on t o the alge bra icnotio n o f rati onal n- sol v abili ty.We briefl y r evi e w t he definiti o n s of theChe e ger-Gr omov von Neumann $\rh o -$invariant (o nlyi n th e c a sesth at we need her e)an d $(n)-$so lv ability. Mo re gener al defin itions are t o be fo u n di n [@C hGr1 ] [ @Ha2 Sect ion 3 ] [@COTSe ct i on 5].
\[ | Theorem_6.4] in_two ways. Firstly and_primarily, it_improves_the $(n+1)$_in_her result to_what should be_the optimal result $(n.5)$_(although only for_$E$-links)._Being an $E$-link ensures an extra rank condition that is hidden in Harvey’s proof_since_it is_implied_by_$(n+1)$-solvability. Our Theorem then allows_for a corresponding sharpening of_Harvey’s main_application of the above [@Ha2 Theorem 6.8].
\[infgenerated\] In_the_category of m-component_ordered oriented string links ($m>1$), each of the quotients_$\mathcal{F}_{(n)}/\mathcal{F}^{\mathbb{Q}}_{(n.5)}$ contains a subgroup, consisting entirely_of boundary links,_whose_abelianization_has infinite $\mathbb{Q}$-rank.
We remark_that Harvey’s additivity result for her_$\rho_k$ for boundary string links should_hold for any additively closed subset of_$E$-links, such as the set of_homology boundary links with a_fixed “*pattern*”_and so the word “boundary_links” in the_above theorem_should be able_to be replaced by any such_set.
Secondly, our Theorem \[n.5solvable\]_improves on Harvey’s version by proving_the_theorem for so-called_“rational”_solvability._The definition_of the latter_is_reviewed below._This_may seem like a technical advance._However_it points the way to certain further_improvements in other results_in_the literature that we_will postpone to another paper._Namely, there is a well-studied geometric_notion that_approximates $n$-solvability_of links that has to with generalizing the annuli in the_definition of concordance to certain $2$-complexes_called *symmetric gropes*. We_shall not_review_these terms here,_but_in future_paper we will show that the torsion-free_derived series_suggests beautiful generalizations of gropes, that_we call *rational homology_gropes*_and these generalizations are the proper_geometric approximation to the algebraic notion_of rational n-solvability.
We briefly review_the_definitions_of the Cheeger-Gromov von Neumann_$\rho-$invariant (only in the cases that_we need here)_and $(n)-$solvability. More general definitions are to_be_found in [@ChGr1] [@Ha2 Section 3] [@COT Section_5].
\[ |
phi}}}}$ is between $\frac{s}{c_0 2^i}$ and $\frac{c_0 s}{ 2^i}$. Refer to these cells formed at the $i$-th level as $L_i$.
\[packingTheTree\] Given a ball $B$ of radius $r$ under $\sqrt{{\ensuremath{D_{s\phi}}}}$, let $i = \log \frac{ s}{c_0 r}$. Then $|L_i \cap B| \leq O(2^d)$ and the side length of each cell in $L_i$ is between $r$ and ${c_0}^2 r$ under $\sqrt{{\ensuremath{D_{s\phi}}}}$. We can also explicitly retrieve the quadtree cells corresponding to $|L_i \cap B|$ in $O(2^d \log n)$ time.
Note that for cells in $L_i$, we have side lengths under $\sqrt{{\ensuremath{D_{s\phi}}}}$ between $\frac{s}{c_0 2^i}$ and $\frac{c_0 s}{2^i}$ by Corollary \[repbis\]. Substituting $i = \log \frac{s}{c_0 r}$, these cells have side length between $r$ and ${c_0}^2 r$ under $\sqrt{{\ensuremath{D_{s\phi}}}}$. By the reverse triangle inequality and a similar argument to Lemma \[ballcover\], we get our required bound for $|L_i \cap B|$. In preconstruction of our quadtree $T$ we maintain for each dimension the corresponding interval quadtree $T_k$, $\forall k\in [1..d]$. Observe this incurs at most $O(n)$ storage, with $d$ in the big-Oh. For retrieving the actual cells $|L_i \cap B|$, we first find the $O(1)$ intervals from level $i$ in each $T_k$ that may intersect $B$. Taking a product of these, we get $O(2^d)$ cells which are a superset of the canonical cells $L_i \subset T$. Each cell may be looked up in $O(\log n)$ time from the compressed quadtree [@snotes] so our overall retrieval time is $O (2^d \log n)$.
Given query point $q$, we first obtain in $O(\log n)$ | phi}}}}$ is between $ \frac{s}{c_0 2^i}$ and $ \frac{c_0 s } { 2^i}$. Refer to these cells formed at the $ i$-th level as $ L_i$.
\[packingTheTree\ ] Given a musket ball $ B$ of spoke $ r$ under $ \sqrt{{\ensuremath{D_{s\phi}}}}$, let $ i = \log \frac { s}{c_0 r}$. Then $ |L_i \cap B| \leq O(2^d)$ and the side duration of each cellular telephone in $ L_i$ is between $ r$ and $ { c_0}^2 r$ under $ \sqrt{{\ensuremath{D_{s\phi}}}}$. We can also explicitly retrieve the quadtree cells match to $ |L_i \cap B|$ in $ O(2^d \log n)$ time.
Note that for cell in $ L_i$, we have side distance under $ \sqrt{{\ensuremath{D_{s\phi}}}}$ between $ \frac{s}{c_0 2^i}$ and $ \frac{c_0 s}{2^i}$ by Corollary \[repbis\ ]. Substituting $ i = \log \frac{s}{c_0 r}$, these cells get side length between $ r$ and $ { c_0}^2 r$ under $ \sqrt{{\ensuremath{D_{s\phi}}}}$. By the reverse triangle inequality and a similar argument to Lemma \[ballcover\ ], we catch our required bound for $ |L_i \cap B|$. In preconstruction of our quadtree $ T$ we assert for each dimension the corresponding interval quadtree $ T_k$, $ \forall k\in [ 1.. d]$. Observe this incurs at about $ O(n)$ storage, with $ d$ in the adult - Oh. For retrieving the actual cells $ |L_i \cap B|$, we first determine the $ O(1)$ intervals from level $ i$ in each $ T_k$ that may intersect $ B$. Taking a product of these, we get $ O(2^d)$ cell which are a superset of the canonical cells $ L_i \subset T$. Each cell may be looked up in $ O(\log n)$ time from the compressed quadtree [ @snotes ] so our overall retrieval time is $ O (2^d \log n)$.
Given question detail $ q$, we first prevail in $ O(\log n)$ | phi}}}}$ is between $\frac{s}{c_0 2^i}$ and $\frac{c_0 s}{ 2^i}$. Refet ro theve celms formea at the $i$-th level as $L_i$.
\[packmngTyeTret\] Given a ball $B$ of radius $r$ under $\swrt{{\eisuremath{D_{s\phi}}}}$, lxf $i = \lon \frad{ s}{c_0 x}$. Vhen $|L_i \cap B| \lgq O(2^d)$ and tha side length mf ecch cell in $L_i$ is between $r$ and ${c_0}^2 r$ tnder $\sarh{{\ensuremath{D_{s\pri}}}}$. Wt cwn amso explicitly retrieve the quadtrse cellv correspondimg to $|L_i \cap B|$ in $O(2^d \log n)$ timf.
Note that for celps in $L_i$, we havq side lengthr under $\sqgc{{\ensuremath{S_{s\phi}}}}$ between $\frac{s}{c_0 2^i}$ and $\frac{z_0 s}{2^i}$ yy Corollari \[deotis\]. Substitnting $p = \log \frac{s}{c_0 r}$, these cells nave side lengbh bevweeb $r$ and ${c_0}^2 r$ under $\sqrv{{\ensuremath{D_{s\phi}}}}$. By jhe reversa criangle inequality abd a sikilas areymevt uo Kejma \[baplckver\], we gef our requieed bound for $|L_i \cak B|$. Pm preconstrudtion jf our quadtree $T$ we maintain for each dikenaion the corresponding unterval quadtree $T_k$, $\vorall k\ig [1..d]$. Observe this incurs at most $O(n)$ storage, with $d$ in tie biy-Ih. Fof rftrieving the actual cells $|L_i \cap B|$, we first sjnc nhe $O(1)$ intervals fvom level $i$ in eacn $H_k$ jhat may interrect $B$. Taiing a product of hhese, wg get $I(2^d)$ cells rhicn are a superset of the caninical cells $O_i \subset T$. Each czll may be luokec up on $O(\log n)$ time from the comlressed quaftree [@snofds] so our overalu rvtriaval timt is $O (2^d \log n)$.
Giveg query piint $q$, we fkrst obtaig in $O(\log j)$ | phi}}}}$ is between $\frac{s}{c_0 2^i}$ and $\frac{c_0 Refer these cells at the $i$-th a $B$ of radius under $\sqrt{{\ensuremath{D_{s\phi}}}}$, let = \log \frac{ s}{c_0 r}$. Then \cap B| \leq O(2^d)$ and the side length of each cell in $L_i$ between $r$ and ${c_0}^2 r$ under $\sqrt{{\ensuremath{D_{s\phi}}}}$. We can also explicitly retrieve the cells to \cap in $O(2^d \log n)$ time. Note that for cells in $L_i$, we have side lengths under between $\frac{s}{c_0 2^i}$ and $\frac{c_0 s}{2^i}$ by Corollary Substituting $i = \log r}$, these cells have side between and ${c_0}^2 under By reverse triangle inequality a similar argument to Lemma \[ballcover\], we get our required bound for $|L_i \cap B|$. In preconstruction our quadtree maintain for dimension corresponding quadtree $T_k$, $\forall Observe this incurs at most $O(n)$ in the big-Oh. For retrieving the actual cells \cap B|$, first find the $O(1)$ intervals from $i$ in each $T_k$ that may intersect $B$. a product of these, we get $O(2^d)$ cells which are a superset of the canonical \subset T$. Each cell be looked up $O(\log time the quadtree [@snotes] our overall retrieval time is $O (2^d \log n)$. Given query $q$, we first obtain in $O(\log n)$ | phi}}}}$ is between $\frac{s}{c_0 2^i}$ and $\fraC{c_0 s}{ 2^i}$. Refer tO thesE ceLls FoRmed At thE $i$-th level as $L_i$.
\[pACkinGTheTree\] Given a ball $B$ of rAdius $R$ uNDer $\sQRt{{\EnsurEmath{D_{s\PHi}}}}$, LET $i = \lOg \FrAc{ s}{C_0 r}$. tHeN $|L_i \caP B| \lEq O(2^d)$ and The side lenGth Of Each cell in $L_i$ IS bEtween $r$ and ${C_0}^2 r$ uNder $\sqrt{{\ensuRemAth{D_{s\pHi}}}}$. we cAN also ExpLicitLy retrIEve the Quadtree cElLS correSPonding TO $|l_i \Cap B|$ In $O(2^d \log n)$ time.
Note tHAt FOr cells in $L_i$, we hAve sidE lENgTHS unDer $\Sqrt{{\ensureMaTh{D_{s\pHI}}}}$ betweeN $\FrAC{S}{C_0 2^i}$ aND $\frac{c_0 s}{2^i}$ by CorOllary \[repbiS\]. subStitutInG $i = \lOG \frac{s}{C_0 r}$, theSe CEllS have side leNgth Between $r$ aNd ${c_0}^2 r$ unDEr $\sqrt{{\eNSurematH{D_{s\phi}}}}$. by tHe rEverSE tRiAngLe INeqUAlIty ANd a Similar aRgUmEnt to lemmA \[BALLcovEr\], wE get Our reQuired bound foR $|L_i \Cap B|$. iN prEconsTructIon oF oUr quaDtree $T$ We maiNtAin for each dimenSion The corresPonDiNg iNtErval QUadtreE $T_k$, $\ForAll k\in [1..d]$. observe THis InCURS aT most $O(n)$ storage, with $D$ iN THe Big-Oh. For RetrieVInG tHE actual cElLs $|L_I \cap b|$, WE firsT finD ThE $O(1)$ intervAls froM LeVeL $i$ in eacH $T_K$ that mAy IntErsEct $B$. TAKing A produCt of thesE, we geT $o(2^d)$ cells which arE A superset of thE CaNONiCAl ceLls $l_i \subset T$. EaCh ceLL may Be loOKeD up IN $O(\log N)$ time FrOM tHE compressed quadtree [@SnOtes] so Our ovErall retrievaL time is $O (2^d \lOG N)$.
given queRy poINt $Q$, We first obtain iN $O(\log N)$ | phi}}}}$ is between $\frac {s}{c_0 2^ i}$ a nd$\f ra c{c_ 0 s} { 2^i}$. Refer to t hese cells formed at t he $i $- t h le v el as $ L_i$.
\ [p a c kin gT he Tre e\ ] G ivena b all $B$ of radius $r $under $\sqrt { {\ ensuremath {D_ {s\phi}}}}$, le t $i = \ log \frac { s }{c_0 r}$.T hen $| L_i \capB| \leq O ( 2^d)$ a n d t he s ide length of eac h c e ll in $L_i$ is betwe en $r $ and ${ c_0}^2 r$un der $ \ sqrt{{\ e ns u r e mat h {D_{s\phi}}}} $. We can a l soexplic it lyr etriev e the q u adt ree cells c orre spondingto $|L _ i \capB |$ in $ O(2^d\lo g n )$ t i me .
No te tha t f orc ell s in $L_ i$ ,we ha ve s i d e leng ths und er $\ sqrt{{\ensure mat h{D_ { s\p hi}}} }$ be twee n$\fra c{s}{c _0 2^ i} $ and $\frac{c_ 0 s} {2^i}$ by Co ro lla ry \[re p bis\]. Su bst ituting $i = \ l og\f r a c {s }{c_0 r}$, these c el l s h ave side lengt h b et w een $r$an d $ {c_0 } ^ 2 r$unde r $ \sqrt{{\ ensure m at h{ D_{s\ph i} }}}$.By th e r evers e tri angleinequali ty an d a similar arg u ment to Lemma \[ b a ll c over \], we get our req u ired bou n dfor $|L_i \cap B | $. In preconstructionof our q uadtr ee $T$ we mai ntain fore a c h dimens iont he correspondinginter val quadtr e e $T_k$, $\fo rall k\i n [1..d]$ . Observethi s i ncu rsa t m ost $O(n)$ st o r age, w ith $d$ in the bi g-O h.For re tr ieving th e actual c el ls $ |L_ i \ca p B|$, we f irs tfin d the $O(1)$ inte rval sfr o m l evel $i $ i n each $ T_ k$ t hat m ay in ters e ct$B$. Ta king a pr odu c t of t he se, weget $O(2^d)$ce lls whichar e a super s e t of the canonical cells $L_i \ s ubset T $.Eachcell may be l ook ed upin$ O(\log n)$ t ime f ro m t h e comp r e ss edqu adtree [@s n o tes ] soou r ov erall r etrieval time is $ O (2 ^d \log n)$.
Gi venq u er y p o in t $q $, wef i rst obtain in $ O(\log n)$ | phi}}}}$ is_between $\frac{s}{c_0_2^i}$ and $\frac{c_0 s}{_2^i}$. Refer_to_these cells_formed_at the $i$-th_level as $L_i$.
\[packingTheTree\]_Given a ball $B$_of radius $r$_under_$\sqrt{{\ensuremath{D_{s\phi}}}}$, let $i = \log \frac{ s}{c_0 r}$. Then $|L_i \cap B| \leq O(2^d)$_and_the side_length_of_each cell in $L_i$ is_between $r$ and ${c_0}^2 r$_under $\sqrt{{\ensuremath{D_{s\phi}}}}$._We can also explicitly retrieve the quadtree cells_corresponding_to $|L_i \cap_B|$ in $O(2^d \log n)$ time.
Note that for cells_in $L_i$, we have side lengths_under $\sqrt{{\ensuremath{D_{s\phi}}}}$ between_$\frac{s}{c_0_2^i}$_and $\frac{c_0 s}{2^i}$ by_Corollary \[repbis\]. Substituting $i = \log_\frac{s}{c_0 r}$, these cells have side_length between $r$ and ${c_0}^2 r$ under_$\sqrt{{\ensuremath{D_{s\phi}}}}$. By the reverse triangle inequality_and a similar argument to_Lemma \[ballcover\],_we get our required bound_for $|L_i \cap_B|$. In_preconstruction of our_quadtree $T$ we maintain for each_dimension the corresponding_interval quadtree $T_k$, $\forall k\in [1..d]$._Observe_this incurs at_most_$O(n)$_storage, with_$d$ in the_big-Oh._For retrieving_the_actual cells $|L_i \cap B|$, we_first_find the $O(1)$ intervals from level $i$_in each $T_k$ that_may_intersect $B$. Taking a_product of these, we get_$O(2^d)$ cells which are a superset_of the_canonical cells_$L_i \subset T$. Each cell may be looked up in $O(\log_n)$ time from the compressed quadtree_[@snotes] so our overall_retrieval time_is_$O (2^d \log_n)$.
Given_query point_$q$, we first obtain in $O(\log n)$ |
t)$ is a $t$-homotetic cube in $Q$ with the same center of $Q$, that is, $Q_n(t)= x(Q)+ t(Q-x(Q))$, $0\leq t\leq 1$. Let $f_n$ be the function defined on $[0,1]$ by $f_n(t):=\mu( Q_n(t))$. The function $f_n$ is a non-decreasing function with $f(0)=0$ since $x(Q)$ is not a point mass and $f(1)=\mu(Q)$.
Let $t_n^0=\sup\{t\in[0,1]\,;\, \mu(Q_n(t))< m\}$. We first observe that the fact that the measure $\mu$ has no point masses, gives that $t_n^0>0$.
There are two possibilities:
1. \[enum:possib1\] $f_n$ is continuous at $t_n^0$.
2. \[enum:possib2\] $f_n$ has a jump discontinuity at $t_n^0$.
Assume that holds. If $t_n^0<1$, we have $$m\geq \lim_{t\rightarrow (t_n^0)^-} f_n(t)= \mu((Q_n(t_n^0))^\circ)= \lim_{t\rightarrow (t_n^0)^+} f_n(t)= \mu(\overline{Q_n(t_n^0)}) \geq m.$$ Then we choose the set $H_Q= (Q_n(t_n^0))^\circ$ and $\mu(H_Q)=m$.
If $t_n^0=1$, then $$\lim_{t\rightarrow 1^-} f_n(t)=\mu((Q)^\circ)=\mu(Q)=m$$ and we also take $H_Q=(Q)^\circ$.
Assume now that holds. If $t_n^0<1$ we have
$$\alpha:=\mu((Q_n(t_n^0))^\circ)=\lim_{t\rightarrow (t_n^0)^-} f_n(t) < \lim_{t\rightarrow (t_n | t)$ is a $ t$-homotetic cube in $ Q$ with the same center of $ Q$, that is, $ Q_n(t)= x(Q)+ t(Q - x(Q))$, $ 0\leq t\leq 1$. Let $ f_n$ be the affair define on $ [ 0,1]$ by $ f_n(t):=\mu (Q_n(t))$. The function $ f_n$ is a non - decreasing function with $ f(0)=0 $ since $ x(Q)$ is not a point multitude and $ f(1)=\mu(Q)$.
Let $ t_n^0=\sup\{t\in[0,1]\,;\, \mu(Q_n(t) ) < m\}$. We first respect that the fact that the meter $ \mu$ has no point mass, gives that $ t_n^0>0$.
There are two possibility:
1. \[enum: possib1\ ] $ f_n$ is continuous at $ t_n^0$.
2. \[enum: possib2\ ] $ f_n$ has a jump discontinuity at $ t_n^0$.
Assume that hold. If $ t_n^0<1 $, we have $ $ m\geq \lim_{t\rightarrow (t_n^0)^- } f_n(t)= \mu((Q_n(t_n^0))^\circ)= \lim_{t\rightarrow (t_n^0)^+ } f_n(t)= \mu(\overline{Q_n(t_n^0) }) \geq m.$$ Then we choose the set $ H_Q= (Q_n(t_n^0))^\circ$ and $ \mu(H_Q)=m$.
If $ t_n^0=1 $, then $ $ \lim_{t\rightarrow 1 ^ - } f_n(t)=\mu((Q)^\circ)=\mu(Q)=m$$ and we also learn $ H_Q=(Q)^\circ$.
Assume now that holds. If $ t_n^0<1 $ we consume
$ $ \alpha:=\mu((Q_n(t_n^0))^\circ)=\lim_{t\rightarrow (t_n^0)^- } f_n(t) < \lim_{t\rightarrow (t_n | t)$ id a $t$-homotetic cube in $Q$ with the same ewnter mf $Q$, tgat is, $Q_v(t)= x(Q)+ t(Q-x(Q))$, $0\leq t\leq 1$. Let $f_n$ ue tye fubction defined on $[0,1]$ by $w_n(t):=\mu( Q_n(t))$. The funxtioi $f_n$ is a non-deccsasing nbnctikk witk $h(0)=0$ since $x(Q)$ is npt a point mass and $f(1)=\mu(Q)$.
Neg $c_n^0=\sup\{t\in[0,1]\,;\, \mu(Q_n(t))< m\}$. We first observe thwt the gaft that the mewsurt $\mt$ haa no point masses, gives that $t_n^0>0$.
Thede are uwo possibilities:
1. \[enum:possib1\] $f_n$ is continuoks ah $t_n^0$.
2. \[enum:possib2\] $f_j$ has a jumk difxontinuity ag $t_n^0$.
Assume that holds. Jf $t_n^0<1$, we have $$m\geq \lim_{t\rightarruw (t_n^0)^-} f_n(t)= \mu((Q_n(t_b^0))^\curc)= \nim_{t\rightarcow (t_n^0)^+} f_n(t)= \mu(\overllme{Q_n(t_n^0)}) \geq m.$$ Yhen we choose thx ser $H_Q= (Q_n(t_n^0))^\circ$ and $\mu(H_X)=m$.
If $t_n^0=1$, then $$\lim_{t\ryghtarrow 1^-} y_n(t)=\mu((Q)^\circ)=\mu(Q)=m$$ and we aoso tdke $V_Q=(Q)^\ckec$.
Arsuje nkw thah hklds. If $t_n^0<1$ we have
$$\alpya:=\mu((Q_n(t_n^0))^\circ)=\lim_{t\rigntwgtow (t_n^0)^-} f_n(t) < \mim_{t\ridhearrow (t_n | t)$ is a $t$-homotetic cube in $Q$ same of $Q$, is, $Q_n(t)= x(Q)+ $f_n$ the function defined $[0,1]$ by $f_n(t):=\mu( The function $f_n$ is a non-decreasing with $f(0)=0$ since $x(Q)$ is not a point mass and $f(1)=\mu(Q)$. Let $t_n^0=\sup\{t\in[0,1]\,;\, m\}$. We first observe that the fact that the measure $\mu$ has no masses, that There two possibilities: 1. \[enum:possib1\] $f_n$ is continuous at $t_n^0$. 2. \[enum:possib2\] $f_n$ has a jump discontinuity $t_n^0$. Assume that holds. If $t_n^0<1$, we have \lim_{t\rightarrow (t_n^0)^-} f_n(t)= \mu((Q_n(t_n^0))^\circ)= (t_n^0)^+} f_n(t)= \mu(\overline{Q_n(t_n^0)}) \geq m.$$ we the set (Q_n(t_n^0))^\circ$ $\mu(H_Q)=m$. $t_n^0=1$, then $$\lim_{t\rightarrow f_n(t)=\mu((Q)^\circ)=\mu(Q)=m$$ and we also take $H_Q=(Q)^\circ$. Assume now that holds. If $t_n^0<1$ we have $$\alpha:=\mu((Q_n(t_n^0))^\circ)=\lim_{t\rightarrow (t_n^0)^-} f_n(t) \lim_{t\rightarrow (t_n | t)$ is a $t$-homotetic cube in $Q$ with The same cenTer of $q$, thAt iS, $Q_N(t)= x(Q)+ T(Q-x(Q))$, $0\Leq t\leq 1$. Let $f_n$ be THe fuNction defined on $[0,1]$ by $f_n(t):=\mu( q_n(t))$. ThE fUNctiON $f_N$ is a nOn-decreASiNG FunCtIoN wiTh $F(0)=0$ SiNce $x(Q)$ Is nOt a poinT mass and $f(1)=\mU(Q)$.
LEt $T_n^0=\sup\{t\in[0,1]\,;\, \mu(Q_n(T))< M\}$. WE first obseRve That the fact tHat The meaSuRe $\mU$ Has no PoiNt masSes, givES that $t_N^0>0$.
There are TwO PossibILities:
1. \[eNUM:pOssiB1\] $f_n$ is continuous at $T_N^0$.
2. \[eNUm:possib2\] $f_n$ has a Jump diScONtINUitY at $T_n^0$.
Assume thAt Holds. iF $t_n^0<1$, we haVE $$m\GEQ \Lim_{T\Rightarrow (t_n^0)^-} f_N(t)= \mu((Q_n(t_n^0))^\cirC)= \Lim_{T\rightArRow (T_N^0)^+} f_n(t)= \mu(\OverlInE{q_n(t_N^0)}) \geq m.$$ Then we ChooSe the set $H_q= (Q_n(t_n^0))^\cIRc$ and $\mu(h_q)=m$.
If $t_n^0=1$, tHen $$\lim_{T\riGhtArroW 1^-} F_n(T)=\mU((Q)^\cIrC)=\Mu(Q)=M$$ AnD we ALso Take $H_Q=(Q)^\cIrC$.
ASsume Now tHAT HOlds. if $t_N^0<1$ we hAve
$$\alPha:=\mu((Q_n(t_n^0))^\circ)=\Lim_{T\rigHTarRow (t_n^0)^-} F_n(t) < \liM_{t\riGhTarroW (t_n | t)$ is a $t$-homotetic cub e in $Q$ w ith t hesam ecent er o f $Q$, that is , $Q_ n(t)= x(Q)+ t(Q-x(Q))$ , $0\ le q t\l e q1$. L et $f_n $ b e the f un cti on de fined on $[0,1] $ by $f_n( t): =\ mu( Q_n(t))$ . T he functio n $ f_n$ is a no n-d ecreas in g f u nctio n w ith $ f(0)=0 $ since $x(Q)$ i sn ot a p o int mas s an d $f (1)=\mu(Q)$.
Let $t _ n^0=\sup\{t\in [0,1]\ ,; \ ,\ m u(Q _n( t))< m\}$. W e fir s t obser v et h a t t h e fact that t he measure$ \mu $ hasno po i nt mas ses,gi v esthat $t_n^0 >0$.
There a re two possibi l ities:
1. \ [en um: poss i b1 \] $f _n $ is co nti n uou s at $t_ n^ 0$ .
2. \[ e n u m :pos sib 2\]$f_n$ has a jump d isc onti n uit y at$t_n^ 0$.
A ssume thatholds .If $t_n^0<1$, w e ha ve $$m\ge q \ li m_{ t\ right a rrow ( t_n ^0) ^-} f_n (t)= \m u ((Q _n ( t _ n^ 0))^\circ)= \lim_{ t\ r i gh tarrow ( t_n^0) ^ +} f _ n(t)= \m u( \ov erli n e {Q_n( t_n^ 0 )} ) \geq m .$$ Th e nwe choose t he set $ H_Q = ( Q_n(t _ n^0) )^\cir c$ and $ \mu(H _ Q)=m$.
If $t_ n ^0=1$, then $ $ \l i m _{ t \rig hta rrow 1^-} f _n(t ) =\mu ((Q) ^ \c irc ) =\mu( Q)=m$ $a nd we also take $H_Q=( Q) ^\circ $.
A ssume now tha t holds. I f $ t_n^0<1$ weh av e
$$\alpha:=\m u((Q_ n(t_n^0))^ \ circ)=\l im_{t \rightar row (t_n^ 0 ) ^-} f_n( t)< \ lim _{t \ r ig htarrow (t_n | t)$ is_a $t$-homotetic_cube in $Q$ with_the same_center_of $Q$,_that_is, $Q_n(t)= x(Q)+_t(Q-x(Q))$, $0\leq t\leq_1$. Let $f_n$ be_the function defined_on_$[0,1]$ by $f_n(t):=\mu( Q_n(t))$. The function $f_n$ is a non-decreasing function with $f(0)=0$ since_$x(Q)$_is not_a_point_mass and $f(1)=\mu(Q)$.
Let $t_n^0=\sup\{t\in[0,1]\,;\, \mu(Q_n(t))<_m\}$. We first observe that_the fact_that the measure $\mu$ has no point masses,_gives_that $t_n^0>0$.
There are_two possibilities:
1. \[enum:possib1\] $f_n$ is continuous at $t_n^0$.
2._ \[enum:possib2\] $f_n$ has a jump_discontinuity at $t_n^0$.
Assume_that_holds._If $t_n^0<1$, we have_$$m\geq \lim_{t\rightarrow (t_n^0)^-} f_n(t)= \mu((Q_n(t_n^0))^\circ)= \lim_{t\rightarrow_(t_n^0)^+} f_n(t)= \mu(\overline{Q_n(t_n^0)}) \geq m.$$ Then_we choose the set $H_Q= (Q_n(t_n^0))^\circ$ and_$\mu(H_Q)=m$.
If $t_n^0=1$, then $$\lim_{t\rightarrow 1^-} f_n(t)=\mu((Q)^\circ)=\mu(Q)=m$$_and we also take $H_Q=(Q)^\circ$.
Assume_now that_holds. If $t_n^0<1$ we have
$$\alpha:=\mu((Q_n(t_n^0))^\circ)=\lim_{t\rightarrow_(t_n^0)^-} f_n(t) <_\lim_{t\rightarrow (t_n |
]{};
(orig) – (2.25,0) \[->\] (2.25,0) – (removeOnePod0); (removeOnePod0) – (removeOnePod0B); (orig) – (2.25,0) \[->\] (2.25,0) |- (playToPod1); (orig) – (2.25,0) \[->\] (2.25,0) |- (playToPod2); (playToPod1) – (playToPod0); (playToPod2) – (playToPod0); (orig) – (2.25,0) \[->\] (2.25,0) |- (removeOnePod21); (removeOnePod21) – (removeOnePod21B); (orig) – (2.25,0) \[->\] (2.25,0) |- (playToPm); (playToPm) – (playToPmB);
**Case 2 :** $P_2 \sim_1$ [$S_{2,2}$]{}
Figure \[fig:pod1EQUIV1pod22\] shows the possible moves on $P_2$ or [$S_{2,2}$]{}, and the answer leading to a ${\mathcal{P}}$-position (for readability, we write $S$ instead of $\hat{S}$ in the figure).
(orig) at (0,0) [ ]{};
(removeTwoPod1) at (5,0)
\(1) at (-1.1,1) ; (2) at (-2.1,0) ; (2b) at (-3.1,0) ; (2c) at (-3.65,0) [$S$]{}; (-3,0) – (-3.75,0.5); (-3,0) – (-3.75,-0.5); (-3.75,0.5) arc (135:225:0.7); (-1.4,0) node [+]{};
(1,0) – (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5) arc (135:225:0. | ] { };
(orig) – (2.25,0) \[->\ ] (2.25,0) – (removeOnePod0); (removeOnePod0) – (removeOnePod0B); (orig) – (2.25,0) \[->\ ] (2.25,0) |- (playToPod1); (orig) – (2.25,0) \[->\ ] (2.25,0) |- (playToPod2); (playToPod1) – (playToPod0); (playToPod2) – (playToPod0); (orig) – (2.25,0) \[->\ ] (2.25,0) |- (removeOnePod21); (removeOnePod21) – (removeOnePod21B); (orig) – (2.25,0) \[->\ ] (2.25,0) |- (playToPm); (playToPm) – (playToPmB);
* * Case 2 :* * $ P_2 \sim_1 $ [ $ S_{2,2}$ ] { }
Figure \[fig: pod1EQUIV1pod22\ ] shows the possible moves on $ P_2 $ or [ $ S_{2,2}$ ] { }, and the solution go to a $ { \mathcal{P}}$-position (for readability, we write $ S$ alternatively of $ \hat{S}$ in the digit).
(orig) at (0,0) [ ] { };
(removeTwoPod1) at (5,0)
\(1) at (-1.1,1); (2) at (-2.1,0); (2b) at (-3.1,0); (2c) at (-3.65,0) [ $ S$ ] { }; (-3,0) – (-3.75,0.5); (-3,0) – (-3.75,-0.5); (-3.75,0.5) arc (135:225:0.7); (-1.4,0) node [ + ] { };
(1,0) – (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5) arc (135:225:0. | ]{};
(orih) – (2.25,0) \[->\] (2.25,0) – (removeOnePod0); (reooveOnePod0) – (removeOnePmd0B); (orjg) – (2.25,0) \[->\] (2.25,0) |- (playToPod1); (orig) – (2.25,0) \[->\] (2.25,0) |- (playVoPoe2); (platToPod1) – (playToPod0); (playGoPod2) – (plwyToPod0); (irig) – (2.25,0) \[->\] (2.25,0) |- (removeOnePod21); (remobcOnePmv21) – (removeOnePod21N); (orig) – (2.25,0) \[->\] (2.25,0) |- (playToPm); (pldyGo'm) – (playToPmB);
**Case 2 :** $P_2 \sim_1$ [$S_{2,2}$]{}
Figure \[fig:[od1EQUIF1pld22\] shows the pjssinje mknew on $P_2$ or [$S_{2,2}$]{}, and the answer lezding tm a ${\mathcal{P}}$-ppsition (for readability, we wrihe $S$ instead of $\hah{S}$ in the fugurq).
(irig) at (0,0) [ ]{};
(reooveTwoPod1) at (5,0)
\(1) at (-1.1,1) ; (2) aj (-2.1,0) ; (2b) at (-3.1,0) ; (2c) at (-3.65,0) [$S$]{}; (-3,0) – (-3.75,0.5); (-3,0) – (-3.75,-0.5); (-3.75,0.5) arc (135:225:0.7); (-1.4,0) vode [+]{};
(1,0) – (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5) arc (135:225:0. | ]{}; (orig) – (2.25,0) \[->\] (2.25,0) – – (orig) – \[->\] (2.25,0) |- (2.25,0) (playToPod2); (playToPod1) – (playToPod2) – (playToPod0); – (2.25,0) \[->\] (2.25,0) |- (removeOnePod21); – (removeOnePod21B); (orig) – (2.25,0) \[->\] (2.25,0) |- (playToPm); (playToPm) – (playToPmB); **Case :** $P_2 \sim_1$ [$S_{2,2}$]{} Figure \[fig:pod1EQUIV1pod22\] shows the possible moves on $P_2$ or and answer to ${\mathcal{P}}$-position (for readability, we write $S$ instead of $\hat{S}$ in the figure). (orig) at (0,0) [ (removeTwoPod1) at (5,0) \(1) at (-1.1,1) ; (2) (-2.1,0) ; (2b) at ; (2c) at (-3.65,0) [$S$]{}; – (-3,0) – (-3.75,0.5) (135:225:0.7); node [+]{}; (1,0) (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5) arc (135:225:0. | ]{};
(orig) – (2.25,0) \[->\] (2.25,0) – (removeOnePod0); (removeONePod0) – (removEOnePOd0B); (OriG) – (2.25,0) \[-&gT;\] (2.25,0) |- (plaYToPOd1); (orig) – (2.25,0) \[->\] (2.25,0) |- (playTopOd2); (plAyToPod1) – (playToPod0); (playTopod2) – (plAytOPod0); (ORiG) – (2.25,0) \[->\] (2.25,0) |- (reMoveOnepOd21); (REMovEONepod21) – (ReMOvEOnePOd21B); (Orig) – (2.25,0) \[->\] (2.25,0) |- (pLayToPm); (plaYTopm) – (PlayToPmB);
**CasE 2 :** $p_2 \sIm_1$ [$S_{2,2}$]{}
Figure \[fIg:pOd1EQUIV1pod22\] shOws The posSiBle MOves oN $P_2$ oR [$S_{2,2}$]{}, and The ansWEr leadIng to a ${\matHcAL{P}}$-posiTIon (for rEADaBiliTy, we write $S$ instead OF $\hAT{S}$ in the figure).
(oRig) at (0,0) [ ]{};
(rEmOVetWOPoD1) at (5,0)
\(1) At (-1.1,1) ; (2) at (-2.1,0) ; (2b) at (-3.1,0) ; (2c) at (-3.65,0) [$s$]{}; (-3,0) – (-3.75,0.5); (-3,0) – (-3.75,-0.5); (-3.75,0.5) aRc (135:225:0.7); (-1.4,0) nodE [+]{};
(1,0) – (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5) Arc (135:225:0. | ]{};
(orig) – (2.25,0) \[ ->\] (2 .25,0 ) – (r em oveO nePo d0); (removeOn e Pod0 ) – (removeOnePod0B);(orig )– (2. 2 5, 0) \[ ->\] (2 . 2 5,0 )|- (p la y To Pod1) ; ( orig) – (2.25,0)\[- &g t;\] (2.25,0 ) | - (playToP od2 ); (playToPo d1) – (pl ay ToP o d0);(pl ayToP od2) – (playT oPod0); ( or i g) – ( 2 .25,0)\ [ -& gt;\ ] (2.25,0) |- (re m ov e OnePod21); (re moveOn eP o d2 1 ) –(re moveOnePod 21 B); ( o rig) –( 2. 2 5 , 0)\ [->\] (2.2 5,0) |- (pl a yTo Pm); ( pl ayT o Pm) –(play To P mB) ;
**Case 2 :** $P_2 \si m_1$ [ $ S_{2,2} $ ]{}
Fi gure \ [fi g:p od1E Q UI V1 pod 22 \ ] s h ow s t h e p ossiblemo ve s on$P_2 $ o r [$S _{2 ,2}$ ]{},and the answe r l eadi n g t o a $ {\mat hcal {P }}$-p ositio n (fo rreadability, we wri te $S$ in ste ad of $ \hat{ S }$ inthe fi gure).
(orig) at(0 , 0 ) [ ]{};
(removeTwoP od 1 ) a t (5,0)
\(1)a t(- 1 .1,1) ;(2 ) a t (- 2 . 1,0); (2 b )at (-3.1 ,0) ;( 2c )at (-3. 65 ,0) [$ S$ ]{} ; ( -3,0) – (- 3.75,0 .5); (-3 ,0) – (-3.75,-0.5);( -3.75,0.5) ar c ( 1 3 5: 2 25:0 .7) ; (-1.4,0)node [+]{ };
( 1, 0)– (0.2 5,0.5 ); (1 , 0) – (0.25,-0.5); ( 0. 25,0.5 ) arc (135:225:0. | ]{};
(orig) –_(2.25,0) \[->\]_(2.25,0) – (removeOnePod0); (removeOnePod0)_– (removeOnePod0B);_(orig)_– (2.25,0)_\[->\]_(2.25,0) |- (playToPod1);_(orig) – (2.25,0)_\[->\] (2.25,0) |- (playToPod2);_(playToPod1) – (playToPod0);_(playToPod2)_– (playToPod0); (orig) – (2.25,0) \[->\] (2.25,0) |- (removeOnePod21); (removeOnePod21) – (removeOnePod21B); (orig) –_(2.25,0)_\[->\] (2.25,0)_|-_(playToPm);_(playToPm) – (playToPmB);
**Case 2 :**_$P_2 \sim_1$ [$S_{2,2}$]{}
Figure \[fig:pod1EQUIV1pod22\] shows the_possible moves_on $P_2$ or [$S_{2,2}$]{}, and the answer leading_to_a ${\mathcal{P}}$-position (for_readability, we write $S$ instead of $\hat{S}$ in the_figure).
(orig) at (0,0) [ ]{};
(removeTwoPod1) at_(5,0)
\(1) at (-1.1,1)_;_(2)_at (-2.1,0) ; (2b)_at (-3.1,0) ; (2c) at (-3.65,0)_[$S$]{}; (-3,0) – (-3.75,0.5); (-3,0) –_(-3.75,-0.5); (-3.75,0.5) arc (135:225:0.7); (-1.4,0) node [+]{};
(1,0)_– (0.25,0.5); (1,0) – (0.25,-0.5); (0.25,0.5)_arc (135:225:0. |
enheimer-Snyder collapse, we assume the shell starts to fall from a turning point $r_0$ at $\tau=0$, where $\dot{r}_s = 0$. We can evaluate equation (\[eq:thinshelleom\]) at the turning point, and thus we get $$M = \frac{\sqrt{f_-(r_0)}-\sqrt{f_+(r_0)}}{r_0^{1-n(1+c)}}\;.$$ This expression with equation (\[eq:rsdot2\]) determines the motion of the thin shell.
Trajectories of the thin shell
------------------------------
Next we wish to inspect the possible trajectories of the shell. Instead of solving the position of the shell $r_s$ explicitly as a function of the coordinate time $t_s$, it is sufficient to look at the expression $\dot{r}_s^2$ in equation (\[eq:rsdot2\]). First of all, since $\dot{r}_s^2$ is a square of the physical velocity, it has to be positive. Additionally, since $\ddot{r}_s = \frac{1}{2}\frac{d}{dr_s}\dot{r}_s^2$, the same expression also encodes information of the acceleration for the trajectory. Thus by computing the acceleration at the turning point, $$\left.\frac{d^2r_s}{d\tau^2}\right|_{r_s=r_0}= - \left[ r_0 \Lambda + \frac{m}{r_0^n}\frac{n-1}{4} + \frac{n(1+c)-1}{4}\frac{M^2}{r_0^{n(1+c)}} - \frac{n\,c\,m^2\,r_0^{2cn-1}}{4\,M^2} \right] \;,$$ we can determine towards which direction the shell starts to move. As long as the expression in the square brackets is positive, the shell starts to fall at the turning point. However, if $$\frac{n\,c\,m^2\,r_0^{2cn-1}}{M^2} \; > \; 4\, r_0 \Lambda + (n-1)\frac{m}{r_0^n} + \left[n(1+c)-1\right]\frac{M^2}{r_0^{n( | enheimer - Snyder collapse, we assume the shell starts to decrease from a turn point $ r_0 $ at $ \tau=0 $, where $ \dot{r}_s = 0$. We can evaluate equality (\[eq: thinshelleom\ ]) at the turning compass point, and thus we get down $ $ M = \frac{\sqrt{f_-(r_0)}-\sqrt{f_+(r_0)}}{r_0^{1 - n(1+c)}}\;.$$ This expression with equality (\[eq: rsdot2\ ]) determine the motion of the flimsy shell.
Trajectories of the thin carapace
------------------------------
Next we wish to inspect the possible trajectories of the shell. Instead of solve the position of the shell $ r_s$ explicitly as a function of the coordinate prison term $ t_s$, it is sufficient to look at the expression $ \dot{r}_s^2 $ in equation (\[eq: rsdot2\ ]). First of all, since $ \dot{r}_s^2 $ is a public square of the physical speed, it has to be positive. Additionally, since $ \ddot{r}_s = \frac{1}{2}\frac{d}{dr_s}\dot{r}_s^2 $, the same expression besides encodes information of the acceleration for the trajectory. Thus by computing the acceleration at the turning point, $ $ \left.\frac{d^2r_s}{d\tau^2}\right|_{r_s = r_0}= - \left [ r_0 \Lambda + \frac{m}{r_0^n}\frac{n-1}{4 } + \frac{n(1+c)-1}{4}\frac{M^2}{r_0^{n(1+c) } } - \frac{n\,c\,m^2\,r_0^{2cn-1}}{4\,M^2 } \right ] \;,$$ we can determine towards which direction the shell starts to move. As long as the expression in the square brackets is positive, the shell start to fall at the turning distributor point. However, if $ $ \frac{n\,c\,m^2\,r_0^{2cn-1}}{M^2 } \; > \; 4\, r_0 \Lambda + (n-1)\frac{m}{r_0^n } + \left[n(1+c)-1\right]\frac{M^2}{r_0^{n ( | enhfimer-Snyder collapse, we xssume the shell startv to fzll from a turning point $r_0$ at $\tau=0$, whxre $\eot{r}_s = 0$. We can evaluate equxtion (\[eq:tjinshellwom\]) et the turning point, and thus sc get $$N = \frac{\sqrt{f_-(r_0)}-\sart{f_+(r_0)}}{r_0^{1-n(1+c)}}\;.$$ Tvis expression wktk equation (\[eq:rsdot2\]) determines the moeion of tje thin shell.
Ttajecuoryes kf the thin shell
------------------------------
Next we wish to ihspect uhe possible trajevtories of the shell. Instewd ov solving the posihion of the shejo $r_s$ explicigly as a fluction of tge coordinate time $t_s$, it is sufwicieut to look qt thf expression $\dot{r}_f^2$ in equatiok (\[eq:rsdmt2\]). Firsy of all, since $\dov{r}_s^2$ us a square of the phbsical velocity, it hws to be [oaitive. Additionaloy, sincg $\ddod{r}_s = \draz{1}{2}\frzc{v}{dr_a}\dot{r}_s^2$, thx same exprsssion also encodes informatiom jd the accelerztion sow the trajectory. Thus by computing the dccsleration at the turnint point, $$\left.\frac{d^2r_s}{d\twu^2}\right|_{r_s=w_0}= - \left[ r_0 \Lambda + \frac{m}{r_0^n}\frac{n-1}{4} + \frac{n(1+c)-1}{4}\frac{M^2}{r_0^{n(1+c)}} - \fsac{n\,c\,j^2\,f_0^{2cn-1}}{4\,N^2} \vighg] \;,$$ ae can determine towards which direction the fgekl starts to movc. As long as the ecpgexfion in the sduare ydadkets is positive, hhe shejl stqrts to fwll st the turning point. Howevee, if $$\frac{n\,c\,m^2\,g_0^{2cn-1}}{M^2} \; > \; 4\, r_0 \Lambda + (n-1)\frcc{m}{r_0^n} + \left[n(1+e)-1\right]\grac{M^2}{t_0^{n( | enheimer-Snyder collapse, we assume the shell starts from turning point at $\tau=0$, where evaluate (\[eq:thinshelleom\]) at the point, and thus get $$M = \frac{\sqrt{f_-(r_0)}-\sqrt{f_+(r_0)}}{r_0^{1-n(1+c)}}\;.$$ This expression equation (\[eq:rsdot2\]) determines the motion of the thin shell. Trajectories of the thin ------------------------------ Next we wish to inspect the possible trajectories of the shell. Instead solving position the $r_s$ explicitly as a function of the coordinate time $t_s$, it is sufficient to look at expression $\dot{r}_s^2$ in equation (\[eq:rsdot2\]). First of all, $\dot{r}_s^2$ is a square the physical velocity, it has be Additionally, since = the expression also encodes of the acceleration for the trajectory. Thus by computing the acceleration at the turning point, $$\left.\frac{d^2r_s}{d\tau^2}\right|_{r_s=r_0}= - r_0 \Lambda + \frac{n(1+c)-1}{4}\frac{M^2}{r_0^{n(1+c)}} \frac{n\,c\,m^2\,r_0^{2cn-1}}{4\,M^2} \;,$$ can determine towards the shell starts to move. As expression in the square brackets is positive, the starts to at the turning point. However, if \; > \; 4\, r_0 \Lambda + (n-1)\frac{m}{r_0^n} \left[n(1+c)-1\right]\frac{M^2}{r_0^{n( | enheimer-Snyder collapse, we aSsume the shEll stArtS to FaLl frOm a tUrning point $r_0$ at $\TAu=0$, whEre $\dot{r}_s = 0$. We can evaluate eQuatiOn (\[EQ:thiNShElleoM\]) at the tURnING poInT, aNd tHuS We Get $$M = \fRac{\Sqrt{f_-(r_0)}-\sQrt{f_+(r_0)}}{r_0^{1-n(1+c)}}\;.$$ ThIs eXpRession with eQUaTion (\[eq:rsdoT2\]) deTermines the mOtiOn of thE tHin SHell.
TRajEctorIes of tHE thin sHell
------------------------------
Next wE wISh to inSPect the POSsIble Trajectories of the SHeLL. Instead of solvIng the PoSItION of The Shell $r_s$ expLiCitly AS a functIOn OF THe cOOrdinate time $t_S$, it is sufficIEnt To look At The EXpressIon $\doT{r}_S^2$ In eQuation (\[eq:rsDot2\]). FIrst of all, Since $\dOT{r}_s^2$ is a sQUare of tHe physIcaL veLociTY, iT hAs tO bE PosITiVe. ADDitIonally, sInCe $\Ddot{r}_S = \fraC{1}{2}\FRAC{d}{dr_S}\doT{r}_s^2$, tHe samE expression alSo eNcodES inFormaTion oF the AcCelerAtion fOr the TrAjectory. Thus by cOmpuTing the acCelErAtiOn At the TUrning PoiNt, $$\lEft.\frac{D^2r_s}{d\tau^2}\RIghT|_{r_S=R_0}= - \LEfT[ r_0 \Lambda + \frac{m}{r_0^n}\fraC{n-1}{4} + \FRAc{N(1+c)-1}{4}\frac{M^2}{r_0^{N(1+c)}} - \frac{N\,C\,m^2\,R_0^{2cN-1}}{4\,m^2} \right] \;,$$ we CaN deTermINE towaRds wHIcH directiOn the sHElL sTarts to MoVe. As loNg As tHe eXpresSIon iN the sqUare bracKets iS Positive, the sheLL starts to fall AT tHE TuRNing PoiNt. However, if $$\Frac{N\,C\,m^2\,r_0^{2cN-1}}{M^2} \; > \; 4\, r_0 \LAMbDa + (n-1)\FRac{m}{r_0^N} + \left[N(1+c)-1\RIgHT]\frac{M^2}{r_0^{n( | enheimer-Snyder collapse,we assumethe s hel l s ta rtsto f all from a tur n ingpoint $r_0$ at $\tau=0 $, wh er e $\d o t{ r}_s= 0$. W e c a n ev al ua teeq u at ion ( \[e q:thins helleom\]) at t he turning p o in t, and thu s w e get $$M =\fr ac{\sq rt {f_ - (r_0) }-\ sqrt{ f_+(r_ 0 )}}{r_ 0^{1-n(1+ c) } }\;.$$ This ex p r es sion with equation (\ [ eq : rsdot2\]) dete rmines t h em o tio n o f the thin s hell.
Trajec t or i e s of the thin shel l
--------- - --- ------ -- --- - -----
Next w e wi sh to inspe ct t he possib le tra j ectorie s of the shell . I nst eado fso lvi ng the po sit i onof the s he ll $r_s $ ex p l i c itly as a f uncti on of the coo rdi nate tim e $t_ s$, i t is s uffic ient t o loo kat the expressi on $ \dot{r}_s ^2$ i n e qu ation (\[eq: rsd ot2 \]). Fi rst ofa ll, s i n c e$\dot{r}_s^2$ is a s q u ar e of the physi c al v e locity,it ha s to b e pos itiv e .Addition ally,s in ce $\ddot {r }_s =\f rac {1} {2}\f r ac{d }{dr_s }\dot{r} _s^2$ , the same expr e ssion also en c od e s i n form ati on of the a ccel e rati on f o rthe traje ctory .T hu s by computing the a cc elerat ion a t the turning point, $$ \ l e ft.\frac {d^2 r _s } {d\tau^2}\righ t|_{r _s=r_0}= - \left[ r _0 \L ambda +\frac{m}{ r _ 0^n}\fra c{n -1} {4} +\ f ra c{n(1+c)-1}{4 } \ frac {M ^2}{r_0 ^{n (1+c)}} -\fr ac{ n\, c\ ,m^2\,r_0 ^{2cn-1} }{ 4\ ,M ^2 } \ right ] \;,$$ w ecan d ete rmine toward s whi ch d ir ec t ion the sh e ll s tart sto mov e.As long ast heexpress ion in th e s q uare b ra ckets i s positive, t he shell sta rt s t o fall a t the tu rning point. However, i f $$\fra c{n \,c\, m^2\ ,r_0^{2cn -1} }{M^2} \; > \; 4 \, r_0 \Lam bd a + ( n-1)\ f r ac {m} {r _0^n} + \l e f t[n (1+c) -1 \rig ht]\fra c{M^2}{r_0^{n( | enheimer-Snyder collapse,_we assume_the shell starts to_fall from_a_turning point_$r_0$_at $\tau=0$, where_$\dot{r}_s = 0$._We can evaluate equation_(\[eq:thinshelleom\]) at the_turning_point, and thus we get $$M = \frac{\sqrt{f_-(r_0)}-\sqrt{f_+(r_0)}}{r_0^{1-n(1+c)}}\;.$$ This expression with equation (\[eq:rsdot2\]) determines_the_motion of_the_thin_shell.
Trajectories of the thin shell
------------------------------
Next_we wish to inspect the_possible trajectories_of the shell. Instead of solving the position_of_the shell $r_s$_explicitly as a function of the coordinate time $t_s$,_it is sufficient to look at_the expression $\dot{r}_s^2$_in_equation_(\[eq:rsdot2\]). First of all,_since $\dot{r}_s^2$ is a square of_the physical velocity, it has to_be positive. Additionally, since $\ddot{r}_s = \frac{1}{2}\frac{d}{dr_s}\dot{r}_s^2$,_the same expression also encodes information_of the acceleration for the_trajectory. Thus_by computing the acceleration at_the turning point,_$$\left.\frac{d^2r_s}{d\tau^2}\right|_{r_s=r_0}= -_\left[ r_0 \Lambda_+ \frac{m}{r_0^n}\frac{n-1}{4} + \frac{n(1+c)-1}{4}\frac{M^2}{r_0^{n(1+c)}} - \frac{n\,c\,m^2\,r_0^{2cn-1}}{4\,M^2}_\right] \;,$$ we_can determine towards which direction the_shell_starts to move._As_long_as the_expression in the_square_brackets is_positive,_the shell starts to fall at_the_turning point. However, if $$\frac{n\,c\,m^2\,r_0^{2cn-1}}{M^2} \; >_\; 4\, r_0 \Lambda_+_(n-1)\frac{m}{r_0^n} + \left[n(1+c)-1\right]\frac{M^2}{r_0^{n( |
. Let $S$ be the universal subbundle on ${\operatorname{Quot}}E \times C$. Let $$c_i(S^\lor) = a_i \otimes 1 + \sum_{j=1}^{2g} b_{i,j} \otimes \delta_j + f_i \otimes \eta$$ be the Künneth decomposition of $c_i(S^\lor)$. In the argot of [@Mar], [@MO], etc. the classes $$\begin{aligned}
a_i & \in & {\operatorname{H}}^{2i}({\operatorname{Quot}}^{r,e} E) \\ b_{i,j} & \in & {\operatorname{H}}^{2i-1}({\operatorname{Quot}}^{r,e} E) \\ f_i & \in & {\operatorname{H}}^{2i-2}({\operatorname{Quot}}^{r,e} E)\end{aligned}$$ are called $a,b$ and $f$ *classes*.
A *tautological class* is an element of the subring ${\overline}{{\operatorname{R}}}^*$ of ${\operatorname{H}}^*({\operatorname{Quot}}E)$ generated by the $a$, $b$, and $f$ classes. The *tautological ring* ${\operatorname{R}}^*$ is the quotient of ${\overline}{{\operatorname{R}}}^*$ by the ideal $$\{ \alpha \in {\overline}{{\operatorname{R}}}^* : \int_{[{\operatorname{Quot}}E]^{\rm vir}} \alpha \beta =0 \; {\rm for \; all} \; \beta \in {\overline}{{\operatorname{R}}}^* \}.$$
Unlike the actual cohomology ring ${\operatorname{H}}^*({\operatorname{Quot}}E)$, the tautological ring is deformation invariant: it does not depend on the curve $C$ or the bundle $E$, except through their discrete invariants. This follows from deformation-invariance properties of the virtual class and the fact that the $a$, $b$, and $f$ classes have globally defined analogues as $C,E$ vary in families which restrict to the given classes at points of the family. The virtual intersection theory of ${\operatorname{Quot}}E$ is the study of the tautological ring, usually through calculations of integrals of tautological classes over the virtual class. A typical result is the Vafa-Intriligator formula for integrals of | . Let $ S$ be the universal subbundle on $ { \operatorname{Quot}}E \times C$. Let $ $ c_i(S^\lor) = a_i \otimes 1 + \sum_{j=1}^{2 g } b_{i, j } \otimes \delta_j + f_i \otimes \eta$$ be the Künneth decay of $ c_i(S^\lor)$. In the slang of [ @Mar ], [ @MO ], etc. the class $ $ \begin{aligned }
a_i & \in & { \operatorname{H}}^{2i}({\operatorname{Quot}}^{r, e } E) \\ b_{i, joule } & \in & { \operatorname{H}}^{2i-1}({\operatorname{Quot}}^{r, e } E) \\ f_i & \in & { \operatorname{H}}^{2i-2}({\operatorname{Quot}}^{r, e } E)\end{aligned}$$ are call $ a, b$ and $ f$ * classes *.
A * tautological course * is an component of the subring $ { \overline}{{\operatorname{R}}}^*$ of $ { \operatorname{H}}^*({\operatorname{Quot}}E)$ generated by the $ a$, $ b$, and $ f$ classes. The * tautological ring * $ { \operatorname{R}}^*$ is the quotient of $ { \overline}{{\operatorname{R}}}^*$ by the ideal $ $ \ { \alpha \in { \overline}{{\operatorname{R}}}^ *: \int_{[{\operatorname{Quot}}E]^{\rm vir } } \alpha \beta = 0 \; { \rm for \; all } \; \beta \in { \overline}{{\operatorname{R}}}^ * \}.$$
Unlike the actual cohomology band $ { \operatorname{H}}^*({\operatorname{Quot}}E)$, the tautological ring is deformation changeless: it does not depend on the curve $ C$ or the bundle $ E$, except through their discrete invariants. This follows from deformation - invariance property of the virtual class and the fact that the $ a$, $ b$, and $ f$ classes have globally defined analogue as $ C, E$ vary in kin which restrict to the pass classes at points of the family. The virtual intersection theory of $ { \operatorname{Quot}}E$ is the report of the tautological ring, usually through calculations of integrals of pleonastic class over the virtual class. A typical result is the Vafa - Intriligator formula for integrals of | . Leh $S$ be the universal subnundle on ${\operatorname{Qnot}}E \tijes C$. Leg $$c_i(S^\lor) = a_i \otimes 1 + \sum_{j=1}^{2g} u_{i,j} \itimew \delta_j + f_i \otimes \etx$$ be the Nünneth dwcomkosition of $c_i(S^\loc)$. In the argot lf [@Mcr], [@MO], etc. the claxses $$\begin{dligned}
a_i & \in & {\upzratorname{H}}^{2i}({\operatorname{Quot}}^{r,e} E) \\ b_{i,t} & \in & {\ppfratorname{H}}^{2i-1}({\opgratognwme{Qhot}}^{r,e} E) \\ f_i & \in & {\operatorname{H}}^{2i-2}({\opedatornake{Quot}}^{r,e} E)\end{sligned}$$ are called $a,b$ and $v$ *clwsses*.
A *tautologicap class* is qn ejwment of the subring ${\onzrline}{{\operajorname{R}}}^*$ of ${\operatorname{H}}^*({\operatofname{Auot}}E)$ genetcred ty the $a$, $b$, end $f$ blasses. The *tautologiwal rinb* ${\operatorname{V}}^*$ is vhe wuotient of ${\overline}{{\o'eratorname{R}}}^*$ by the ydeal $$\{ \al[hc \in {\overline}{{\operatorbane{R}}}^* : \hnt_{[{\o[eragirnxme{Supt}}S]^{\rm vig}} \ampha \beta =0 \; {\rm for \; qll} \; \beta \in {\overlime}{{\jirratorname{R}}}^* \}.$$
Hnlike tre actual cohomology ring ${\operatorname{H}}^*({\mpedatorname{Quot}}E)$, the tautilogical ring is defotmation indariant: it does not depend on the curve $C$ or the tundlx $D$, ergcpt gyrlugh their discrete invariants. This follows fwkm dvformation-invariakce properties of yhf fyrtual class xnd thz fzct that the $a$, $b$, ajd $f$ clwsses have glofallu defined analogues as $C,E$ vqry in familpes qhich restrict to che given clcsses st poonts of the family. The rirtuam intersectlon theorg of ${\operatorname{Duon}}E$ iv the study of the tautolodical rinj, usuclly thruugh calcujations of intensals of tautologicwl clcsses over the girtual class. A typical result mx the Vafa-Injrinigdtor foriula nor integrals os | . Let $S$ be the universal subbundle \times Let $$c_i(S^\lor) a_i \otimes 1 + \otimes \eta$$ be Künneth decomposition of In the argot of [@Mar], [@MO], the classes $$\begin{aligned} a_i & \in & {\operatorname{H}}^{2i}({\operatorname{Quot}}^{r,e} E) \\ b_{i,j} & \in {\operatorname{H}}^{2i-1}({\operatorname{Quot}}^{r,e} E) \\ f_i & \in & {\operatorname{H}}^{2i-2}({\operatorname{Quot}}^{r,e} E)\end{aligned}$$ are called $a,b$ and *classes*. *tautological is element of the subring ${\overline}{{\operatorname{R}}}^*$ of ${\operatorname{H}}^*({\operatorname{Quot}}E)$ generated by the $a$, $b$, and $f$ classes. The ring* ${\operatorname{R}}^*$ is the quotient of ${\overline}{{\operatorname{R}}}^*$ by ideal $$\{ \alpha \in : \int_{[{\operatorname{Quot}}E]^{\rm vir}} \alpha \beta \; for \; \; \in \}.$$ Unlike the cohomology ring ${\operatorname{H}}^*({\operatorname{Quot}}E)$, the tautological ring is deformation invariant: it does not depend on the curve $C$ the bundle through their invariants. follows deformation-invariance properties of class and the fact that the $f$ classes have globally defined analogues as $C,E$ in families restrict to the given classes at of the family. The virtual intersection theory of is the study of the tautological ring, usually through calculations of integrals of tautological classes virtual class. A typical is the Vafa-Intriligator for of | . Let $S$ be the universal subbundLe on ${\operatOrnamE{QuOt}}E \TiMes C$. let $$c_I(S^\lor) = a_i \otimes 1 + \sUM_{j=1}^{2g} b_{I,j} \otimes \delta_j + f_i \otimes \Eta$$ be ThE künnETh DecomPositioN Of $C_I(s^\loR)$. IN tHe aRgOT oF [@Mar], [@Mo], etC. the claSses $$\begin{aLigNeD}
a_i & \in & {\operatoRNaMe{H}}^{2i}({\operatOrnAme{Quot}}^{r,e} E) \\ b_{i,J} & \in & {\OperatOrNamE{h}}^{2i-1}({\opeRatOrnamE{Quot}}^{r,E} e) \\ f_i & \in & {\oPeratornaMe{h}}^{2I-2}({\operaTOrname{QUOT}}^{r,E} E)\enD{aligned}$$ are called $A,B$ aND $f$ *classes*.
A *tautOlogicAl CLaSS* Is aN elEment of the SuBring ${\OVerline}{{\OPeRATOrnAMe{R}}}^*$ of ${\operatorName{H}}^*({\operatORnaMe{Quot}}e)$ gEneRAted by The $a$, $b$, AnD $F$ clAsses. The *tauToloGical ring* ${\OperatORname{R}}^*$ iS The quotIent of ${\OveRliNe}{{\opERaToRnaMe{r}}}^*$ By tHE iDeaL $$\{ \AlpHa \in {\overLiNe}{{\OperaTornAME{r}}}^* : \Int_{[{\oPerAtorName{QUot}}E]^{\rm vir}} \alphA \beTa =0 \; {\rm FOr \; aLl} \; \betA \in {\ovErliNe}{{\OperaTornamE{R}}}^* \}.$$
UnlIkE the actual cohomOlogY ring ${\operAtoRnAme{h}}^*({\oPeratORname{QUot}}e)$, thE tautolOgical rINg iS dEFORmAtion invariant: it doEs NOT dEpend on tHe curvE $c$ oR tHE bundle $E$, ExCepT thrOUGh theIr diSCrEte invarIants. THIs FoLlows frOm DeformAtIon-InvArianCE proPertieS of the viRtual CLass and the fact THat the $a$, $b$, and $f$ cLAsSES hAVe glObaLly defined aNaloGUes aS $C,E$ vARy In fAMilieS whicH rEStRIct to the given classeS aT pointS of thE family. The virTual intersECTIon theorY of ${\oPErATorname{Quot}}E$ is The stUdy of the taUTologicaL ring, Usually tHrough calCULations oF inTegRalS of TAUtOlogical classES Over ThE virtuaL clAss. A typIcaL reSulT is ThE Vafa-IntrIligator FoRmUlA fOr iNtegrALs of | . Let $S$ be the universal subbundle on $ {\o per at orna me{Q uot}}E \timesC $. L et $$c_i(S^\lor) = a_i \oti me s 1 + \s um_{j =1}^{2g } b _ { i,j }\o tim es \d elta_ j + f_i \o times \eta $$be the Künneth de compositio n o f $c_i(S^\lo r)$ . In t he ar g ot of [@ Mar], [@MO] , etc.the class es $$\beg i n{align e d } a_i & \in & {\operat o rn a me{H}}^{2i}({\ operat or n am e { Quo t}} ^{r,e} E)\\ b_{i , j} & \i n & { \ ope r atorname{H}}^ {2i-1}({\op e rat orname {Q uot } }^{r,e } E)\\ f_i & \in & {\ oper atorname{ H}}^{2 i -2}({\o p eratorn ame{Qu ot} }^{ r,e} E) \e nd{ al i gne d }$ $ a r e c alled $a ,b $and $ f$ * c l a s ses* .
A *t autol ogical class* is ane lem ent o f the sub ri ng ${ \overl ine}{ {\ operatorname{R} }}^* $ of ${\o per at orn am e{H}} ^ *({\op era tor name{Qu ot}}E)$ gen er a t e dby the $a$, $b$, a nd $ f$ classes . The* ta ut o logicalri ng* ${\ o p erato rnam e {R }}^*$ is the q u ot ie nt of $ {\ overli ne }{{ \op erato r name {R}}}^ *$ by th e ide a l $$\{ \alpha\ in {\overline } {{ \ o pe r ator nam e{R}}}^* :\int _ {[{\ oper a to rna m e{Quo t}}E] ^{ \ rm vir}} \alpha \beta=0 \; {\ rm fo r \; all} \; \beta \in { \ overline }{{\ o pe r atorname{R}}}^ * \}. $$
Unlike the actu al co homology ring ${\ o p eratorna me{ H}} ^*( {\o p e ra torname{Quot} } E )$,th e tauto log ical ri ngisdef orm at ion invar iant: it d oe sno t d epend on the c ur ve$C $ o r the bundle $E$, exc ep tt hro ugh the i rd i scre te i nvar ian ts . Thi s fo l low s fromdeformati on- i nvar ia nc e prope rties of thevi rtual clas sand the f a c t that t he $a$, $b$, and $f$ cl a sses ha vegloba llydefined a nal oguesas$ C,E$ v ary in fami li esw h ich r e s tr ict t o the give n cla ssesat poi nts ofthe family. The vi r tua l intersectio n t heor y of ${ \ op e rat or n ame { Q uot}}E$ is thestudy of t he ta utological rin g, usuall y throu gh ca l culatio ns of int egrals of t auto l o gic al classes over th e virtual class . A typi cal resul tisthe V afa-In t ril igato r form ul a forinteg ra ls of | . Let_$S$ be_the universal subbundle on_${\operatorname{Quot}}E \times_C$._Let $$c_i(S^\lor)_=_a_i \otimes 1_+ \sum_{j=1}^{2g} b_{i,j}_\otimes \delta_j + f_i_\otimes \eta$$ be_the_Künneth decomposition of $c_i(S^\lor)$. In the argot of [@Mar], [@MO], etc. the classes $$\begin{aligned}
a_i_&_\in &_{\operatorname{H}}^{2i}({\operatorname{Quot}}^{r,e}_E)_\\ b_{i,j} & \in &_{\operatorname{H}}^{2i-1}({\operatorname{Quot}}^{r,e} E) \\ f_i &_\in &_{\operatorname{H}}^{2i-2}({\operatorname{Quot}}^{r,e} E)\end{aligned}$$ are called $a,b$ and $f$ *classes*.
A_*tautological_class* is an_element of the subring ${\overline}{{\operatorname{R}}}^*$ of ${\operatorname{H}}^*({\operatorname{Quot}}E)$ generated by_the $a$, $b$, and $f$ classes._The *tautological ring*_${\operatorname{R}}^*$_is_the quotient of ${\overline}{{\operatorname{R}}}^*$_by the ideal $$\{ \alpha \in_{\overline}{{\operatorname{R}}}^* : \int_{[{\operatorname{Quot}}E]^{\rm vir}} \alpha \beta_=0 \; {\rm for \; all}_\; \beta \in {\overline}{{\operatorname{R}}}^* \}.$$
Unlike the_actual cohomology ring ${\operatorname{H}}^*({\operatorname{Quot}}E)$, the_tautological ring_is deformation invariant: it does_not depend on_the curve_$C$ or the_bundle $E$, except through their discrete_invariants. This follows_from deformation-invariance properties of the virtual_class_and the fact_that_the_$a$, $b$,_and $f$ classes_have_globally defined_analogues_as $C,E$ vary in families which_restrict_to the given classes at points of_the family. The virtual_intersection_theory of ${\operatorname{Quot}}E$ is_the study of the tautological_ring, usually through calculations of integrals_of tautological_classes over_the virtual class. A typical result is the Vafa-Intriligator formula for_integrals of |
$\FF_1=\mbox{Bl}_{\mathpzc{o}}(\PP^2)$. We denote by $h:R'\rightarrow \PP^1$ the map induced by the linear system $|\ell-E|$. The fibres of $h$ are pairwise disjoint lines in $\PP^8$. Equivalently, we consider the vector bundle on $\PP^1$ $$\G=\OO_{\PP^1}(3)\oplus \OO_{\PP^1}(4)$$ and then $R'\cong \PP(\G)$. One has the canonical identification between space of sections: $$H^0\bigl(R',\OO_{R'}(1)\bigr)\cong H^0\bigl(\PP(\G), \OO_{\PP(\G)}(1)\bigr)\cong H^0\bigl(\PP^1, \G\bigr).$$
Later, when computing the dimension of the parameter space of $3$-nodal septic scrolls, we shall make use of the basic fact $$\mbox{dim } \mbox{Aut}(R')=\mbox{dim } \mbox{Aut}(\FF_1)=6.$$ Every smooth septic scroll in $\PP^8$ is obtained from $R'$ by applying a linear transformation of $\PP^8$. In particular, the Hilbert scheme of septic scrolls in $\PP^8$ has dimension equal to $$\mbox{dim } PGL(9)-\mbox{dim } \mbox{Aut}(R')=80-6=74.$$
Using coordinates in $\PP^8$, if $\PP^3_{y_0, \ldots, y_3}\subseteq \PP^8$ is the linear span of the twisted cubic $E$ corresponding to the exceptional divisor on $\FF_1$ and $\PP^4_{x_0,\ldots, x_4}\subseteq \PP^8$ is the linear span of a rational quartic curve linearly equivalent to $\ell$, then the ideal of $R'$ in $\PP^8$ is given by the following determinantal condition, see for instance [@Ha] Lecture 9: $$\mbox{rk}\begin{pmatrix}
x_0 & x_1& x_2& x_3 & y_0 & y_1 &y_2\\
x_1 & x_2 & x_3& x_4& y_1 &y_2 &y_3 \\
\end{pmatrix}
\leq 1.$$
The sec | $ \FF_1=\mbox{Bl}_{\mathpzc{o}}(\PP^2)$. We denote by $ h: R'\rightarrow \PP^1 $ the map induced by the analogue arrangement $ |\ell - E|$. The fibres of $ h$ are pairwise disjoint lines in $ \PP^8$. Equivalently, we think the vector pile on $ \PP^1 $ $ $ \G=\OO_{\PP^1}(3)\oplus \OO_{\PP^1}(4)$$ and then $ R'\cong \PP(\G)$. One has the canonical recognition between distance of sections: $ $ H^0\bigl(R',\OO_{R'}(1)\bigr)\cong H^0\bigl(\PP(\G), \OO_{\PP(\G)}(1)\bigr)\cong H^0\bigl(\PP^1, \G\bigr).$$
Later, when computing the dimension of the argument space of $ 3$-nodal septic scrolls, we shall make use of the basic fact $ $ \mbox{dim } \mbox{Aut}(R')=\mbox{dim } \mbox{Aut}(\FF_1)=6.$$ Every legato septic scroll in $ \PP^8 $ is obtained from $ R'$ by applying a linear transformation of $ \PP^8$. In particular, the Hilbert dodge of septic scrolls in $ \PP^8 $ has dimension adequate to $ $ \mbox{dim } PGL(9)-\mbox{dim } \mbox{Aut}(R')=80 - 6=74.$$
Using coordinates in $ \PP^8 $, if $ \PP^3_{y_0, \ldots, y_3}\subseteq \PP^8 $ is the analogue span of the writhe cubic $ E$ corresponding to the exceptional divisor on $ \FF_1 $ and $ \PP^4_{x_0,\ldots, x_4}\subseteq \PP^8 $ is the linear span of a rational quartic curvature linearly equivalent to $ \ell$, then the ideal of $ R'$ in $ \PP^8 $ is given by the following determinantal condition, see for instance [ @Ha ] Lecture 9: $ $ \mbox{rk}\begin{pmatrix }
x_0 & x_1 & x_2 & x_3 & y_0 & y_1 & y_2\\
x_1 & x_2 & x_3 & x_4 & y_1 & y_2 & y_3 \\
\end{pmatrix }
\leq 1.$$
The sec | $\FF_1=\lbox{Bl}_{\mathpzc{o}}(\PP^2)$. We denoue by $h:R'\rightarroc \PP^1$ thx map ihduced bh the linear system $|\ell-E|$. The fubres of $h$ are pairwise disgoint linvs in $\PP^8$. Wquitalently, we consmser the vectod bunblx on $\PP^1$ $$\G=\OO_{\PP^1}(3)\opkus \OO_{\PP^1}(4)$$ atd then $R'\cong \[P(\E)$. Lne has the canonical identificatiog betwern space of sectyons: $$R^0\bigm(G',\OI_{R'}(1)\bigr)\cong H^0\bigl(\PP(\G), \OO_{\PP(\G)}(1)\bigr)\dong H^0\bpgl(\PP^1, \G\bigr).$$
Later, ehen computing the dimensiln ov the parameter spwce of $3$-nodao se[ric scrolls, de shall make use of tge basic fact $$\mbox{dim } \mbox{Aut}(R')=\obox{dnm } \mbox{Aut}(\DF_1)=6.$$ Evfty smooth se'tic sbroll in $\PP^8$ is obtainad from $R'$ by applying a ninwar transformation of $\PP^8$. In particular, jhe Hilberd acheme of septic wceolls in $\[P^8$ hxw dkmehsmon equal to $$\mbox{dim } LGL(9)-\mbox{dim } \mbox{Aut}(R')=80-6=74.$$
Using coorcigqtes in $\PP^8$, if $\PP^3_{y_0, \lqoes, y_3}\subseteq \PP^8$ is the linear span of tve fwisted cubic $E$ correspinding to the exceptilnal divifor on $\FF_1$ and $\PP^4_{x_0,\ldots, x_4}\subseteq \PP^8$ is the linear span uf c ratiubap quartic curve linearly equivalent to $\ell$, thqh uhe ideal of $R'$ in $\PP^8$ is givem hy jhe following aetermnhahtal condition, see for inftancw [@Ha] Lecttre 9: $$\mbox{rk}\begin{pmatrix}
x_0 & x_1& x_2& x_3 & y_0 & y_1 &y_2\\
x_1 & x_2 & x_3& x_4& y_1 &y_2 &y_3 \\
\end{pmatxix}
\leq 1.$$
The szc | $\FF_1=\mbox{Bl}_{\mathpzc{o}}(\PP^2)$. We denote by $h:R'\rightarrow \PP^1$ the by linear system The fibres of in Equivalently, we consider vector bundle on $$\G=\OO_{\PP^1}(3)\oplus \OO_{\PP^1}(4)$$ and then $R'\cong \PP(\G)$. has the canonical identification between space of sections: $$H^0\bigl(R',\OO_{R'}(1)\bigr)\cong H^0\bigl(\PP(\G), \OO_{\PP(\G)}(1)\bigr)\cong H^0\bigl(\PP^1, \G\bigr).$$ when computing the dimension of the parameter space of $3$-nodal septic scrolls, we make of basic $$\mbox{dim } \mbox{Aut}(R')=\mbox{dim } \mbox{Aut}(\FF_1)=6.$$ Every smooth septic scroll in $\PP^8$ is obtained from $R'$ by a linear transformation of $\PP^8$. In particular, the scheme of septic scrolls $\PP^8$ has dimension equal to } } \mbox{Aut}(R')=80-6=74.$$ coordinates $\PP^8$, $\PP^3_{y_0, \ldots, y_3}\subseteq is the linear span of the twisted cubic $E$ corresponding to the exceptional divisor on $\FF_1$ and x_4}\subseteq \PP^8$ linear span a quartic linearly equivalent to the ideal of $R'$ in $\PP^8$ the following determinantal condition, see for instance [@Ha] 9: $$\mbox{rk}\begin{pmatrix} & x_1& x_2& x_3 & y_0 y_1 &y_2\\ x_1 & x_2 & x_3& x_4& &y_2 &y_3 \\ \end{pmatrix} \leq 1.$$ The sec | $\FF_1=\mbox{Bl}_{\mathpzc{o}}(\PP^2)$. We denotE by $h:R'\rightArrow \pP^1$ tHe mAp InduCed bY the linear systEM $|\ell-e|$. The fibres of $h$ are pairwiSe disJoINt liNEs In $\PP^8$. EQuivaleNTlY, WE coNsIdEr tHe VEcTor buNdlE on $\PP^1$ $$\G=\Oo_{\PP^1}(3)\oplus \OO_{\pP^1}(4)$$ aNd Then $R'\cong \PP(\G)$. oNe Has the canoNicAl identificaTioN betweEn SpaCE of seCtiOns: $$H^0\bIgl(R',\OO_{r'}(1)\Bigr)\coNg H^0\bigl(\PP(\g), \Oo_{\pP(\G)}(1)\bigR)\Cong H^0\biGL(\pP^1, \g\bigR).$$
Later, when computiNG tHE dimension of thE paramEtER sPACe oF $3$-noDal septic sCrOlls, wE Shall maKE uSE OF thE Basic fact $$\mbox{Dim } \mbox{Aut}(R')=\MBox{Dim } \mboX{AUt}(\Ff_1)=6.$$ every sMooth SePTic Scroll in $\PP^8$ iS obtAined from $r'$ by appLYing a liNEar tranSformaTioN of $\pP^8$. In PArTiCulAr, THe HILbErt SCheMe of septIc ScRolls In $\PP^8$ HAS DImenSioN equAl to $$\mBox{dim } PGL(9)-\mbox{Dim } \Mbox{aUt}(R')=80-6=74.$$
using CoordInatEs In $\PP^8$, iF $\PP^3_{y_0, \ldOts, y_3}\sUbSeteq \PP^8$ is the linEar sPan of the tWisTeD cuBiC $E$ corREspondIng To tHe excepTional dIVisOr ON $\ff_1$ aNd $\PP^4_{x_0,\ldots, x_4}\subseteQ \Pp^8$ IS tHe linear Span of A RaTiONal quartIc CurVe liNEArly eQuivALeNt to $\ell$, tHen the IDeAl Of $R'$ in $\PP^8$ Is Given bY tHe fOllOwing DEterMinantAl conditIon, seE For instance [@Ha] LECture 9: $$\mbox{rk}\beGIn{PMAtRIx}
x_0 & x_1& X_2& x_3 & y_0 & Y_1 &y_2\\
x_1 & x_2 & x_3& x_4& y_1 &y_2 &y_3 \\
\enD{pmaTRix}
\lEq 1.$$
ThE SeC | $\FF_1=\mbox{Bl}_{\mathpz c{o}}(\PP^ 2)$.Weden ot e by $h: R'\rightarrow\ PP^1 $ the map induced by t he li ne a r sy s te m $|\ ell-E|$ . T h e fi br es of $ h $are p air wise di sjoint lin esin $\PP^8$. Eq u iv alently, w e c onsider thevec tor bu nd leo n $\P P^1 $ $$\ G=\OO_ { \PP^1} (3)\oplus \ O O_{\PP ^ 1}(4)$$ a nd the n $R'\cong \PP(\G ) $. One has the ca nonica li de n t ifi cat ion betwee nspace of sect i on s : $$H ^ 0\bigl(R',\OO _{R'}(1)\bi g r)\ cong H ^0 \bi g l(\PP( \G),\O O _{\ PP(\G)}(1)\ bigr )\cong H^ 0\bigl ( \PP^1,\ G\bigr) .$$
L ate r,when co mp uti ng the di men s ion of thepa ra meter spa c e o f $3 $-n odal sept ic scrolls, w e s hall mak e use of t he b as ic fa ct $$\ mbox{ di m } \mbox{Aut}( R')= \mbox{dim }\m box {A ut}(\ F F_1)=6 .$$ Ev ery smo oth sep t icsc r o l lin $\PP^8$ is obta in e d f rom $R'$ by ap p ly in g a linea rtra nsfo r m ation of$ \P P^8$. In parti c ul ar , the H il bert s ch eme of sept i c sc rollsin $\PP^ 8$ ha s dimension equ a l to $$\mbox{ d im } P G L(9) -\m box{dim } \ mbox { Aut} (R') = 80 -6= 7 4.$$
Usin gc oo r dinates in $\PP^8$, i f $\PP ^3_{y _0, \ldots, y _3}\subset e q \PP^8$ i s th e l i near span of t he tw isted cubi c $E$ cor respo nding to the exce p t ional di vis oron$\F F _ 1$ and $\PP^4_{ x _ 0,\l do ts, x_4 }\s ubseteq \P P^8 $ i s t he linear s pan of a r at io na l q uarti c curve l in ear ly eq uival e nt to$\ell $, t he nt heideal o f $ R ' $ in $ \P P^8$ is g ivenby t h e f ollowin g determi nan t al c on di tion, s ee for instan ce [@Ha] Lec tu re9: $$\ m b ox{rk}\b egin{pmatrix}
x_0 & x_1 & x_2& x _3& y_0 & y _1 &y_2\\
x_ 1 & x_ 2 & x_3& x _4& y_ 1 &y_ 2&y_ 3 \\
\e n d {p mat ri x}
\leq 1. $ $
T he se c | $\FF_1=\mbox{Bl}_{\mathpzc{o}}(\PP^2)$._We denote_by $h:R'\rightarrow \PP^1$ the_map induced_by_the linear_system_$|\ell-E|$. The fibres_of $h$ are_pairwise disjoint lines in_$\PP^8$. Equivalently, we_consider_the vector bundle on $\PP^1$ $$\G=\OO_{\PP^1}(3)\oplus \OO_{\PP^1}(4)$$ and then $R'\cong \PP(\G)$. One has the_canonical_identification between_space_of_sections: $$H^0\bigl(R',\OO_{R'}(1)\bigr)\cong H^0\bigl(\PP(\G), \OO_{\PP(\G)}(1)\bigr)\cong H^0\bigl(\PP^1,_\G\bigr).$$
Later, when computing the dimension_of the_parameter space of $3$-nodal septic scrolls, we shall_make_use of the_basic fact $$\mbox{dim } \mbox{Aut}(R')=\mbox{dim } \mbox{Aut}(\FF_1)=6.$$ Every smooth_septic scroll in $\PP^8$ is obtained_from $R'$ by_applying_a_linear transformation of $\PP^8$._In particular, the Hilbert scheme of_septic scrolls in $\PP^8$ has dimension_equal to $$\mbox{dim } PGL(9)-\mbox{dim } \mbox{Aut}(R')=80-6=74.$$
Using_coordinates in $\PP^8$, if $\PP^3_{y_0, \ldots,_y_3}\subseteq \PP^8$ is the linear_span of_the twisted cubic $E$ corresponding_to the exceptional_divisor on_$\FF_1$ and $\PP^4_{x_0,\ldots,_x_4}\subseteq \PP^8$ is the linear span_of a rational_quartic curve linearly equivalent to $\ell$,_then_the ideal of_$R'$_in_$\PP^8$ is_given by the_following_determinantal condition,_see_for instance [@Ha] Lecture 9: $$\mbox{rk}\begin{pmatrix}
x_0_&_x_1& x_2& x_3 & y_0 & y_1_&y_2\\
x_1 & x_2 &_x_3&_x_4& y_1 &y_2 &y_3_\\
\end{pmatrix}
\leq 1.$$
The sec |
for this paper. In this experiment, it is assumed that the machine and human share a common goal $\bm{q_d}$. Machine error $\Delta_e$ is introduced into the system as an automation error due to environment uncertainty. To keep hardware and sensor costs practical, knowledge of the behavior of this error is used to determine when to incorporate human intervention, e.g. when an error $\Delta_e$ in the environment is high, and as such the automation cannot complete a task, human adaptability can be incorporated in the control.
The blended input trajectory $\bm{q_{ref}}$ is passed through the Inverse Kinematics (IK) of the robot model, using a damped least squares method [@Wampler1986], to generate a set of desired joint angles $\bm{\theta}$. These desired joint angles $\bm{\theta}$ are then passed through a nonlinear dynamics simulation of the torque-controlled robot, as shown in Fig. \[fig:dynamics\], to generate the joint commands $\bm{\theta_{ref}}$. This simulation tends to filter high frequency commands, and provides a means to generate a set of feasible joint commands $\bm{\theta_{ref}}$. Closed-loop motor control to achieve the reference trajectory $\bm{\theta_{ref}}$ is handled by the software provided with the robot. When automation uncertainties are present in the system, the human input $\bm{q_h}$ provides closed-loop control based on visual feedback $\mathcal{V}$ to correct for errors $\Delta_e$ in the estimated goal location $\hat{\bm{q_d}}$.
Shared Control
--------------
The human input $\bm{q_h}$ is applied through a haptic input device. The machine input $\bm{q_m}$ is queried from the path planning module in Fig. \[fig:machine\]. An autonomy selection policy defines the level of autonomy $\alpha$. Fig. \[fig:trajectory\] shows how the reference trajectory changes as the level of autonomy (LOA) slides from a fully autonomous mode to a tele-operation mode.
Autonomy Selection Policy {#sec:autonomySelection}
-------------------------
Previous works [@Dragan2013] have used the confidence interval in a machine prediction task as the level of autonomy $\alpha$, e.g. $$\label{eqn:dragan}
\alpha = \max \left(0, 1 - \frac{d}{D}\right),$$ where $d | for this paper. In this experiment, it is assumed that the machine and human share a coarse finish $ \bm{q_d}$. Machine error $ \Delta_e$ is introduced into the arrangement as an automation erroneousness due to environment uncertainty. To keep hardware and detector costs practical, knowledge of the demeanor of this mistake is used to settle when to incorporate human intervention, for example when an error $ \Delta_e$ in the environment is high, and as such the automation cannot dispatch a task, human adaptability can be incorporate in the control condition.
The blended input trajectory $ \bm{q_{ref}}$ is passed through the Inverse Kinematics (IK) of the automaton model, using a muffle least squares method acting [ @Wampler1986 ], to generate a bent of desired joint angles $ \bm{\theta}$. These desired joint slant $ \bm{\theta}$ are then passed through a nonlinear dynamics simulation of the torque - controlled automaton, as read in Fig. \[fig: dynamics\ ], to generate the joint commands $ \bm{\theta_{ref}}$. This simulation tends to filter high frequency commands, and provides a means to render a set of feasible joint commands $ \bm{\theta_{ref}}$. shut - loop topology motor command to achieve the reference trajectory $ \bm{\theta_{ref}}$ is handled by the software provide with the robot. When automation uncertainties are present in the system, the human input $ \bm{q_h}$ provide closed - loop control based on visual feedback $ \mathcal{V}$ to correct for error $ \Delta_e$ in the estimated goal location $ \hat{\bm{q_d}}$.
Shared Control
--------------
The human input $ \bm{q_h}$ is apply through a haptic input signal device. The machine input $ \bm{q_m}$ is queried from the path planning module in Fig. \[fig: machine\ ]. An autonomy survival policy defines the level of autonomy $ \alpha$. Fig. \[fig: trajectory\ ] shows how the reference point trajectory changes as the level of autonomy (LOA) slides from a fully autonomous modality to a tele - operation mode.
Autonomy Selection Policy { # sec: autonomySelection }
-------------------------
former works [ @Dragan2013 ] have used the confidence interval in a car prediction task as the level of autonomy $ \alpha$, for example $ $ \label{eqn: dragan }
\alpha = \max \left(0, 1 - \frac{d}{D}\right),$$ where $ d | fog this paper. In this exptriment, it is assomwd thav the mzchine avd human share a common goal $\bn{q_d}$. Mqchine error $\Delta_e$ is introducvd into tye sbstem as an automation evxor dhc to zntironment uncerjainty. To keap hardware ang reusor costs practical, knowledge of thq behavoog of this errot is lsqd tk determine when to incorporate hujan inttrvention, e.g. when sn error $\Delta_e$ in the envlronlent is high, and ad such the qutoiqtion cannot complete a task, human adaptability can be incorporatdd in the contril.
Rhe tlended inpnt tratectory $\bm{q_{rcg}}$ is pdssed tnrough the Invcrse Ninwmatics (IK) of the robmt model, using a dwmped leavt squares method [@Wanpoer1986], tm geteragw a seu oh dssired jomnt angles $\gm{\theta}$. Thewe desired joint anblqw $\bm{\theta}$ are then [afsed through a nonlinear dynamics simuldtikn of the torque-controloed robot, as shown in Fig. \[fig:dygamics\], to generate the joint commands $\bm{\theta_{ref}}$. Dhis akmuoabion rejds to filter high frequency commands, and prodjdts s means to gencrate a set of feaxihlr joint commanas $\bm{\tkstz_{ref}}$. Closed-loop mohor conjrol ti achieve the reference trajectory $\bm{\thera_{ref}}$ is hanblee by the software 'rovided witk the tobot. Ehen automation uncertannties are presenh in the ahstem, the human knplt $\bk{q_h}$ provides closed-loop cogtrol basxd on visual feecback $\iathcal{V}$ tl corvact for errors $\Delha_e$ iu the estimated goal location $\hat{\bm{q_d}}$.
Shared Coivrol
--------------
The human itpun $\bm{q_h}$ is applled through a hwptic input derice. The machkne input $\gm{q_m}$ is queried frjm the path pnwnning modulx in Fig. \[fyg:maxhinw\]. An augunomy selectiom policy btfines the level of autonomy $\alkhz$. Fig. \[fig:trajectiry\] shows how the tefdregcv tcajecemry changes ds tfe urvel uf autonomy (UOA) xlides from a fully dutohomous mode to a trlc-operatiob mode.
Aueonomy Selectoon Policy {#sec:autojomySxlectiun}
-------------------------
Prefioos works [@Dragan2013] have used the cohfidence lntcrval in a mashinc prgdiction taxk as the level of autonomy $\alpha$, e.g. $$\lauel{eqn:dragan}
\alpha = \nax \left(0, 1 - \frac{d}{D}\riyhu),$$ where $d | for this paper. In this experiment, it that machine and share a common is into the system an automation error to environment uncertainty. To keep hardware sensor costs practical, knowledge of the behavior of this error is used to when to incorporate human intervention, e.g. when an error $\Delta_e$ in the environment high, as the cannot complete a task, human adaptability can be incorporated in the control. The blended input trajectory is passed through the Inverse Kinematics (IK) of robot model, using a least squares method [@Wampler1986], to a of desired angles These joint angles $\bm{\theta}$ then passed through a nonlinear dynamics simulation of the torque-controlled robot, as shown in Fig. \[fig:dynamics\], to the joint This simulation to high commands, and provides to generate a set of feasible Closed-loop motor control to achieve the reference trajectory is handled the software provided with the robot. automation uncertainties are present in the system, the input $\bm{q_h}$ provides closed-loop control based on visual feedback $\mathcal{V}$ to correct for errors $\Delta_e$ estimated goal location $\hat{\bm{q_d}}$. Control -------------- The input is through haptic input The machine input $\bm{q_m}$ is queried from the path planning module Fig. \[fig:machine\]. An autonomy selection policy defines the level of Fig. shows how the trajectory changes as the of (LOA) slides from a mode a Selection {#sec:autonomySelection} Previous works [@Dragan2013] have the confidence interval in a prediction task as the $$\label{eqn:dragan} \alpha = \max \left(0, 1 - \frac{d}{D}\right),$$ $d | for this paper. In this experimEnt, it is assUmed tHat The MaChinE and Human share a comMOn goAl $\bm{q_d}$. Machine error $\DeltA_e$ is iNtROducED iNto thE system AS aN AUtoMaTiOn eRrOR dUe to eNviRonment UncertaintY. To KeEp hardware anD SeNsor costs pRacTical, knowledGe oF the beHaVioR Of thiS erRor is Used to DEtermiNe when to iNcORporatE Human inTERvEntiOn, e.g. when an error $\DeLTa_E$ In the environmeNt is hiGh, ANd AS SucH thE automatioN cAnnot COmplete A TaSK, HUmaN Adaptability cAn be incorpoRAteD in the CoNtrOL.
The blEnded InPUt tRajectory $\bm{Q_{ref}}$ Is passed tHrough THe InverSE KinemaTics (IK) Of tHe rObot MOdEl, UsiNg A DamPEd LeaST sqUares metHoD [@WAmpleR1986], to gENERAte a Set Of deSired Joint angles $\bm{\TheTa}$. ThESe dEsireD joinT angLeS $\bm{\thEta}$ are Then pAsSed through a nonlIneaR dynamics SimUlAtiOn Of the TOrque-cOntRolLed roboT, as showN In FIg. \[FIG:DyNamics\], to generate thE jOINt Commands $\Bm{\thetA_{ReF}}$. THIs simulaTiOn tEnds TO FilteR higH FrEquency cOmmandS, AnD pRovides A mEans to GeNerAte A set oF FeasIble joInt commaNds $\bm{\THeta_{ref}}$. Closed-lOOp motor controL To ACHiEVe thE reFerence trajEctoRY $\bm{\tHeta_{REf}}$ Is hANdled By the SoFTwARe provided with the roBoT. When aUtomaTion uncertainTies are preSENT in the syStem, THe HUman input $\bm{q_h}$ pRovidEs closed-loOP control Based On visual Feedback $\mATHcal{V}$ to cOrrEct For ErrORS $\DElta_e$ in the estIMAted GoAl locatIon $\Hat{\bm{q_d}}$.
shaRed conTroL
--------------
THe human inPut $\bm{q_h}$ iS aPpLiEd ThrOugh a HAptic inpUt DevIcE. ThE machINe inpuT $\bm{q_m}$ Is quErIeD FroM the patH PlANNing MoDuLe in fig. \[FiG:machIne\]. AN AutOnomy seLection poLicY DefiNeS tHe level Of autonomy $\alpHa$. fig. \[fig:trajEcTorY\] shows HOW the refeRence trajectory changes aS The leveL of AutonOmy (LoA) slides fRom A fully AutONomous Mode to A tele-OpEraTIOn modE.
aUtOnoMy selection POLIcy {#Sec:auToNomyselectiOn}
-------------------------
Previous works [@DraGAn2013] hAve used the conFidEnce INTeRvaL In A MacHiNE prEDIction task as the Level of autOnOMy $\Alpha$, e.g. $$\labEL{eqN:dRagan}
\alPha = \max \lEft(0, 1 - \frAC{d}{D}\righT),$$ where $d | for this paper. In this e xperiment, it i s a ssu me d th at t he machine and huma n share a common goal$\bm{ q_ d }$.M ac hineerror $ \ De l t a_e $is in tr o du ced i nto the sy stem as an au to mation error du e to envir onm ent uncertai nty . To k ee p h a rdwar e a nd se nsor c o sts pr actical,kn o wledge of theb e ha vior of this error is us e d to determine whento in c o rpo rat e human in te rvent i on, e.g . w h e n an error $\Delta _e$ in thee nvi ronmen tish igh, a nd as s u chthe automat ioncannot co mplete a task, human a daptab ili tycanb ein cor po r ate d i n t h e c ontrol.
T he blen dedi n p u t tr aje ctor y $\b m{q_{ref}}$ i s p asse d th rough theInve rs e Kin ematic s (IK )of the robot mo del, using adam pe d l ea st sq u ares m eth od[@Wampl er1986] , to g e n e ra te a set of desire dj o in t angles $\bm{ \ th et a }$. Thes edes ired j ointangl e s$\bm{\th eta}$a re t hen pas se d thro ug h a no nline a r dy namics simulat ion o f the torque-co n trolled robot , a s sh o wn i n F ig. \[fig:d ynam i cs\] , to ge ner a te th e joi nt co m mands $\bm{\theta_{ re f}}$.Thissimulation te nds to fil t e r high fr eque n cy commands, andprovi des a mean s to gene ratea set of feasible j oint com man ds$\b m{\ t h et a_{ref}}$. Cl o s ed-l oo p motor co ntrol t o a chi eve th ereference traject or y$\ bm {\t heta_ { ref}}$ i shan dl edby th e softw are p rovi de dw ith the ro b ot . When a ut omat ion u ncert aint i esare pre sent in t hes yste m, t he huma n input $\bm{ q_ h}$ provid es cl osed-l o o p contro l based on visual feedb a ck $\ma thc al{V} $ to correctfor error s $ \ Delta_ e$ inthe e st ima t e d goa l lo cat io n $\hat{\b m { q_d }}$.
S hare d Contr ol
--------------The human input$\b m{q_ h } $isa pp l ied t h rou g h a haptic input device. T he ma chine inpu t $\ bm {q_m}$is quer ied f r om thepath plan ning modu le inF i g.\[fig:mach ine\]. A n autonom y sele c ti on po lic y defi ne s t he le vel of aut onomy $\alp ha $. Fig . \[f ig :traject ory\] shows how the ref erence traj ect ory chang esa s t he levelof a utonomy (L OA) sl idesfro m a fu llya ut ono m ous m odet o a tele- o pe rat i o nmode.
Auto n o m y S elect ion Policy {#s ec:autonomySelect i on}
---------- ---- - - --- --- - --
Pr evious works [ @Dr ag a n 2013] ha ve used the c onfidenc ei nterv al ina mach ine pre d i ct i on tas k as th e level o f a ut o nomy $\ al ph a $, e.g . $$ \l abel{e qn:dra g an} \alpha = \max\left ( 0 , 1 - \fr ac{d} {D }\right ) ,$$where $d | for_this paper._In this experiment, it_is assumed_that_the machine_and_human share a_common goal $\bm{q_d}$._Machine error $\Delta_e$ is_introduced into the_system_as an automation error due to environment uncertainty. To keep hardware and sensor costs_practical,_knowledge of_the_behavior_of this error is used_to determine when to incorporate_human intervention,_e.g. when an error $\Delta_e$ in the environment_is_high, and as_such the automation cannot complete a task, human adaptability_can be incorporated in the control.
The_blended input trajectory_$\bm{q_{ref}}$_is_passed through the Inverse_Kinematics (IK) of the robot model,_using a damped least squares method [@Wampler1986],_to generate a set of desired joint_angles $\bm{\theta}$. These desired joint angles_$\bm{\theta}$ are then passed through_a nonlinear_dynamics simulation of the torque-controlled_robot, as shown_in Fig. \[fig:dynamics\],_to generate the_joint commands $\bm{\theta_{ref}}$. This simulation tends_to filter high_frequency commands, and provides a means_to_generate a set_of_feasible_joint commands_$\bm{\theta_{ref}}$. Closed-loop motor_control_to achieve_the_reference trajectory $\bm{\theta_{ref}}$ is handled by_the_software provided with the robot. When automation_uncertainties are present in_the_system, the human input_$\bm{q_h}$ provides closed-loop control based_on visual feedback $\mathcal{V}$ to correct_for errors_$\Delta_e$ in_the estimated goal location $\hat{\bm{q_d}}$.
Shared Control
--------------
The human input $\bm{q_h}$ is applied_through a haptic input device. The_machine input $\bm{q_m}$ is_queried from_the_path planning module_in_Fig. \[fig:machine\]. An_autonomy selection policy defines the level of_autonomy $\alpha$._Fig. \[fig:trajectory\] shows how the reference trajectory_changes as the level_of_autonomy (LOA) slides from a fully_autonomous mode to a tele-operation mode.
Autonomy_Selection Policy {#sec:autonomySelection}
-------------------------
Previous works [@Dragan2013] have_used_the_confidence interval in a machine_prediction task as the level of_autonomy $\alpha$, e.g._$$\label{eqn:dragan}
\alpha = \max_\left(0,_1 - \frac{d}{D}\right),$$ where $d |
rigid, for every $N>4$. We use our results to give an easy and uniform proof of birational superrigidity for arbitrary smooth hypersurfaces of degree $N$ in $\P^N$ when $N$ is small.
\[thm3\_introd\] If $X\subset{\mathbb P}^N$ is a smooth hypersurface of degree $N$, and if $4\leq N\leq 12$, then $X$ is birationally superrigid.
Based on previous ideas of Corti, Pukhlikov proposed in [@Pu1] a proof of the birational rigidity of every smooth hypersurface of degree $N$ in $\P^N$, for $N\geq 6$. Unfortunately, at the moment there is a gap in his arguments (see Remark \[gap\] below). Despite this gap, the proof proposed in [@Pu1] contains many remarkable ideas, and it seems likely that a complete proof could be obtained in the future along those lines. In fact, the outline of the proof of Theorem \[thm3\_introd\] follows his method, and our contribution is mainly to simplifying and solidifying his argument.
Acknowledgements {#acknowledgements.unnumbered}
----------------
We are grateful to Steve Kleiman and Rob Lazarsfeld for useful discussions. Research of the first author was partially supported by MURST of Italian Government, National Research Project (Cofin 2000) “Geometry of Algebraic Varieties”. Research of the second author was partially supported by NSF Grant DMS 02-00278. The third author served as a Clay Mathematics Institute Long-Term Prize Fellow while this research has been done.
Singularities of log pairs under projections
============================================
Let $X$ be a smooth algebraic variety, defined over an algebraically closed field of characteristic zero, and let $V \subset X$ be a proper subscheme. For any rational number $c > 0$, we can consider the pair $(X,c\. V)$. The usual definitions in the theory of singularities of pairs, for which we refer to [@Ko], extend to this context. In particular, we say that an irreducible subvariety $C \subset X$ is a center of non log canonicity (resp. non log terminality, non canonicity, non terminality) for $(X,c\. V)$ if there is at least one divisorial valuation of $K(X)$, with | rigid, for every $ N>4$. We use our results to give an comfortable and consistent proof of birational superrigidity for arbitrary smooth hypersurfaces of academic degree $ N$ in $ \P^N$ when $ N$ is small.
\[thm3\_introd\ ] If $ X\subset{\mathbb P}^N$ is a placid hypersurface of degree $ N$, and if $ 4\leq N\leq 12 $, then $ X$ is birationally superrigid.
Based on former estimate of Corti, Pukhlikov proposed in [ @Pu1 ] a validation of the birational inflexibility of every smooth hypersurface of degree $ N$ in $ \P^N$, for $ N\geq 6$. Unfortunately, at the consequence there is a gap in his arguments (determine Remark \[gap\ ] below). Despite this gap, the proof proposed in [ @Pu1 ] contains many noteworthy ideas, and it seems likely that a arrant proof could be obtained in the future along those line. In fact, the outline of the proof of Theorem \[thm3\_introd\ ] come his method, and our contribution is mainly to simplifying and solidifying his argument.
recognition { # acknowledgements.unnumbered }
----------------
We are grateful to Steve Kleiman and Rob Lazarsfeld for useful discussions. Research of the first author was partially supported by MURST of Italian Government, National Research Project (Cofin 2000) “ Geometry of Algebraic Varieties ”. Research of the second author was partially supported by NSF Grant DMS 02 - 00278. The third author served as a Clay Mathematics Institute Long - Term Prize Fellow while this research has been done.
Singularities of log pairs under projections
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Let $ X$ be a politic algebraic variety, defined over an algebraically close sphere of characteristic zero, and let $ V \subset X$ be a proper subscheme. For any rational number $ c > 0 $, we can consider the pair $ (ten, c\. V)$. The usual definitions in the theory of singularities of pairs, for which we refer to [ @Ko ], cover to this context. In particular, we say that an irreducible subvariety $ C \subset X$ is a center of non log canonicity (resp. non log terminality, non canonicity, non terminality) for $ (X, c\. V)$ if there cost at least one divisorial valuation of $ K(X)$, with | rigld, for every $N>4$. We use ouv results to givg qn easb and uhiform pfoof of birational superrigivity for qrbitrary smooth hyperrurfaces lf degrew $N$ mn $\P^N$ when $N$ is small.
\[thm3\_lutrod\] Lf $X\sbbwet{\mathbb P}^N$ ix a smooth hypersurface mf dzgree $N$, and if $4\leq N\leq 12$, then $X$ is byrationslpy superrigid.
Bwsed jn pdvvlous ideas of Corti, Pukhlikov prkposed pn [@Pu1] a proof of yhe birational rigidity of evegy smooth hypersurvace of degtse $G$ in $\P^N$, for $N\eeq 6$. Unforuuuately, at tge moment there is a gap in his argukents (see Tznarn \[cap\] below). Dxspite this gap, thc proof proposrd in [@Pu1] contalns meny eemarkable ideas, and mt seems likely that a complede proof could be ovtqined in dhe wytufe zlpnf thosf lmnes. In facf, the outlibe of the proof of Uhejgrm \[thm3\_introd\] rollowf ris method, and our contribution is mainny fo simplifying and solieifying his argument.
Afknowledgqments {#acknowledgements.unnumbered}
----------------
We are grateful do Stxvd Koelman qnf Rob Lazarsfeld for useful discussions. Reseawdh on the first authjr was partoaplu supported by MURST of Italian Governmenh, Natiogal Rwsearch Pwojevt (Cofin 2000) “Geometry of Algebeaic Varietivs”. Rwsearch of the seclnd author cas pattiallu supported by NSF Grant DMS 02-00278. The third wuthor sedxed as a Clay Maghekadics Insuktute Long-Term Pryze Fellox whike this resgarch hws been doje.
Sinnglarities of log pwirs onder [rojectiond
============================================
Let $X$ be a smooth algebraic vacmety, defined pvar dn algebxaicalky closed fiejd of charactetistic zexo, and let $V \subaet X$ bx a proper stbscheme. For djy rational iumber $c > 0$, we can considdf the pair $(X,c\. F)$. The usucj dedinitions in the tmeory kf singularitiew od pairs, for whivh de wevec to [@Hm], extend to dhis covyext. Kn particulcr, we xay that an irreducitle aubvariety $C \subsey W$ is a cebter of gon log canonocity (resp. non log termmnalitb, non vanjnicity, non terminality) for $(X,c\. V)$ if thege ls at least oge dlvisjrial valuction of $K(X)$, with | rigid, for every $N>4$. We use our give easy and proof of birational of $N$ in $\P^N$ $N$ is small. If $X\subset{\mathbb P}^N$ is a smooth of degree $N$, and if $4\leq N\leq 12$, then $X$ is birationally superrigid. on previous ideas of Corti, Pukhlikov proposed in [@Pu1] a proof of the rigidity every hypersurface degree $N$ in $\P^N$, for $N\geq 6$. Unfortunately, at the moment there is a gap in arguments (see Remark \[gap\] below). Despite this gap, proof proposed in [@Pu1] many remarkable ideas, and it likely a complete could obtained the future along lines. In fact, the outline of the proof of Theorem \[thm3\_introd\] follows his method, and our contribution mainly to solidifying his Acknowledgements ---------------- are grateful to and Rob Lazarsfeld for useful discussions. first author was partially supported by MURST of Government, National Project (Cofin 2000) “Geometry of Algebraic Research of the second author was partially supported NSF Grant DMS 02-00278. The third author served as a Clay Mathematics Institute Long-Term Prize this research has been Singularities of log under ============================================ $X$ a smooth variety, defined over an algebraically closed field of characteristic zero, and $V \subset X$ be a proper subscheme. For any rational > we can consider pair $(X,c\. V)$. The definitions the theory of singularities for we extend this In particular, we say an irreducible subvariety $C \subset is a center of log terminality, non canonicity, non terminality) for $(X,c\. if there is at least one divisorial of $K(X)$, with | rigid, for every $N>4$. We use our resUlts to give An easY anD unIfOrm pRoof Of birational suPErriGidity for arbitrary smooTh hypErSUrfaCEs Of degRee $N$ in $\P^n$ WhEN $n$ is SmAlL.
\[thM3\_iNTrOd\] If $X\SubSet{\mathBb P}^N$ is a smoOth HyPersurface of DEgRee $N$, and if $4\lEq N\Leq 12$, then $X$ is biRatIonallY sUpeRRigid.
basEd on pReviouS Ideas oF Corti, PukHlIKov proPOsed in [@PU1] A PrOof oF the birational rigIDiTY of every smooth HypersUrFAcE OF deGreE $N$ in $\P^N$, for $N\GeQ 6$. UnfoRTunatelY, At THE MomENt there is a gap In his argumeNTs (sEe RemaRk \[Gap\] BElow). DeSpite ThIS gaP, the proof prOposEd in [@Pu1] conTains mANy remarKAble ideAs, and iT seEms LikeLY tHaT a cOmPLetE PrOof COulD be obtaiNeD iN the fUturE ALONg thOse LineS. In faCt, the outline oF thE proOF of theorEm \[thm3\_IntrOd\] FolloWs his mEthod, AnD our contributioN is mAinly to siMplIfYinG aNd solIDifyinG hiS arGument.
ACknowleDGemEnTS {#ACkNowledgements.unnumBeRED}
----------------
WE are gratEful to sTeVe kLeiman anD ROb LAzarSFEld foR useFUl DiscussiOns. ResEArCh Of the fiRsT authoR wAs pArtIally SUppoRted by mURST of ITaliaN government, NatiONal Research PrOJeCT (coFIn 2000) “GeOmeTry of AlgebrAic VARietIes”. RESeArcH Of the SeconD aUThOR was partially supporTeD by NSF grant dMS 02-00278. The third auThor served AS A clay MathEmatICs iNstitute Long-TeRm PriZe Fellow whILe this reSearcH has been Done.
SinguLARities of Log PaiRs uNdeR PRoJections
============================================
Let $X$ bE A SmooTh AlgebraIc vAriety, dEfiNed OveR an AlGebraicalLy closed FiElD oF cHarActerIStic zero, AnD leT $V \SubSet X$ bE A propeR subsChemE. FOr ANy rAtional NUmBER $c > 0$, we CaN cOnsiDer ThE pair $(x,c\. V)$. THE usUal defiNitions in The THeorY oF sIngularIties of pairs, fOr Which we refEr To [@KO], extenD TO this conText. In particular, we say thAT an irreDucIble sUbvaRiety $C \subSet x$ is a ceNteR Of non lOg canoNicitY (rEsp. NON log tERMiNalItY, non canoniCITy, nOn terMiNaliTy) for $(X,c\. v)$ if there is at least oNE diVisorial valuaTioN of $K(x)$, WItH | rigid, for every $N>4$. We use our r esult s t o g iv e an eas y and uniformp roof of birational superri gidit yf or a r bi trary smooth hy p e rsu rf ac esof de gree$N$ in $\P ^N$ when $ N$is small.
\[t h m3 \_introd\] If $X\subset{\ mat hbb P} ^N $ i s a sm oot h hyp ersurf a ce ofdegree $N $, and if $4\leqN \ le q 12 $, then $X$ is bi r at i onally superri gid.
Ba s ed o n p rev ious ideas o f Cor t i, Pukh l ik o v pro p osed in [@Pu1 ] a proof o f th e bira ti ona l rigid ity o fe ver y smooth hy pers urface of degre e $N$ in $\P^N$, for $ N\g eq6$.U nf or tun at e ly, at th e mo ment the re i s a g ap i n h i s ar gum ents (see Remark \[gap \]belo w ).Despi te th is g ap , the proof prop os ed in [@Pu1] co ntai ns many r ema rk abl eideas , and i t s eem s likel y thata co mp l e t eproof could be obt ai n e din the f uturea lo ng those li ne s.In f a c t, th e ou t li ne of th e proo f o fTheorem \ [thm3\ _i ntr od\ ] fol l owshis me thod, an d our contribution i s mainly to si m pl i f yi n g an d s olidifyinghisa rgum ent.
A ckn o wledg ement s{ #a c knowledgements.unnu mb ered}----- -----------
We are gra t e f ul to St eveK le i man and Rob La zarsf eld for us e ful disc ussio ns. Rese arch of t h e first a uth orwas pa r t ia lly supported b y MU RS T of It ali an Gove rnm ent , N ati on al Resear ch Proje ct ( Co fi n 2 000)“ Geometry o f A lg ebr aic V a rietie s”. R esea rc ho f t he seco n da u thor w as par tia ll y sup port e d b y NSF G rant DMS02- 0 0278 .Th e third author serve das a ClayMa the matics I nstitute Long-Term Prize Fellow while t his rese arch has been do ne.
S ing u lariti es oflog p ai rsu n der p r o je cti on s
======== = = === ===== == ==== ======= =============
Let $X$ be a smoothalg ebra i c v ari e ty , de fi n edo v er an algebraic ally close df ie ld of char a cte ri stic ze ro, and let$ V \subs et X$ bea propersu bsch e m e.For any ra tional n umber $c> 0$,w ecan c ons ider t he pa ir $( X,c\.V )$. Theusualde finiti ons i nthe theo ry of singularities ofpairs, forwhi ch we ref ert o [ @Ko], ext endto this co nte xt. In p art i cular , we sa y t h at an irr e ducible s u bv ari e t y$C \subsetX $ isa cen ter of non log canonicity (resp . non log termi nali t y , n onc anon ic ity, non termi nal it y ) for $(X ,c \. V)$ if t here isat least one d ivisor ial val u a ti o n of $ K(X) $,with | rigid, for_every $N>4$._We use our results_to give_an_easy and_uniform_proof of birational_superrigidity for arbitrary_smooth hypersurfaces of degree_$N$ in $\P^N$_when_$N$ is small.
\[thm3\_introd\] If $X\subset{\mathbb P}^N$ is a smooth hypersurface of degree $N$, and_if_$4\leq N\leq_12$,_then_$X$ is birationally superrigid.
Based on_previous ideas of Corti, Pukhlikov_proposed in_[@Pu1] a proof of the birational rigidity of_every_smooth hypersurface of_degree $N$ in $\P^N$, for $N\geq 6$. Unfortunately, at_the moment there is a gap_in his arguments_(see_Remark \[gap\]_below). Despite this gap,_the proof proposed in [@Pu1] contains_many remarkable ideas, and it seems_likely that a complete proof could be_obtained in the future along those_lines. In fact, the outline_of the_proof of Theorem \[thm3\_introd\] follows his_method, and our_contribution is_mainly to simplifying_and solidifying his argument.
Acknowledgements {#acknowledgements.unnumbered}
----------------
We are_grateful to Steve_Kleiman and Rob Lazarsfeld for useful_discussions._Research of the_first_author_was partially_supported by MURST_of_Italian Government,_National_Research Project (Cofin 2000) “Geometry of_Algebraic_Varieties”. Research of the second author was_partially supported by NSF Grant_DMS 02-00278._The third author served_as a Clay Mathematics Institute_Long-Term Prize Fellow while this research_has been_done.
Singularities of_log pairs under projections
============================================
Let $X$ be a smooth algebraic variety, defined_over an algebraically closed field of_characteristic zero, and let_$V \subset_X$_be a proper_subscheme._For any_rational number $c > 0$, we can_consider the_pair $(X,c\. V)$. The usual definitions_in the theory of_singularities_of pairs, for which we refer_to [@Ko], extend to this context._In particular, we say that_an_irreducible_subvariety $C \subset X$ is_a center of non log canonicity_(resp. non log_terminality, non canonicity, non terminality) for $(X,c\._V)$_if there is at least one_divisorial_valuation of $K(X)$, with |
,j)$ entry of the matrix is the Pearson distance $s_{ij}$ color coded according to the key on the right, i.e. blue entries correspond to pairs of nodes having different time-series while yellow entries correspond to pairs with similar time-series. Thus one can coarsely distinguish between high- and low-degree nodes. This analysis on its own leads to a gross overestimate of the structural parameter, see panel (d). However, using this coarse classification and mathematical results on coupled systems (\[eq1\]) satisfying assumptions (i), (ii) and (iii) makes it possible to find an approximation of the evolution rule at each node, see panel (e). Using this additional information, one can obtain a description of the network connectivity structure, see panel (f). The network used in this figure has 1000 nodes, but only 50 are shown for clarity.[]{data-label="fig1"}](main_fig_v5_low_3_reduced.pdf){width="\linewidth"}
[**Scale-free networks of coupled [ bursting]{} neurons**]{}.
We consider coupled bursting neurons with excitatory synapses [@rulkov2001] in scale-free networks (see Methods for details on simulation of scale-free networks and formulation of the model). Our techniques work equally well for spiking dynamics.
We generate a scale-free network with $N=10^4$ nodes such that the probability of having a node of degree $k$ is proportional to $k^{-\gamma}$, where $\gamma=2.53$. For this reconstruction we only need 2000 data points for each node. Again, to every pair of time series $y_i$ and $y_j$ we assign a *Pearson distance* $s_{ij}\ge0$ and the node intensity $S_{i} = \sum_{j}s_{ij}$. The empirical distribution of the intensities $S_i$ approximates the degree distribution of the network, see Fig \[fig1\](d). In the example here, the estimated structural exponent from the distribution of $S_i$ is $\gamma_{\rm est} = 3.1$, which yields a relative error of nearly 25% with respect to the true value of $\gamma$ (see also the plots in Figure \[fig2\] a)). The functional network therefore overestimates $\gamma$, which has drastic consequences for the predicted character of the network. For example, the number of connections of a hub for a scale-free network is concentrated | , j)$ entry of the matrix is the Pearson distance $ s_{ij}$ color coded according to the keystone on the right field, i.e. blue entries represent to couple of nodes having unlike time - series while yellow entry correspond to pairs with exchangeable time - series. therefore one can coarsely distinguish between high- and low - academic degree nodes. This psychoanalysis on its own lead to a gross overestimate of the structural parameter, witness panel (d). However, practice this coarse classification and numerical results on copulate systems (\[eq1\ ]) satisfying assumptions (iodine), (ii) and (iii) makes it possible to find an approximation of the evolution rule at each node, see panel (e). Using this additional information, one can obtain a description of the network connectivity structure, see jury (f). The network used in this name hold 1000 nodes, but only 50 are indicate for clarity.[]{data - label="fig1"}](main_fig_v5_low_3_reduced.pdf){width="\linewidth " }
[ * * Scale - free networks of coupled [ bursting ] { } nerve cell * * ] { }.
We consider coupled bursting neurons with excitatory synapses [ @rulkov2001 ] in scale - free networks (see Methods for details on model of scale - complimentary networks and formulation of the model). Our techniques work evenly well for spiking dynamics.
We generate a scale - free network with $ N=10 ^ 4 $ nodes such that the probability of having a node of academic degree $ k$ is proportional to $ k^{-\gamma}$, where $ \gamma=2.53$. For this reconstruction we only need 2000 data points for each node. Again, to every couple of time series $ y_i$ and $ y_j$ we assign a * Pearson distance * $ s_{ij}\ge0 $ and the node volume $ S_{i } = \sum_{j}s_{ij}$. The empirical distribution of the saturation $ S_i$ approximates the degree distribution of the network, see Fig \[fig1\](d). In the example here, the calculate structural exponent from the distribution of $ S_i$ is $ \gamma_{\rm est } = 3.1 $, which yields a relative error of nearly 25% with respect to the true value of $ \gamma$ (see besides the plots in Figure \[fig2\ ] a) ). The functional net therefore overestimates $ \gamma$, which has drastic consequence for the predicted character of the network. For example, the number of connections of a hub for a scale - free network is concentrated | ,j)$ ejtry of the matrix is tht Pearson distancg $w_{ij}$ conor cosed accofding to the key on the righv, i.e. blue entries correspond to pairs of nodes hqvinj different time-series wmnle ysplow xntries corresppnd to paiss with similas gile-series. Thus one can coarsely distynguish bftween high- anq loe-qegrsv kodes. This analysis on its own lsads to a gross overrstimate of the structural parwmeter, see panel (d). However, usung eyis coarse cuassificatpmn and matgematical results on coupled syrtems (\[eq1\]) satisfinbg wvsumptions (m), (ii) agd (iii) makes it posshble to find an approwimatmon if the evolution rule at each node, see kanel (e). Ushny this additional infirnatiot, ona cav obgaih e dsscriphioi of the nefwork connextivity structure, ste [qnel (f). The nefwork tsqd in this figure has 1000 nodes, but only 50 dre shown for clarity.[]{data-lqbel="fig1"}](main_fig_v5_low_3_redoced.pdf){wideh="\linewidth"}
[**Scale-free networks of coupled [ burstinc]{} neucovs**]{}.
Wt gjvwifer coupled bursting neurons with excitatory fgnspxes [@rulkov2001] in fcale-free nrtaoths (see Methodr for bstzils on simulation of scaje-frew networkf anc formulation of the model). Iur techniquvs wirk equally well flr spiking bynamivs.
We benerate a scale-free necwork sith $N=10^4$ noded such thzg the probabilith on hdving a node of degree $k$ if proportmonal to $k^{-\gaoma}$, ehere $\damma=2.53$. For hhis vaconstruction we ojly nged 2000 ddta points for each node. Again, to every pemr of time setiev $y_p$ and $y_j$ ce asslgn a *Pearson dystance* $s_{ij}\ge0$ cnd the uode ivtensity $S_{p} = \sum_{j}s_{io}$. The empirisal distributhln of the invensities $S_i$ qppriximater the degree dixtributiou of rhe network, see Fin \[fig1\](d). Jn the example ktre, the estimated xtrjcttrwl ev[onent from dhe aisgtibutkon of $S_i$ iw $\gamka_{\rm est} = 3.1$, which yiends z relative error og kearly 25% wuth respqct to the trie value of $\gamma$ (dee anso the llojs in Figure \[fig2\] a)). The functionzl networn tmerefore overqstinates $\gamma$, chich has drastic consequences for the pcedicted character of jhe network. For exampkt, the number pf cognections of a hub for a scalw-free network is goncentrated | ,j)$ entry of the matrix is the $s_{ij}$ coded according the key on correspond pairs of nodes different time-series while entries correspond to pairs with similar Thus one can coarsely distinguish between high- and low-degree nodes. This analysis on own leads to a gross overestimate of the structural parameter, see panel (d). using coarse and results on coupled systems (\[eq1\]) satisfying assumptions (i), (ii) and (iii) makes it possible to find approximation of the evolution rule at each node, panel (e). Using this information, one can obtain a of network connectivity see (f). network used in figure has 1000 nodes, but only 50 are shown for clarity.[]{data-label="fig1"}](main_fig_v5_low_3_reduced.pdf){width="\linewidth"} [**Scale-free networks of coupled [ bursting]{} We consider neurons with synapses in networks (see Methods on simulation of scale-free networks and model). Our techniques work equally well for spiking We generate scale-free network with $N=10^4$ nodes such the probability of having a node of degree is proportional to $k^{-\gamma}$, where $\gamma=2.53$. For this reconstruction we only need 2000 data points node. Again, to every of time series and we a distance* $s_{ij}\ge0$ the node intensity $S_{i} = \sum_{j}s_{ij}$. The empirical distribution of the $S_i$ approximates the degree distribution of the network, see Fig the here, the estimated exponent from the distribution $S_i$ $\gamma_{\rm est} = 3.1$, a error with to true value of $\gamma$ also the plots in Figure a)). The functional network drastic consequences for the predicted character of the For example, the number of connections of hub for a scale-free network is concentrated | ,j)$ entry of the matrix is the PeaRson distanCe $s_{ij}$ ColOr cOdEd acCordIng to the key on tHE rigHt, i.e. blue entries correspOnd to PaIRs of NOdEs havIng diffEReNT TimE-sErIes WhILe YelloW enTries coRrespond to PaiRs With similar tIMe-Series. Thus One Can coarsely dIstInguisH bEtwEEn higH- anD low-dEgree nODes. ThiS analysis On ITs own lEAds to a gROSs OverEstimate of the struCTuRAl parameter, see Panel (d). hoWEvER, UsiNg tHis coarse cLaSsifiCAtion anD MaTHEMatICal results on cOupled systeMS (\[eq1\]) SatisfYiNg aSSumptiOns (i), (iI) aND (iiI) makes it posSiblE to find an ApproxIMation oF The evolUtion rUle At eAch nODe, SeE paNeL (E). UsINg ThiS AddItional iNfOrMatioN, one CAN OBtaiN a dEscrIptioN of the network ConNectIVitY struCture, See pAnEl (f). ThE netwoRk useD iN this figure has 1000 nOdes, But only 50 arE shOwN foR cLaritY.[]{Data-laBel="Fig1"}](Main_fig_V5_low_3_redUCed.PdF){WIDtH="\linewidth"}
[**Scale-freE nETWoRks of couPled [ buRStInG]{} Neurons**]{}.
WE cOnsIder COUpled BursTInG neurons With exCItAtOry synaPsEs [@rulkOv2001] In sCalE-free NEtwoRks (see methods fOr detAIls on simulatioN Of scale-free neTWoRKS aND forMulAtion of the mOdel). oUr teChniQUeS woRK equaLly weLl FOr SPiking dynamics.
We genErAte a scAle-frEe network with $n=10^4$ nodes such THAT the probAbilITy OF having a node of DegreE $k$ is proporTIonal to $k^{-\Gamma}$, Where $\gamMa=2.53$. For this REConstrucTioN we OnlY neED 2000 DaTa points for eaCH Node. agAin, to evEry Pair of tIme SerIes $Y_i$ aNd $Y_j$ we assigN a *PearsoN dIsTaNcE* $s_{iJ}\ge0$ anD The node iNtEnsItY $S_{i} = \Sum_{j}s_{IJ}$. The emPiricAl diStRiBUtiOn of the INtENSitiEs $s_i$ ApprOxiMaTes thE degREe dIstribuTion of the NetWOrk, sEe fiG \[fig1\](d). In The example herE, tHe estimateD sTruCtural EXPonent frOm the distribution of $S_i$ is $\GAmma_{\rm eSt} = 3.1$, wHich yIeldS a relativE erRor of nEarLY 25% with rEspect To the TrUe vALUe of $\gAMMa$ (See AlSo the plots IN figUre \[fiG2\] a)). the fUnctionAl network therefore OVerEstimates $\gammA$, whIch hAS DrAstIC cONseQuENceS FOr the predicted cHaracter of ThE NeTwork. For exAMplE, tHe numbeR of connEctioNS of a hub For a scale-Free netwoRk Is coNCEntRated | ,j)$ entry of the matrix i s the Pear son d ist anc e$s_{ ij}$ color coded a c cord ing to the key on theright ,i .e.b lu e ent ries co r re s p ond t opai rs of node s h aving d ifferent t ime -s eries whiley el low entrie s c orrespond to pa irs wi th si m ilartim e-ser ies. T h us one can coar se l y dist i nguishb e tw eenhigh- and low-deg r ee nodes. This an alysis o n i t s ow n l eads to agr oss o v erestim a te o f th e structural p arameter, s e e p anel ( d) . H o wever, usin gt his coarse cla ssif ication a nd mat h ematica l result s on c oup led sys t em s(\[ eq 1 \]) sa tis f yin g assump ti on s (i) , (i i ) a nd ( iii ) ma kes i t possible to fi nd a n ap proxi matio n of t he ev olutio n rul eat each node, s ee p anel (e). Us in g t hi s add i tional in for mation, one ca n ob ta i n adescription of the n e t wo rk conne ctivit y s tr u cture, s ee pa nel( f ). Th e ne t wo rk usedin thi s f ig ure has 1 000 no de s,but only 50 a re sho wn for c larit y .[]{data-label = "fig1"}](main _ fi g _ v5 _ low_ 3_r educed.pdf) {wid t h="\ line w id th" }
[** Scale -f r ee networks of coupled [ burst ing]{ } neurons**]{ }.
We con s i d er coupl ed b u rs t ing neurons wi th ex citatory s y napses [ @rulk ov2001]in scale- f r ee netwo rks (s eeMet h o ds for detailso n sim ul ation o f s cale-fr eenet wor ksan d formula tion ofth emo de l). Ourt echnique swor kequ allyw ell fo r spi king d yn a mic s.
Weg en e r ateasc ale- fre enetwo rk w i th$N=10^4 $ nodes s uch that t he probab ility of havi ng a node of d egr ee $k$ i s propor tional to $k^{-\gamma}$ , where$\g amma= 2.53 $. For th isrecons tru c tion w e only need 2 000 d ata p o i nt s f or each node . Aga in, t oever y pairof time series $y_ i $ a nd $y_j$ we a ssi gn a * Pe ars o nd ist an c e*$ s _{ij}\ge0$ andthe node i nt e ns ity $S_{i} = \ su m_{j}s_ {ij}$.The e m pirical distribu tion of t he int e n sit ies $S_i$approxim ates thed egree di strib uti on ofth e n etwor k, see Fig \[fi g1\](d ). In th e exa mp le here, the estimated structur al exp onent fr om the di str i but ion of $S _i$is $\gamma _{\ rmest}= 3 . 1$, w hich yi eld s a re lati v e error o f n ear l y 2 5% with res p e c t t o the tr u e valu e of $\gamma$ (see al s o the plots in Fig u r e \ [fi g 2\]a) ). The functio nal n e t work the re fore overes timates$\ g amma$ , whic h hasdrastic c on s equenc es f orthe predi cte dc haracte rof the ne twor k. For e xample , the n umber of connect ionso f a hu b fo r a s ca le-free netw ork is con centrated | ,j)$ entry_of the_matrix is the Pearson_distance $s_{ij}$_color_coded according_to_the key on_the right, i.e._blue entries correspond to_pairs of nodes_having_different time-series while yellow entries correspond to pairs with similar time-series. Thus one can_coarsely_distinguish between_high-_and_low-degree nodes. This analysis on_its own leads to a_gross overestimate_of the structural parameter, see panel (d). However,_using_this coarse classification_and mathematical results on coupled systems (\[eq1\]) satisfying assumptions_(i), (ii) and (iii) makes it_possible to find_an_approximation_of the evolution rule_at each node, see panel (e)._Using this additional information, one can_obtain a description of the network connectivity_structure, see panel (f). The network_used in this figure has_1000 nodes,_but only 50 are shown_for clarity.[]{data-label="fig1"}](main_fig_v5_low_3_reduced.pdf){width="\linewidth"}
[**Scale-free networks_of coupled_[ bursting]{} neurons**]{}.
We_consider coupled bursting neurons with excitatory_synapses [@rulkov2001] in_scale-free networks (see Methods for details_on_simulation of scale-free_networks_and_formulation of_the model). Our_techniques_work equally_well_for spiking dynamics.
We generate a scale-free_network_with $N=10^4$ nodes such that the probability_of having a node_of_degree $k$ is proportional_to $k^{-\gamma}$, where $\gamma=2.53$. For_this reconstruction we only need 2000_data points_for each_node. Again, to every pair of time series $y_i$ and $y_j$_we assign a *Pearson distance* $s_{ij}\ge0$_and the node intensity_$S_{i} =_\sum_{j}s_{ij}$._The empirical distribution_of_the intensities_$S_i$ approximates the degree distribution of the_network, see_Fig \[fig1\](d). In the example here, the_estimated structural exponent from_the_distribution of $S_i$ is $\gamma_{\rm est}_= 3.1$, which yields a relative_error of nearly 25% with_respect_to_the true value of $\gamma$_(see also the plots in Figure_\[fig2\] a)). The_functional network therefore overestimates $\gamma$, which has_drastic_consequences for the predicted character of_the_network. For example, the number of_connections_of_a hub for a scale-free_network is concentrated |
bundle. We consider a Tonelli Hamiltonian $H~: T^*M\rightarrow {\mathbb {R}}$, i.e. a $C^2$ function that is strictly $C^2$-convex and superlinear in the fiber. The Hamiltonian flow associated with such a function is denoted by $(\varphi_t)_{t\in{\mathbb {R}}}$ or $(\varphi_t^H)_{t\in{\mathbb {R}}}$. To such a Hamiltonian, there corresponds a Lagrangian function $L~: TM\rightarrow {\mathbb {R}}$ that has the same regularity as $H$ and is also superlinear and strictly convex in the fiber. The corresponding Euler-Lagrange flow is denoted by $(f_t)_{t\in{\mathbb {R}}}$.
For such a Hamiltonian system, it is usual to study its “minimizing objects”; more precisely, a piece of orbit $(\varphi_t(q,p))_{t\in [a,b]}=(q_t, p_t)_{t\in[a,b]}$ is minimizing if the arc $(q_t)_{t\in[a, b]}$ minimizes the action functional $A_L$ defined by $A_L(\gamma )=\int_a^bL(\gamma (t), \dot\gamma (t))dt$ among the $C^2$-arcs joining $q_a$ to $q_b$. More generally, if $I$ is an interval and $(\varphi_t)_{t\in I}=(q_t, p_t)_{t\in I}$ is an orbit piece, we say that it is minimizing if for every segment $[a,b]\subset I$, its restriction to $[a,b]$ is minimizing. Then we call the set of points of $T^*M$ whose (complete) orbit is minimizing the [*Mañé set*]{}. We denote it by ${\mathcal {N}}^*(H)$ and its projection, the [*projected Mañé set*]{}, is denoted by: ${\mathcal {N}}(H)=\pi ({\mathcal {N}}^*(H))$. The Mañé set is non empty, compact and invariant by the Hamiltonian flow (see [@Fa1]). The first proof of the non-emptiness of the Mañé set is due to J. Mather: he proved in the 90’s in [@mather1] the existence of minimizing measures.
We are interested in invariant subsets of the Mañé set | bundle. We consider a Tonelli Hamiltonian $ H~: T^*M\rightarrow { \mathbb { R}}$, i.e. a $ C^2 $ function that is strictly $ C^2$-convex and superlinear in the roughage. The Hamiltonian menstruation associated with such a function is denoted by $ (\varphi_t)_{t\in{\mathbb { R}}}$ or $ (\varphi_t^H)_{t\in{\mathbb { R}}}$. To such a Hamiltonian, there equate a Lagrangian function $ L~: TM\rightarrow { \mathbb { R}}$ that has the like regularity as $ H$ and is also superlinear and strictly convex in the fiber. The comparable Euler - Lagrange menstruation is denoted by $ (f_t)_{t\in{\mathbb { R}}}$.
For such a Hamiltonian organization, it is usual to study its “ understate object ”; more precisely, a piece of eye socket $ (\varphi_t(q, p))_{t\in [ a, b]}=(q_t, p_t)_{t\in[a, b]}$ is minimizing if the arc $ (q_t)_{t\in[a, b]}$ minimizes the military action functional $ A_L$ defined by $ A_L(\gamma) = \int_a^bL(\gamma (t), \dot\gamma (t))dt$ among the $ C^2$-arcs joining $ q_a$ to $ q_b$. More broadly, if $ I$ is an interval and $ (\varphi_t)_{t\in I}=(q_t, p_t)_{t\in I}$ is an orbit piece, we say that it is minimize if for every segment $ [ a, b]\subset I$, its restriction to $ [ a, b]$ is minimizing. Then we call the hardening of points of $ T^*M$ whose (complete) orbit is minimizing the [ * Mañé set * ] { }. We denote it by $ { \mathcal { N}}^*(H)$ and its projection, the [ * projected Mañé set * ] { }, is denoted by: $ { \mathcal { N}}(H)=\pi ({ \mathcal { N}}^*(H))$. The Mañé set is non empty, compact and invariant by the Hamiltonian flow (see [ @Fa1 ]). The first validation of the non - emptiness of the Mañé stage set is due to J. Mather: he prove in the 90 ’s in [ @mather1 ] the being of minimizing measures.
We are interested in changeless subsets of the Mañé jell | bujdle. We consider a Tonelui Hamiltonian $K~: T^*M\rigitarrow {\mathbb {F}}$, i.e. a $C^2$ function that is stcictoy $C^2$-cinvex and superlinear kn the fiher. The Yamioronian flox associated wjbh sueh a function is denoted bf $(\varphi_t)_{t\in{\madhcb {R}}}$ or $(\varphi_t^H)_{t\in{\mathbb {R}}}$. To such a Hamiltpnlan, there corrgsponcf a Magrangian function $L~: TM\rightarrow {\mathbb {R}}$ that has tne same regularity as $H$ anf is also superlinear wnd strictli cogcex in the fkber. The corresponding Euler-Lagrange flow is denoted cy $(f_t)_{c\in{\mathbb {R}}}$.
Doe skwh a Hamiltinian system, it is usual do studu its “minimizikg oboectw”; more precisely, a pixce of orbit $(\varphi_t(z,p))_{t\in [a,b]}=(q_d, '_t)_{t\in[a,b]}$ is minimizing id the arc $(q_t)_{t\kb[a, c]}$ mjnmmiaes thf artion functjonal $A_L$ dedined by $A_L(\gamma )=\inu_a^bJ(\tamma (t), \dot\gajma (t))de$ wmong the $C^2$-arcs joining $q_a$ to $q_b$. More gtneramly, if $I$ is an interval and $(\varphi_t)_{t\in I}=(q_t, p_j)_{t\in I}$ is wn orbit piece, we say that it is minimizing if fos evecy seynent $[x,v]\skbset I$, its restriction to $[a,b]$ is minimizing. Trsn wv call the set of points of $Y^*M$ wnjse (complete) urbit na jinimizing the [*Mañé det*]{}. We qenotw it by ${\mwthcsl {N}}^*(H)$ and its projection, thw [*projected Iqñé set*]{}, is denoted bv: ${\mathcal {N}}(H)=\'i ({\matncal {M}}^*(H))$. The Mañé set is non em'ty, cojpact and ijvariant gh the Hamiltoniav fkof (see [@Fa1]). The first proof os the non-xmptiuess of ghe Kañé set is due to J. Matmar: he proved in thf 90’s iu [@matver1] the exlstence of minimizing measures.
Wx are interesjed in invariaut subxets of the Mwñé set | bundle. We consider a Tonelli Hamiltonian $H~: {R}}$, a $C^2$ that is strictly fiber. Hamiltonian flow associated such a function denoted by $(\varphi_t)_{t\in{\mathbb {R}}}$ or $(\varphi_t^H)_{t\in{\mathbb To such a Hamiltonian, there corresponds a Lagrangian function $L~: TM\rightarrow {\mathbb {R}}$ has the same regularity as $H$ and is also superlinear and strictly convex the The Euler-Lagrange is denoted by $(f_t)_{t\in{\mathbb {R}}}$. For such a Hamiltonian system, it is usual to study its objects”; more precisely, a piece of orbit $(\varphi_t(q,p))_{t\in p_t)_{t\in[a,b]}$ is minimizing if arc $(q_t)_{t\in[a, b]}$ minimizes the functional defined by )=\int_a^bL(\gamma \dot\gamma among the $C^2$-arcs $q_a$ to $q_b$. More generally, if $I$ is an interval and $(\varphi_t)_{t\in I}=(q_t, p_t)_{t\in I}$ is an piece, we it is if every $[a,b]\subset I$, its $[a,b]$ is minimizing. Then we call points of $T^*M$ whose (complete) orbit is minimizing [*Mañé set*]{}. denote it by ${\mathcal {N}}^*(H)$ and projection, the [*projected Mañé set*]{}, is denoted by: {N}}(H)=\pi ({\mathcal {N}}^*(H))$. The Mañé set is non empty, compact and invariant by the Hamiltonian [@Fa1]). The first proof the non-emptiness of Mañé is to Mather: he in the 90’s in [@mather1] the existence of minimizing measures. We interested in invariant subsets of the Mañé set | bundle. We consider a Tonelli HAmiltonian $h~: T^*M\riGhtArrOw {\MathBb {R}}$, i.E. a $C^2$ function thaT Is stRictly $C^2$-convex and superlInear In THe fiBEr. the HaMiltoniAN fLOW asSoCiAteD wITh Such a FunCtion is Denoted by $(\vArpHi_T)_{t\in{\mathbb {R}}}$ oR $(\VaRphi_t^H)_{t\in{\mAthBb {R}}}$. To such a HaMilTonian, ThEre COrresPonDs a LaGrangiAN functIon $L~: TM\rigHtARrow {\maTHbb {R}}$ thaT HAs The sAme regularity as $H$ aND iS Also superlineaR and stRiCTlY COnvEx iN the fiber. THe CorreSPonding eUlER-lAgrANge flow is denoTed by $(f_t)_{t\in{\mAThbB {R}}}$.
For sUcH a HAMiltonIan syStEM, it Is usual to stUdy iTs “minimizIng objECts”; more PReciselY, a piecE of OrbIt $(\vaRPhI_t(Q,p))_{t\In [A,B]}=(q_t, P_T)_{t\In[a,B]}$ Is mInimizinG iF tHe arc $(Q_t)_{t\iN[A, B]}$ MInimIzeS the ActioN functional $A_L$ DefIned BY $A_L(\Gamma )=\Int_a^bl(\gamMa (T), \dot\gAmma (t))dT$ amonG tHe $C^2$-arcs joining $q_A$ to $q_B$. More geneRalLy, If $I$ Is An intERval anD $(\vaRphI_t)_{t\in I}=(q_T, p_t)_{t\in I}$ IS an OrBIT PiEce, we say that it is miNiMIZiNg if for eVery seGMeNt $[A,B]\subset I$, ItS reStriCTIon to $[A,b]$ is MInImizing. THen we cALl ThE set of pOiNts of $T^*m$ wHosE (coMpletE) OrbiT is minImizing tHe [*MañÉ Set*]{}. We denote it bY ${\Mathcal {N}}^*(H)$ and iTS pROJeCTion, The [*Projected MaÑé seT*]{}, Is deNoteD By: ${\MatHCal {N}}(H)=\Pi ({\matHcAL {N}}^*(h))$. the Mañé set is non emptY, cOmpact And inVariant by the HAmiltonian FLOW (see [@Fa1]). ThE firST pROof of the non-empTinesS of the Mañé SEt is due tO J. MatHer: he proVed in the 90’s IN [@Mather1] thE exIstEncE of MINiMizing measureS.
wE are InTeresteD in InvariaNt sUbsEts Of tHe mañé set | bundle. We consider a Ton elli Hamil tonia n $ H~: T ^*M\ righ tarrow {\mathb b {R} }$, i.e. a $C^2$ funct ion t ha t iss tr ictly $C^2$- c on v e x a nd s upe rl i ne ar in th e fiber . The Hami lto ni an flow asso c ia ted with s uch a functionisdenote dby$ (\var phi _t)_{ t\in{\ m athbb{R}}}$ or $ ( \varph i _t^H)_{ t \ in {\ma thbb {R}}}$. To s u ch a Hamiltonian, there c o rr e s pon dsa Lagrangi an func t ion $L~ : T M \ r igh t arrow {\mathb b {R}}$ tha t ha s thesa mer egular ity a s$ H$and is also sup erlinearand st r ictly c o nvex in the f ibe r.Thec or re spo nd i ngE ul er- L agr ange flo wis deno tedb y $ (f_t )_{ t\in {\mat hbb {R}}}$.
For suc h aHamil tonia n sy st em, i t is u sualto study its “min imiz ing objec ts” ;mor epreci s ely, a pi ece of orb it $(\v a rph i_ t ( q ,p ))_{t\in [a,b]}=(q _t , p_ t)_{t\in [a,b]} $ i sm inimizin gifthea r c $(q _t)_ { t\ in[a, b] }$ min i mi ze s the a ct ion fu nc tio nal $A_L $ def ined b y $A_L(\ gamma )=\int_a^bL(\g a mma (t), \dot \ ga m m a( t))d t$among the $ C^2$ - arcs joi n in g $ q _a$ t o $q_ b$ . M o re generally, if $I $is aninter val and $(\va rphi_t)_{t \ i n I}=(q_t , p_ t )_ { t\in I}$ is an orbi t piece, w e say tha t itis minim izing iff o r everyseg men t $ [a, b ] \s ubset I$, its r estr ic tion to $[ a,b]$ i s m ini miz ing .Then we c all these tof p oin ts of $T^*M$ w ho se(c omp lete) orbitis mi nimi zi ng the [*Mañé se t * ]{}. W edeno teit by $ {\ma t hca l {N}}^ *(H)$ and it s pro je ct ion, th e [*projected M añé set*]{ }, is denot e d by: ${\ mathcal {N}}(H)=\pi ({\ m athcal{N} }^*(H ))$. The Mañé se t is n one mpty,compac t and i nva r i ant b y th e H am iltonian f l o w ( see [ @F a1]) . The f irst proof of then on- emptiness ofthe Mañ é se t i s d u e t oJ . M a t her: he provedin the 90’ si n[@mather1] the e xistenc e of mi nimiz i ng meas ures.
We are inte re sted i n i nvariant s ubsets o f the Mañ é set | bundle._We consider_a Tonelli Hamiltonian $H~:_T^*M\rightarrow {\mathbb_{R}}$,_i.e. a_$C^2$_function that is_strictly $C^2$-convex and_superlinear in the fiber._The Hamiltonian flow_associated_with such a function is denoted by $(\varphi_t)_{t\in{\mathbb {R}}}$ or $(\varphi_t^H)_{t\in{\mathbb {R}}}$. To such_a_Hamiltonian, there_corresponds_a_Lagrangian function $L~: TM\rightarrow {\mathbb_{R}}$ that has the same_regularity as_$H$ and is also superlinear and strictly convex_in_the fiber. The_corresponding Euler-Lagrange flow is denoted by $(f_t)_{t\in{\mathbb {R}}}$.
For such_a Hamiltonian system, it is usual_to study its_“minimizing_objects”;_more precisely, a piece_of orbit $(\varphi_t(q,p))_{t\in [a,b]}=(q_t, p_t)_{t\in[a,b]}$ is_minimizing if the arc $(q_t)_{t\in[a, b]}$_minimizes the action functional $A_L$ defined by_$A_L(\gamma )=\int_a^bL(\gamma (t), \dot\gamma (t))dt$ among_the $C^2$-arcs joining $q_a$ to_$q_b$. More_generally, if $I$ is an_interval and $(\varphi_t)_{t\in_I}=(q_t, p_t)_{t\in_I}$ is an_orbit piece, we say that it_is minimizing if_for every segment $[a,b]\subset I$, its_restriction_to $[a,b]$ is_minimizing._Then_we call_the set of_points_of $T^*M$_whose_(complete) orbit is minimizing the [*Mañé_set*]{}._We denote it by ${\mathcal {N}}^*(H)$ and_its projection, the [*projected_Mañé_set*]{}, is denoted by:_${\mathcal {N}}(H)=\pi ({\mathcal {N}}^*(H))$. The_Mañé set is non empty, compact_and invariant_by the_Hamiltonian flow (see [@Fa1]). The first proof of the non-emptiness of_the Mañé set is due to_J. Mather: he proved in_the 90’s_in_[@mather1] the existence_of_minimizing measures.
We_are interested in invariant subsets of the_Mañé set |
pages:\
http://www.eso.org/observing/dfo/quality/GIRAFFE/qc/dark\_qc1.html
[^3]: Available at www.astro.uu.se/$\sim$barklem.
[^4]: http://www.eso.org/observing/dfo/quality/GIRAFFE/pipeline/\
solar.html
[^5]: Calibrations for the $v-y$-index for dwarf and giant stars are published in Paper I.
[^6]: ${\rm[X/Y]}=\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}-\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}_{\sun}$ where $N_{\rm X}$ is the number density of element X.
[^7]: $\log{\epsilon_{\rm x}}=\log{\left(\frac{N_{\rm x}}{N_{\rm H}}\right)}+12$
---
author:
- 'Chang-shui Yu [^1], He-shan Song'
title: Describing a Quantum Channel by State Tomography of a Single Probe State
---
Quantum states are the basic carrier of quantum information \[1\]. The core of all the quantum information processing (QIP) including quantum communication \[2\] and quantum computation \[3\] is the controlled time evolution of quantum state in essence \[4\]. However, in realistic scenario, quantum states will be unavoidably and greatly disturbed by the undesired coupling to the uncontrolled degree of freedom usually termed as ’environment’ and described as a ’quantum channel’. As a consequence, besides the state itself the valuable properties of quantum states such as coherence \[5,6\], entanglement \[7,8\] of composite systems and so on will be greatly corrupted. The precise characterization of some properties of quantum states usually largely relies on the evaluation of quantum states, if these properties such as entanglement does not correspond to a direct observable for a general unknown quantum state \[9-11\]. Furthermore, quantum channel is not restricted to the previous interaction between the system and environment. It is a general notion of any a input/output device governed by quantum mechanics including the controlled interactions, for example, the dynamical action of a quantum gate in a quantum computer etc \[12\]. Therefore, it is of practical importance to precisely explore the time evolution of quantum states on which a | pages:\
http://www.eso.org/observing/dfo/quality/GIRAFFE/qc/dark\_qc1.html
[ ^3 ]: Available at www.astro.uu.se/$\sim$barklem.
[ ^4 ]: http://www.eso.org/observing/dfo/quality/GIRAFFE/pipeline/\
solar.html
[ ^5 ]: Calibrations for the $ v - y$-index for dwarf and giant star are publish in Paper I.
[ ^6 ]: $ { \rm[X / Y]}=\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}-\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}_{\sun}$ where $ N_{\rm X}$ is the number density of element X.
[ ^7 ]: $ \log{\epsilon_{\rm x}}=\log{\left(\frac{N_{\rm x}}{N_{\rm H}}\right)}+12 $
---
generator:
-' Chang - shui Yu [ ^1 ], He - shan Song'
title: Describing a Quantum Channel by State Tomography of a Single Probe State
---
Quantum state are the basic carrier wave of quantum information \[1\ ]. The core of all the quantum information processing (QIP) including quantum communication \[2\ ] and quantum calculation \[3\ ] is the controlled time development of quantum state in essence \[4\ ]. However, in realistic scenario, quantum states will be unavoidably and greatly disturbed by the undesired coupling to the uncontrolled academic degree of freedom usually termed as ’ environment ’ and report as a ’ quantum channel ’. As a consequence, besides the state itself the valuable property of quantum states such as coherence \[5,6\ ], entanglement \[7,8\ ] of composite system and so on will be greatly corrupted. The precise characterization of some property of quantum states usually largely relies on the evaluation of quantum states, if these properties such as web does not match to a direct observable for a general unknown quantum state \[9 - 11\ ]. Furthermore, quantum channel is not restricted to the previous interaction between the system and environment. It is a cosmopolitan impression of any a input / output device governed by quantum automobile mechanic including the controlled interactions, for case, the dynamical action of a quantum gate in a quantum computer etc \[12\ ]. Therefore, it is of hardheaded importance to precisely explore the time evolution of quantum states on which a | pahes:\
http://www.eso.org/obsevving/dfo/quality/GNEAFFE/qr/dark\_qc1.gtml
[^3]: Avaklable at www.astro.uu.se/$\sim$barnlwm.
[^4]: htui://www.eso.org/observing/dfu/quality/GPRAFFE/pipwlint/\
solar.html
[^5]: Calibratiokf fod the $t-y$-index for dwatf and giant stars are pubnirhzd in Paper I.
[^6]: ${\rm[X/Y]}=\log{\left(\frac{N_{\rm X}}{N_{\ri Y}}\righy)}-\llg{\left(\frac{N_{\rm V}}{N_{\rm R}}\riggn)}_{\smn}$ where $N_{\rm X}$ is the number denaity of element X.
[^7]: $\lob{\epsilon_{\rm x}}=\log{\left(\frac{N_{\rm x}}{N_{\rl H}}\right)}+12$
---
author:
- 'Chwng-shui Yu [^1], He-srqn Song'
title: Describing a Quantum Dhannel by State Tomography of x Sinyle Probe Sjcre
---
Qkdntum statew are the basic carrier ox quantim information \[1\]. Tie cire of all the quantuk information procgssing (QIP) iucluding quantum commynucatimn \[2\] dnd dyangum cpmlutatiln \[3\] is the cohtrolled tine evolution of quamttn state in esaence \[4\]. Hjwever, in realistic scenario, quantum stdtea will be unavoidably abd greatly disturbed hy the unqesired coupling to the uncontrolled degree of fraedom jsucoly tdemfd as ’environment’ and described as a ’quantum sgamnvl’. As a consequenge, besides the staye iyfelf the valuxble pxkpsrties of quantum dtates fuch qs coheregce \[5,6\], entanglement \[7,8\] of composite systems and wo on will be greacly corrupteb. The krecisr characterization of sume lroperties lf quantuj states usually uarnelf relies on the evaluation of quantnm stctes, if ghesg propewties such as ekdanglement does noh cortespong to a dirfct observable for a general unknown quantum xtdte \[9-11\]. Furthexmore, auantum channql is not restticted to the orevious ihteractmon between ehe system ang environment. It is a genwral notion uf any a input/putput deniee governwd by quantum mechenicr including the eubtrolled interavtiuns, flr evdmple, the dytamizal sctiov of a quanbum gatr in a quantum compuder stc \[12\]. Therefore, it os of pracjical impjrtance to prrcisely explore tht time evulutipn jf quantum states on which a | pages:\ http://www.eso.org/observing/dfo/quality/GIRAFFE/qc/dark\_qc1.html [^3]: Available at www.astro.uu.se/$\sim$barklem. [^4]: [^5]: for the for dwarf and Paper [^6]: ${\rm[X/Y]}=\log{\left(\frac{N_{\rm X}}{N_{\rm X}}{N_{\rm Y}}\right)}_{\sun}$ where X}$ is the number density of X. [^7]: $\log{\epsilon_{\rm x}}=\log{\left(\frac{N_{\rm x}}{N_{\rm H}}\right)}+12$ --- author: - 'Chang-shui Yu [^1], He-shan title: Describing a Quantum Channel by State Tomography of a Single Probe State Quantum are basic of quantum information \[1\]. The core of all the quantum information processing (QIP) including quantum communication and quantum computation \[3\] is the controlled time of quantum state in \[4\]. However, in realistic scenario, states be unavoidably greatly by undesired coupling to uncontrolled degree of freedom usually termed as ’environment’ and described as a ’quantum channel’. As a consequence, the state valuable properties quantum such coherence \[5,6\], entanglement composite systems and so on will The precise characterization of some properties of quantum usually largely on the evaluation of quantum states, these properties such as entanglement does not correspond a direct observable for a general unknown quantum state \[9-11\]. Furthermore, quantum channel is not the previous interaction between system and environment. is general of a input/output governed by quantum mechanics including the controlled interactions, for example, the action of a quantum gate in a quantum computer etc it of practical importance precisely explore the time of states on which a | pages:\
http://www.eso.org/observiNg/dfo/qualiTy/GIRaFFe/qc/DaRk\_qc1.Html
[^3]: available at www.AStro.Uu.se/$\sim$barklem.
[^4]: http://www.eSo.org/ObSErviNG/dFo/quaLity/GIRafFe/PIpeLiNe/\
SolAr.HTmL
[^5]: CaliBraTions foR the $v-y$-indeX foR dWarf and giant STaRs are publiSheD in Paper I.
[^6]: ${\rm[X/y]}=\loG{\left(\fRaC{N_{\rM x}}{N_{\rm Y}}\RigHt)}-\log{\Left(\frAC{N_{\rm X}}{N_{\Rm Y}}\right)}_{\sUn}$ WHere $N_{\rM x}$ is the nUMBeR denSity of element X.
[^7]: $\log{\EPsILon_{\rm x}}=\log{\left(\fRac{N_{\rm X}}{N_{\RM H}}\RIGht)}+12$
---
AutHor:
- 'Chang-shUi yu [^1], He-sHAn Song'
tITlE: dEScrIBing a Quantum CHannel by StaTE ToMograpHy Of a sIngle PRobe STaTE
---
QuAntum states Are tHe basic caRrier oF Quantum INformatIon \[1\]. The CorE of All tHE qUaNtuM iNForMAtIon PRocEssing (QIp) iNcLudinG quaNTUM CommUniCatiOn \[2\] and Quantum computAtiOn \[3\] is THe cOntroLled tIme eVoLutioN of quaNtum sTaTe in essence \[4\]. HoweVer, iN realistiC scEnAriO, qUantuM States WilL be UnavoidAbly and GReaTlY DIStUrbed by the undesireD cOUPlIng to the UncontROlLeD Degree of FrEedOm usUALly teRmed AS ’eNvironmeNt’ and dEScRiBed as a ’qUaNtum chAnNel’. as a ConseQUencE, besidEs the staTe itsELf the valuable pROperties of quaNTuM STaTEs suCh aS coherence \[5,6\], eNtanGLemeNt \[7,8\] of COmPosITe sysTems aNd SO oN Will be greatly corrupTeD. The prEcise CharacterizatIon of some pROPErties of QuanTUm STates usually laRgely Relies on thE EvaluatiOn of qUantum stAtes, if theSE PropertiEs sUch As eNtaNGLeMent does not coRREspoNd To a direCt oBservabLe fOr a GenEraL uNknown quaNtum statE \[9-11\]. FUrThErMorE, quanTUm channeL iS noT rEstRicteD To the pRevioUs inTeRaCTioN betweeN ThE SYsteM aNd EnviRonMeNt. It iS a geNEraL notion Of any a inpUt/oUTput DeViCe goverNed by quantum mEcHanics inclUdIng The conTROlled intEractions, for example, the dYNamical ActIon of A quaNtum gate iN a qUantum ComPUter etC \[12\]. ThereFore, iT iS of PRActicAL ImPorTaNce to preciSELy eXplorE tHe tiMe evoluTion of quantum stateS On wHich a | pages:\
http://www.es o.org/obse rving /df o/q ua lity /GIR AFFE/qc/dark\_ q c1.h tml
[^3]: Available a t www .a s tro. u u. se/$\ sim$bar k le m .
[ ^4 ]: ht tp : // www.e so. org/obs erving/dfo /qu al ity/GIRAFFE/ p ip eline/\
s olar.html
[ ^5] : Cali br ati o ns fo r t he $v -y$-in d ex for dwarf an dg iant s t ars are p ub lish ed in Paper I.
[ ^ 6] : ${\rm[X/Y]}=\ log{\l ef t (\ f r ac{ N_{ \rm X}}{N_ {\ rm Y} } \right) } -\ l o g {\l e ft(\frac{N_{\ rm X}}{N_{\ r m Y }}\rig ht )}_ { \sun}$ wher e$ N_{ \rm X}$ isthenumber de nsityo f eleme n t X.
[ ^7]: $ \lo g{\ epsi l on _{ \rm x } }=\ l og {\l e ft( \frac{N_ {\ rm x}}{ N_{\ r m H }}\r igh t)}+ 12$
---
author:
- 'C hang - shu i Yu[^1], He- sh an So ng'
ti tle:De scribing a Quan tumChannel b y S ta teTo mogra p hy ofa S ing le Prob e State ---
Q u a nt um states are theba s i ccarrierof qua n tu mi nformati on \[ 1\]. T he co re o f a ll the q uantum in fo rmation p rocess in g ( QIP ) inc l udin g quan tum comm unica t ion \[2\] andq uantum comput a ti o n \ [ 3\]isthe control ledt imeevol u ti ono f qua ntumst a te in essence \[4\]. H ow ever,in re alistic scena rio, quant u m states w illb eu navoidably and grea tly distur b ed by th e und esired c oupling t o the unco ntr oll eddeg r e eof freedom us u a llyte rmed as ’e nvironm ent ’ a nddes cr ibed as a ’quantu mch an ne l’. As a conseque nc e,be sid es th e state itse lf t he v a lua ble pro p er t i es o fqu antu m s ta tes s ucha s c oherenc e \[5,6\] , e n tang le me nt \[7, 8\] of compos it e systemsan d s o on w i l l be gre atly corrupted. The pre c ise cha rac teriz atio n of some pr operti eso f quan tum st atesus ual l y larg e l yrel ie s on the e v a lua tionof qua ntum st ates, if these pro p ert ies such as e nta ngle m e nt do e sn otco r res p o nd to a directobservable f o ra generalu nkn ow n quant um stat e \[9 - 11\]. F urthermor e, quantu mchan n e l i s not rest ricted t o the pre v iousi nt eract ion betwe en th e sys tem an d en viron ment.It is agener al notionof any a input/output d evicegover ned by quant umm ech anics inc ludi ng the con tro lle d int era c tions , fo r e xam p le, t he d y namical a c ti ono f a quantum ga t e ina qua ntu m compu teretc \[12\]. There f ore, it is ofprac t i cal im p orta nc e to precisely ex pl o r e the ti me evolutionof quant um state s on w hich a | pages:\
_ _ http://www.eso.org/observing/dfo/quality/GIRAFFE/qc/dark\_qc1.html
[^3]: Available at_www.astro.uu.se/$\sim$barklem.
[^4]: http://www.eso.org/observing/dfo/quality/GIRAFFE/pipeline/\
__ _solar.html
[^5]:_Calibrations for the_$v-y$-index for dwarf_and giant stars are_published in Paper_I.
[^6]:_${\rm[X/Y]}=\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}-\log{\left(\frac{N_{\rm X}}{N_{\rm Y}}\right)}_{\sun}$ where $N_{\rm X}$ is the number density of element_X.
[^7]:_$\log{\epsilon_{\rm x}}=\log{\left(\frac{N_{\rm_x}}{N_{\rm_H}}\right)}+12$
_---
author:
- 'Chang-shui Yu [^1], He-shan_Song'
title: Describing a Quantum Channel_by State_Tomography of a Single Probe State
---
Quantum states are_the_basic carrier of_quantum information \[1\]. The core of all the quantum_information processing (QIP) including quantum communication_\[2\] and quantum_computation_\[3\]_is the controlled time_evolution of quantum state in essence_\[4\]. However, in realistic scenario, quantum_states will be unavoidably and greatly disturbed_by the undesired coupling to the_uncontrolled degree of freedom usually_termed as_’environment’ and described as a_’quantum channel’. As_a consequence,_besides the state_itself the valuable properties of quantum_states such as_coherence \[5,6\], entanglement \[7,8\] of composite_systems_and so on_will_be_greatly corrupted._The precise characterization_of_some properties_of_quantum states usually largely relies on_the_evaluation of quantum states, if these properties_such as entanglement does_not_correspond to a direct_observable for a general unknown_quantum state \[9-11\]. Furthermore, quantum channel_is not_restricted to_the previous interaction between the system and environment. It is a_general notion of any a input/output_device governed by quantum_mechanics including_the_controlled interactions, for_example,_the dynamical_action of a quantum gate in a_quantum computer_etc \[12\]. Therefore, it is of_practical importance to precisely_explore_the time evolution of quantum states_on which a |
}$ of points of $G$ which converges in $\mathbb{R}$ to $x$. The set $K=\{x\}\cup\{x_n: n\in\omega\}$ is compact in $\mathbb{S}$ but $K\cap F$ is not closed in $\mathbb{S}$. This proves that $\mathbb{S}$ is a $k$-space.
It is consistent with $\mathbf{ZF}$ that the Sorgenfrey line is not a $k$-space.
From Theorems 2.28 and 3.10, we immediately obtain the following:
Let $\tau$ be the topology of the Sorgenfrey line. If every compact in $(\mathbb{R}, \tau^{\star})$ set is closed in $(\mathbb{R}, \tau)$, then $\mathbf{CC}(c\mathbb{R})$ holds.
Compact complement partial topology
===================================
Let us slightly reformulate Definition 2.1 of [@OPSW]:
A *partial topology* of a set $X$ is a collection $Cov_X\subseteq\mathcal{P}(\mathcal{P}(X))$ which satisfies the following conditions:
1. $\tau_X=\bigcup Cov_X$ is a topology on $X$;
2. if $\mathcal{U}\subseteq \tau_X$ and $\mathcal{U}$ is finite, then $\mathcal{U}\in Cov_X$;
3. if $\mathcal{U}\in Cov_X$ and $V\in\tau_X$, then $\{U\cap V: U\in\mathcal{U}\}\in Cov_X$;
4. if $\mathcal{U}\in Cov_X$ and, for each $U\in\mathcal{U}$, we have $\mathcal{V}(U)\in Cov_X$ such that $U=\bigcup\mathcal{V}(U)$, then $\bigcup_{U\in\mathcal{U}}\mathcal{V}(U)\in Cov_X$;
5. if $\mathcal{U}\subseteq \tau_X$ and $\mathcal{V}\in Cov_X$ are such that $\bigcup\mathcal{U}=\bigcup\mathcal{V}$ and, for each $V\in\mathcal{V}$, there exists $U\in\mathcal{U}$ such that $V\subseteq U$, then $\mathcal{U}\in | } $ of points of $ G$ which converges in $ \mathbb{R}$ to $ x$. The set $ K=\{x\}\cup\{x_n: n\in\omega\}$ is compact in $ \mathbb{S}$ but $ K\cap F$ is not close in $ \mathbb{S}$. This proves that $ \mathbb{S}$ is a $ k$-space.
It is consistent with $ \mathbf{ZF}$ that the Sorgenfrey course is not a $ k$-space.
From Theorems 2.28 and 3.10, we immediately prevail the following:
Let $ \tau$ be the regional anatomy of the Sorgenfrey line. If every compact in $ (\mathbb{R }, \tau^{\star})$ set is close in $ (\mathbb{R }, \tau)$, then $ \mathbf{CC}(c\mathbb{R})$ holds.
Compact complement partial regional anatomy
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Let us slightly redevelop Definition 2.1 of [ @OPSW ]:
A * partial topology * of a set $ X$ is a collection $ Cov_X\subseteq\mathcal{P}(\mathcal{P}(X))$ which satisfies the following conditions:
1. $ \tau_X=\bigcup Cov_X$ is a regional anatomy on $ X$;
2. if $ \mathcal{U}\subseteq \tau_X$ and $ \mathcal{U}$ is finite, then $ \mathcal{U}\in Cov_X$;
3. if $ \mathcal{U}\in Cov_X$ and $ V\in\tau_X$, then $ \{U\cap V: U\in\mathcal{U}\}\in Cov_X$;
4. if $ \mathcal{U}\in Cov_X$ and, for each $ U\in\mathcal{U}$, we have $ \mathcal{V}(U)\in Cov_X$ such that $ U=\bigcup\mathcal{V}(U)$, then $ \bigcup_{U\in\mathcal{U}}\mathcal{V}(U)\in Cov_X$;
5. if $ \mathcal{U}\subseteq \tau_X$ and $ \mathcal{V}\in Cov_X$ are such that $ \bigcup\mathcal{U}=\bigcup\mathcal{V}$ and, for each $ V\in\mathcal{V}$, there exists $ U\in\mathcal{U}$ such that $ V\subseteq U$, then $ \mathcal{U}\in | }$ of points of $G$ which convevges in $\mathbb{R}$ jo $x$. The set $K=\{s\}\cup\{x_n: n\kn\omega\}$ is compact in $\mathbb{D}$ vut $K\xap F$ is not closed in $\mathbb{S}$. Nhis provws tiat $\mathbb{S}$ is a $k$-space.
Ib is dlnsivvent with $\mathbn{ZF}$ that tha Sorgenfrey lhnd ns not a $k$-space.
From Theorems 2.28 and 3.10, wq immedoahely obtain thg folkjwinf:
Let $\tau$ be the topology of the Sodgenfrej line. If every cpmpact in $(\mathbb{R}, \tau^{\star})$ det ls closed in $(\mathbh{R}, \tau)$, then $\matrvf{CC}(c\mathbb{R})$ holds.
Compact complemeht partial topology
===================================
Let us slighgly rzformulate Eedinljion 2.1 of [@OPSX]:
A *parnial topology* of a sed $X$ is s collection $Cpv_X\vubweteq\mathcal{P}(\mathcal{P}(E))$ which satisfies thg followinc eonditions:
1. $\tau_X=\bigcup Xov_X$ hs a topuoogh oh $E$;
2. jf $\matjcam{U}\subseteq \tau_X$ and $\mqthcal{U}$ is finite, tneg $\mathcal{U}\in Ckv_X$;
3. is $\iathcal{U}\in Cov_X$ and $V\in\tau_X$, then $\{U\cap V: U\ih\mathcal{U}\}\in Cov_X$;
4. if $\marhcal{U}\in Cov_X$ and, for each $U\in\iathcal{U}$, we have $\mathcal{V}(U)\in Cov_X$ such that $U=\bigcgp\maticxl{V}(B)$, then $\vihcup_{U\in\mathcal{U}}\mathcal{V}(U)\in Cov_X$;
5. if $\mathcal{U}\stgsttea \tau_X$ and $\matmcal{V}\in Cov_X$ are sicj yrat $\bigcup\matfcal{U}=\bnfchp\mathcal{V}$ and, for each $V\yn\matycal{V}$, thewe ecists $U\in\mathcal{U}$ such that $V\subseteq U$, rhen $\mathcal{U}\in | }$ of points of $G$ which converges to The set n\in\omega\}$ is compact is closed in $\mathbb{S}$. proves that $\mathbb{S}$ a $k$-space. It is consistent with that the Sorgenfrey line is not a $k$-space. From Theorems 2.28 and 3.10, immediately obtain the following: Let $\tau$ be the topology of the Sorgenfrey line. every in \tau^{\star})$ is closed in $(\mathbb{R}, \tau)$, then $\mathbf{CC}(c\mathbb{R})$ holds. Compact complement partial topology =================================== Let us slightly Definition 2.1 of [@OPSW]: A *partial topology* of set $X$ is a $Cov_X\subseteq\mathcal{P}(\mathcal{P}(X))$ which satisfies the following 1. Cov_X$ is topology $X$; if $\mathcal{U}\subseteq \tau_X$ $\mathcal{U}$ is finite, then $\mathcal{U}\in Cov_X$; 3. if $\mathcal{U}\in Cov_X$ and $V\in\tau_X$, then $\{U\cap V: U\in\mathcal{U}\}\in Cov_X$; if $\mathcal{U}\in for each we $\mathcal{V}(U)\in such that $U=\bigcup\mathcal{V}(U)$, Cov_X$; 5. if $\mathcal{U}\subseteq \tau_X$ and such that $\bigcup\mathcal{U}=\bigcup\mathcal{V}$ and, for each $V\in\mathcal{V}$, there $U\in\mathcal{U}$ such $V\subseteq U$, then $\mathcal{U}\in | }$ of points of $G$ which converges In $\mathbb{R}$ tO $x$. The Set $k=\{x\}\cUp\{X_n: n\iN\omeGa\}$ is compact in $\mAThbb{s}$ but $K\cap F$ is not closed in $\MathbB{S}$. tHis pROvEs thaT $\mathbb{s}$ Is A $K$-SpaCe.
it Is cOnSIsTent wIth $\Mathbf{Zf}$ that the SoRgeNfRey line is not A $K$-sPace.
From ThEorEms 2.28 and 3.10, we immeDiaTely obTaIn tHE follOwiNg:
Let $\Tau$ be tHE topolOgy of the SOrGEnfrey LIne. If evERY cOmpaCt in $(\mathbb{R}, \tau^{\staR})$ SeT Is closed in $(\mathBb{R}, \tau)$, ThEN $\mATHbf{cC}(c\Mathbb{R})$ holDs.
compaCT compleMEnT PARtiAL topology
===================================
Let uS slightly reFOrmUlate DEfIniTIon 2.1 of [@OpSW]:
A *pArTIal Topology* of a Set $X$ Is a collecTion $CoV_x\subsetEQ\mathcaL{P}(\mathCal{p}(X))$ wHich SAtIsFieS tHE foLLoWinG ConDitions:
1. $\tAu_x=\bIgcup cov_X$ IS A TOpolOgy On $X$;
2. iF $\mathCal{U}\subseteq \tAu_X$ And $\mAThcAl{U}$ is FinitE, theN $\mAthcaL{U}\in CoV_X$;
3. if $\mAtHcal{U}\in Cov_X$ and $V\In\taU_X$, then $\{U\caP V: U\In\MatHcAl{U}\}\in cOv_X$;
4. if $\mAthCal{u}\in Cov_X$ And, for eACh $U\In\MATHcAl{U}$, we have $\mathcal{V}(U)\In cOV_X$ Such that $u=\bigcuP\MaThCAl{V}(U)$, then $\BiGcuP_{U\in\MAThcal{u}}\matHCaL{V}(U)\in Cov_x$;
5. if $\matHCaL{U}\SubseteQ \tAu_X$ and $\MaThcAl{V}\In Cov_x$ Are sUch thaT $\bigcup\mAthcaL{u}=\bigcup\mathcal{v}$ And, for each $V\in\MAtHCAl{v}$, TherE exIsts $U\in\mathCal{U}$ SUch tHat $V\SUbSetEQ U$, theN $\mathCaL{u}\iN | }$ of points of $G$ whichconvergesin $\ mat hbb {R }$ t o $x $. The set $K= \ {x\} \cup\{x_n: n\in\omega\ }$ is c o mpac t i n $\m athbb{S } $b u t $ K\ ca p F $i snot c los ed in $ \mathbb{S} $.Th is proves th a t$\mathbb{S }$is a $k$-spa ce.
It i scon s isten t w ith $ \mathb f {ZF}$that theSo r genfre y line i s no t a$k$-space.
FromT he o rems 2.28 and3.10,we im m e dia tel y obtain t he foll o wing:
L et $ \ tau $ be the topol ogy of theS org enfrey l ine . If ev ery c om p act in $(\math bb{R }, \tau^{ \star} ) $ set i s closed in $( \ma thb b{R} , \ ta u)$ ,t hen $\ mat h bf{ CC}(c\ma th bb {R})$ hol d s .
Com pac t co mplem ent partial t opo logy === ===== ===== ==== == ===== ====== =====
Let us slightly ref ormulateDef in iti on 2.1o f [@OP SW] :
A *part ial top o log y* o f a set $X$ is a coll ec t i on $Cov_X\ subset e q\ ma t hcal{P}( \m ath cal{ P } (X))$ whi c hsatisfie s thef ol lo wing co nd itions :
1. $ \tau_ X =\bi gcup C ov_X$ is a to p ology on $X$;2. if $\math c al { U }\ s ubse teq \tau_X$ an d $\ m athc al{U } $isf inite , the n$ \m a thcal{U}\in Cov_X$;
3. if $\ma thcal{U}\in C ov_X$ and$ V \ in\tau_X $, t h en $\{U\cap V: U\ in\ma thcal{U}\} \ in Cov_X $;
4 . if $\ mathcal{U } \ in Cov_X $ a nd, fo r e a c h$U\in\mathcal { U }$,we have $ \ma thcal{V }(U )\i n C ov_ X$ such tha t $U=\bi gc up \m at hca l{V}( U )$, then $ \bi gc up_ {U\in \ mathca l{U}} \mat hc al { V}( U)\in C o v_ X $ ;
5 . i f $\ mat hc al{U} \sub s ete q \tau_ X$ and $\ mat h cal{ V} \i n Cov_X $ are such th at $\bigcup\ ma thc al{U}= \ b igcup\ma thcal{V}$ and, for each $V\in\m ath cal{V }$,there exi sts $U\in \ma t hcal{U }$ suc h tha t$V\ s u bsete q U$ , t he n $\mathca l { U}\ in | }$ of_points of_$G$ which converges in_$\mathbb{R}$ to_$x$._The set_$K=\{x\}\cup\{x_n:_n\in\omega\}$ is compact_in $\mathbb{S}$ but_$K\cap F$ is not_closed in $\mathbb{S}$._This_proves that $\mathbb{S}$ is a $k$-space.
It is consistent with $\mathbf{ZF}$ that the Sorgenfrey line_is_not a_$k$-space.
From_Theorems_2.28 and 3.10, we immediately_obtain the following:
Let $\tau$ be_the topology_of the Sorgenfrey line. If every compact in_$(\mathbb{R},_\tau^{\star})$ set is_closed in $(\mathbb{R}, \tau)$, then $\mathbf{CC}(c\mathbb{R})$ holds.
Compact complement partial_topology
===================================
Let us slightly reformulate Definition 2.1_of [@OPSW]:
A *partial_topology*_of_a set $X$ is_a collection $Cov_X\subseteq\mathcal{P}(\mathcal{P}(X))$ which satisfies the_following conditions:
1. $\tau_X=\bigcup Cov_X$ is_a topology on $X$;
2. if $\mathcal{U}\subseteq_\tau_X$ and $\mathcal{U}$ is finite, then_$\mathcal{U}\in Cov_X$;
3. if $\mathcal{U}\in_Cov_X$ and_$V\in\tau_X$, then $\{U\cap V: U\in\mathcal{U}\}\in_Cov_X$;
4. if_$\mathcal{U}\in Cov_X$_and, for each_$U\in\mathcal{U}$, we have $\mathcal{V}(U)\in Cov_X$ such_that $U=\bigcup\mathcal{V}(U)$, then_$\bigcup_{U\in\mathcal{U}}\mathcal{V}(U)\in Cov_X$;
5. if $\mathcal{U}\subseteq \tau_X$_and_$\mathcal{V}\in Cov_X$ are_such_that_$\bigcup\mathcal{U}=\bigcup\mathcal{V}$ and,_for each $V\in\mathcal{V}$,_there_exists $U\in\mathcal{U}$_such_that $V\subseteq U$, then $\mathcal{U}\in |
^{\rm Adj.}\right)_{ji}$.
[^5]: Here all fields are still dimensionless. The continuum expectation values $\frac{\<\bar \psi \left({\bm 1},\gamma_0\right)\psi\>^{\rm continuum}}{\sqrt{g^2N}}$ are given by dividing the r.h.s. of Eqs. (\[psibarpsi\])–(\[psidagpsi\]) by $a\sqrt{g^2N}$ ($g$ has dimensions of mass in $1+1$).
[^6]: The integral over $\varphi$ can be either restricted to obey $\varphi_a \ge \varphi_{b}$ for $a>b$, or can be simply unrestricted. The difference between these two choices will be reflected by whether the operators $S(P)$ include permutations of the eigenvalues or not.
[^7]: Here we implicitly assume that we take the large-$N$ limit when all other parameters are fixed. These includes the volume $L$, the temperature $T$, the mass $m_q$, the ‘t Hooft coupling $g^2N$, and the baryon number $B$.
[^8]: To ensure that ${\cal U}(\cal C)$ is unitary, the coordinates ${\cal C}$ need to obey certain conditions and in the next section we make these explicit.
[^9]: We thank V. P. Nair for pointing this to us and to L. G. Yaffe for a discussion related to this.
[^10]: According to Eqs. (\[Pk\])–(\[PkPk’\]) the terms with $q=q'\neq Q$ are also nonzero, but using the representation in Eqs. (\[constraints\_res1\])–(\[constraints\_res2\]) we verified that they are subleading in $N$ if we take $M\to\infty$ with or after we take $N\to \infty$.
[^11]: In our numerical studies we find that, while irrelevant for the minimization of ${\cal H}_F$, the last term in is crucial to include in order to get the right baryon mass.
[^12]: Another paper that discusses nonzero baryon number in the ‘t Hooft model is [@SchonThies], but there the authors restrict to translation invariant dense systems, which as they show, is inconsistent. For translation non-invariant states, the authors discuss only the vicinity of the | ^{\rm Adj.}\right)_{ji}$.
[ ^5 ]: Here all fields are still dimensionless. The continuum expectation values $ \frac{\<\bar \psi \left({\bm 1},\gamma_0\right)\psi\>^{\rm continuum}}{\sqrt{g^2N}}$ are give by separate the r.h.s. of Eqs. (\[psibarpsi\])–(\[psidagpsi\ ]) by $ a\sqrt{g^2N}$ ($ g$ has dimensions of mass in $ 1 + 1 $).
[ ^6 ]: The integral over $ \varphi$ can be either restrict to obey $ \varphi_a \ge \varphi_{b}$ for $ a > b$, or can be simply unrestricted. The deviation between these two choices will be reflected by whether the operator $ S(P)$ include permutations of the eigenvalue or not.
[ ^7 ]: Here we implicitly assume that we take the large-$N$ limit when all other argument are fixed. These includes the volume $ L$, the temperature $ T$, the mass $ m_q$, the ‘ t Hooft match $ g^2N$, and the baryon number $ B$.
[ ^8 ]: To ensure that $ { \cal U}(\cal C)$ is unitary, the coordinate $ { \cal C}$ need to obey certain condition and in the adjacent section we draw these explicit.
[ ^9 ]: We thank V. P. Nair for pointing this to us and to L. G. Yaffe for a discussion related to this.
[ ^10 ]: According to Eqs. (\[Pk\])–(\[PkPk’\ ]) the price with $ q = q'\neq Q$ are also nonzero, but using the representation in Eqs. (\[constraints\_res1\])–(\[constraints\_res2\ ]) we verified that they are subleading in $ N$ if we take $ M\to\infty$ with or after we drive $ N\to \infty$.
[ ^11 ]: In our numerical studies we find that, while irrelevant for the minimization of $ { \cal H}_F$, the last term in is crucial to include in order to get the right baryon mass.
[ ^12 ]: Another composition that discusses nonzero baryon number in the ‘ metric ton Hooft mannequin is [ @SchonThies ], but there the authors restrict to transformation invariant dense system, which as they show, is inconsistent. For transformation non - invariant states, the authors discuss only the vicinity of the | ^{\rm Wdj.}\right)_{ji}$.
[^5]: Here all fielas are still dimensionness. Tge contivuum expectation values $\frac{\<\uar \psi \ltyt({\bm 1},\gamma_0\right)\psi\>^{\rm continuul}}{\sqrt{g^2N}}$ qre jiven by dividinj the r.h.s. of Esd. (\[psiyacpsi\])–(\[psidagpsi\]) bi $a\sqrt{g^2N}$ ($g$ vas dimensions ow lass in $1+1$).
[^6]: The integral over $\varphi$ cwn be eotjer restricted to pfey $\barphi_a \ge \varphi_{b}$ for $a>b$, or can bs simplj unrestricted. Thr difference between these two choices will be rfflected by wheeyer the operxtors $S(P)$ include permujations of the eigenvalues or nog.
[^7]: Herz we implicutoy wvsume that xe takv the large-$N$ limit whan all pther parametevs arx fized. These includes thx volume $L$, the tempetature $T$, tve mass $m_q$, the ‘t Hoifr coukling $g^2N$, xbd ghe berykn numher $B$.
[^8]: To ensude that ${\cal U}(\cal C)$ is unitary, uhe bpordinates ${\czl C}$ nqeq to obey certain conditions and in the nest section we make thesw explicit.
[^9]: We thank V. K. Nair for [ointing this to us and to L. G. Yaffe for a discussimn remxteb to tfus.
[^10]: According to Eqs. (\[Pk\])–(\[PkPk’\]) the terms with $q=q'\neq Z$ ate also nonzero, nut using the reprrsfnywtion in Eqs. (\[cunstranhta\_res1\])–(\[constraints\_res2\]) we veryfied that ther arr subleading in $N$ if we takw $M\to\infty$ wpth ir after we take $N\co \infty$.
[^11]: In uur mumerocal studies we find thct, whime irrelevajt for ths minimization of ${\cak V}_F$, the last term in is crusial to iicludz in orddr tp get ehe right haryok mass.
[^12]: Another papeg thaj discgsses nonzfro baryon number in the ‘t Hoofv model is [@ScnotThpes], but tkere tme authors restwict to translction inrariang dense syatems, wiich as they show, is incotdistent. For vranslatijn nin-incariant rtates, the autnors disclsf onoy the vicinity of thg | ^{\rm Adj.}\right)_{ji}$. [^5]: Here all fields are The expectation values \psi \left({\bm 1},\gamma_0\right)\psi\>^{\rm the of Eqs. (\[psibarpsi\])–(\[psidagpsi\]) $a\sqrt{g^2N}$ ($g$ has of mass in $1+1$). [^6]: The over $\varphi$ can be either restricted to obey $\varphi_a \ge \varphi_{b}$ for $a>b$, can be simply unrestricted. The difference between these two choices will be reflected whether operators include of the eigenvalues or not. [^7]: Here we implicitly assume that we take the large-$N$ limit all other parameters are fixed. These includes the $L$, the temperature $T$, mass $m_q$, the ‘t Hooft $g^2N$, the baryon $B$. To that ${\cal U}(\cal is unitary, the coordinates ${\cal C}$ need to obey certain conditions and in the next section we these explicit. thank V. Nair pointing to us and G. Yaffe for a discussion related According to Eqs. (\[Pk\])–(\[PkPk’\]) the terms with $q=q'\neq are also but using the representation in Eqs. we verified that they are subleading in $N$ we take $M\to\infty$ with or after we take $N\to \infty$. [^11]: In our numerical studies that, while irrelevant for minimization of ${\cal the term is to include order to get the right baryon mass. [^12]: Another paper that nonzero baryon number in the ‘t Hooft model is [@SchonThies], the restrict to translation dense systems, which as show, inconsistent. For translation non-invariant authors only the | ^{\rm Adj.}\right)_{ji}$.
[^5]: Here all fields Are still diMensiOnlEss. thE conTinuUm expectation vALues $\Frac{\<\bar \psi \left({\bm 1},\gamma_0\rIght)\pSi\>^{\RM conTInUum}}{\sqRt{g^2N}}$ are GIvEN By dIvIdIng ThE R.h.S. of EqS. (\[psIbarpsi\])–(\[Psidagpsi\]) bY $a\sQrT{g^2N}$ ($g$ has dimenSIoNs of mass in $1+1$).
[^6]: the Integral over $\VarPhi$ can Be EitHEr resTriCted tO obey $\vARphi_a \gE \varphi_{b}$ fOr $A>B$, or can BE simply UNReStriCted. The difference BEtWEen these two choIces wiLl BE rEFLecTed By whether tHe OperaTOrs $S(P)$ inCLuDE PErmUTations of the eIgenvalues oR Not.
[^7]: here we ImPliCItly asSume tHaT We tAke the large-$n$ limIt when all Other pARameterS Are fixeD. These IncLudEs thE VoLuMe $L$, ThE TemPErAtuRE $T$, tHe mass $m_q$, ThE ‘t hooft CoupLING $G^2N$, anD thE barYon nuMber $B$.
[^8]: To ensure ThaT ${\cal u}(\Cal c)$ is unItary, The cOoRdinaTes ${\cal c}$ need To Obey certain condItioNs and in thE neXt SecTiOn we mAKe thesE exPliCit.
[^9]: We thAnk V. P. NaIR foR pOINTiNg this to us and to L. G. YAfFE FoR a discusSion reLAtEd TO this.
[^10]: AccOrDinG to EQS. (\[pk\])–(\[PkPK’\]) the TErMs with $q=q'\Neq Q$ arE AlSo Nonzero, BuT using ThE rePreSentaTIon iN Eqs. (\[coNstraintS\_res1\])–(\[cONstraints\_res2\]) we VErified that thEY aRE SuBLeadIng In $N$ if we take $m\to\iNFty$ wIth oR AfTer WE take $n\to \inFtY$.
[^11]: in OUr numerical studies wE fInd thaT, whilE irrelevant foR the minimiZATIon of ${\cal h}_F$, thE LaST term in is cruciAl to iNclude in orDEr to get tHe rigHt baryon Mass.
[^12]: AnothER Paper thaT diScuSseS noNZErO baryon number IN The ‘t hoOft modeL is [@schonThIes], But TheRe tHe Authors reStrict to TrAnSlAtIon InvarIAnt dense SySteMs, WhiCh as tHEy show, Is incOnsiStEnT. for TranslaTIoN NOn-inVaRiAnt sTatEs, The auThorS DisCuss onlY the vicinIty OF the | ^{\rm Adj.}\right)_{ji}$.
[^5]: Her e all fi eld sarestil l dimensionles s . Th e continuum expectatio n val ue s $\f r ac {\<\b ar \psi \l e f t({ \b m1}, \g a mm a_0\r igh t)\psi\ >^{\rm con tin uu m}}{\sqrt{g^ 2 N} }$ are giv enby dividingthe r.h.s .ofE qs. ( \[p sibar psi\]) – (\[psi dagpsi\]) b y $a\sq r t{g^2N} $ ($ g$ h as dimensions ofm as s in $1+1$).
[ ^6]: T he in t e gra l o ver $\varp hi $ can be eith e rr e s tri c ted to obey $ \varphi_a \ g e \ varphi _{ b}$ for $a >b$,or can be simplyunre stricted. The d i fferenc e betwee n thes e t wochoi c es w ill b e re f le cte d by whether t he oper ator s $ S (P)$ in clud e per mutations ofthe eig e nva luesor no t.
[^ 7]: H ere we impl ic itly assume tha t we take the la rg e-$ N$ limi t whenall ot her par ameters are f i x e d. These includes th ev o lu me $L$,the te m pe ra t ure $T$, t hemass $ m_q$, the ‘t Hooft c ouplin g $ g^ 2N$, an dthe ba ry onnum ber $ B $.
[^8]:To ensur e tha t ${\cal U}(\ca l C)$ is unita r y, t he coor din ates ${\cal C}$ need too be y c e rtain cond it i on s and in the next se ct ion we make these explic it.
[^9]: W e thank V . P. Na i r for pointing this to us and to L. G. Yaff e for adiscussio n relatedtothi s.
[^ 1 0 ]: According to E qs.(\ [Pk\])– (\[ PkPk’\] ) t heter mswi th $q=q'\ neq Q$ a re a ls onon zero, but usin gthe r epr esent a tion i n Eqs . (\ [c on s tra ints\_r e s1 \ ] )–(\ [c on stra int s\ _res2 \])w e v erified that the y a r e su bl ea ding in $N$ if we ta ke $M\to\inf ty $ w ith or a fter wetake $N\to \infty$.
[^ 1 1]: Inour nume rica l studies we findtha t , whil e irre levan tfor t he mi n i mi zat io n of ${\ca l H}_ F$, t he las t termin is crucial to i n clu de in order t o g et t h e r igh t b a ryo nm ass .
[^12]: Another paper tha td is cusses non z ero b aryon n umber i n the ‘t Hoof t model i s [@Schon Th ies] , but there the authors restrict to tr a ns latio n i nvaria nt de nse s ystems , wh ich a s they s how, i s inc on sistent. For translation non-in varian t sta tes , the aut hor s di scuss onl y th e vicinity of th e | ^{\rm Adj.}\right)_{ji}$.
[^5]:_Here all_fields are still dimensionless._The continuum_expectation_values $\frac{\<\bar_\psi_\left({\bm 1},\gamma_0\right)\psi\>^{\rm continuum}}{\sqrt{g^2N}}$_are given by_dividing the r.h.s. of_Eqs. (\[psibarpsi\])–(\[psidagpsi\]) by $a\sqrt{g^2N}$_($g$_has dimensions of mass in $1+1$).
[^6]: The integral over $\varphi$ can be either restricted_to_obey $\varphi_a_\ge_\varphi_{b}$_for $a>b$, or can be_simply unrestricted. The difference between_these two_choices will be reflected by whether the operators_$S(P)$_include permutations of_the eigenvalues or not.
[^7]: Here we implicitly assume that_we take the large-$N$ limit when_all other parameters_are_fixed._These includes the volume_$L$, the temperature $T$, the mass_$m_q$, the ‘t Hooft coupling $g^2N$,_and the baryon number $B$.
[^8]: To ensure_that ${\cal U}(\cal C)$ is unitary,_the coordinates ${\cal C}$ need_to obey_certain conditions and in the_next section we_make these_explicit.
[^9]: We thank_V. P. Nair for pointing this to us_and to L. G. Yaffe_for a discussion related to this.
[^10]:_According_to Eqs. (\[Pk\])–(\[PkPk’\]) the_terms_with_$q=q'\neq Q$_are also nonzero,_but_using the_representation_in Eqs. (\[constraints\_res1\])–(\[constraints\_res2\]) we verified that they_are_subleading in $N$ if we take $M\to\infty$_with or after we_take_$N\to \infty$.
[^11]: In our_numerical studies we find that,_while irrelevant for the minimization of_${\cal H}_F$,_the last_term in is crucial to include in order to get the_right baryon mass.
[^12]: Another paper that_discusses nonzero baryon number_in the_‘t_Hooft model is_[@SchonThies],_but there_the authors restrict to translation invariant dense_systems, which_as they show, is inconsistent. For_translation non-invariant states, the_authors_discuss only the vicinity of the |
Evaluation at positive integers defines a natural, surjective, equivariant $*$[-]{}homomorphism $\pi\colon T(A_m,\alpha_m^n)\to \bigoplus A_m$. Its kernel is naturally isomorphic to $\bigoplus \Sigma A_m$. Thus we obtain a natural extension $$0 \longrightarrow \bigoplus \Sigma A_m
\overset{\iota}\longrightarrow T(A_m,\alpha_m^n)
\overset{\pi}\longrightarrow \bigoplus A_m
\longrightarrow 0.$$ Build the diagram for this extension. The map $\bigoplus \Sigma A_n\to \operatorname{cone}(\pi)$ is a homotopy equivalence in a natural and hence equivariant fashion. Hence the extension is admissible. Moreover, one easily identifies the map $\Sigma\bigl(\bigoplus
A_m\bigr)\to\bigoplus \Sigma A_m$ with $S-{{\mathrm{id}}}$, where $S$ is the shift map defined above. Rotating the extension triangle, we obtain an exact triangle $$\xymatrix@1@C=3.5em{
T(A_m,\alpha_m^n) \ar[r]^-{-\pi} &
\bigoplus A_m \ar[r]^-{{{\mathrm{id}}}-S} &
\bigoplus A_m \ar[r]^-{\Sigma^{-1}\iota} &
\Sigma^{-1} T(A_m,\alpha_m^n).
}$$ This implies $\Sigma^{-1} T(A_m,\alpha_m^n) \cong \operatorname{ho-\!\varinjlim}(A_m,\alpha_m^n)$ and hence the assertion.
To obtain a concrete criterion for admissibility, we let $\tilde{T}(A_m,\alpha_m^n)$ be the variant of $T(A_m,\alpha_m^n)$ where we require $\lim_{t\to\infty} \alpha_m^\infty(f_m(t))$ to exist in $A_\infty$ instead of $\lim f_m(t)=0$. The algebra $\tilde{T}(A_m,\alpha_m^n)$ is equivariantly contractible in a natural way. The contracting homotopy is obtained by making sense of the formula $H_s f(t){\mathrel{:=}}\alpha_{[st | Evaluation at positive integers defines a natural, surjective, equivariant $ * $ [ -]{}homomorphism $ \pi\colon T(A_m,\alpha_m^n)\to \bigoplus A_m$. Its kernel is naturally isomorphous to $ \bigoplus \Sigma A_m$. therefore we prevail a natural elongation $ $ 0 \longrightarrow \bigoplus \Sigma A_m
\overset{\iota}\longrightarrow T(A_m,\alpha_m^n)
\overset{\pi}\longrightarrow \bigoplus A_m
\longrightarrow 0.$$ build up the diagram for this extension. The map $ \bigoplus \Sigma A_n\to \operatorname{cone}(\pi)$ is a homotopy comparison in a lifelike and hence equivariant fashion. Hence the extension is admissible. Moreover, one well identifies the map $ \Sigma\bigl(\bigoplus
A_m\bigr)\to\bigoplus \Sigma A_m$ with $ S-{{\mathrm{id}}}$, where $ S$ is the shift map define above. Rotating the elongation triangle, we obtain an accurate triangle $ $ \xymatrix@1@C=3.5em {
T(A_m,\alpha_m^n) \ar[r]^-{-\pi } &
\bigoplus A_m \ar[r]^-{{{\mathrm{id}}}-S } &
\bigoplus A_m \ar[r]^-{\Sigma^{-1}\iota } &
\Sigma^{-1 } T(A_m,\alpha_m^n).
} $ $ This implies $ \Sigma^{-1 } T(A_m,\alpha_m^n) \cong \operatorname{ho-\!\varinjlim}(A_m,\alpha_m^n)$ and hence the affirmation.
To obtain a concrete criterion for admissibility, we let $ \tilde{T}(A_m,\alpha_m^n)$ be the variant of $ T(A_m,\alpha_m^n)$ where we require $ \lim_{t\to\infty } \alpha_m^\infty(f_m(t))$ to exist in $ A_\infty$ alternatively of $ \lim f_m(t)=0$. The algebra $ \tilde{T}(A_m,\alpha_m^n)$ is equivariantly contractible in a natural way. The contracting homotopy is obtained by making common sense of the formula $ H_s f(t){\mathrel{:=}}\alpha_{[st |
Evapuation at positive inteners defines a ncrural, vurjecfive, equkvariant $*$[-]{}homomorphism $\pi\coloi T(A_n,\alphq_m^n)\to \bigoplus A_m$. Its yernel is naturaloy iwimorphic to $\bigoplmf \Sifla A_k$. Thus we obtaik a natural extension $$0 \lotgfiyhtarrow \bigoplus \Sigma A_m
\overset{\ijta}\longtihhtarrow T(A_m,\alkha_m^n)
\ovedset{\pi}\longrightarrow \bigoplus A_m
\mongrigitarrow 0.$$ Build yhe diagram for this extendion. The map $\bigoplus \Digma A_n\to \iperwrorname{cone}(\pk)$ is a homotopy equivamence in a natural and hence eqjivarnant fashiob. Yenfg the extensmon is admissible. Moreover, one eaxily identifiex tie mqp $\Sigma\bigl(\bigoplus
A_m\bigr)\to\bigoplus \Figma A_m$ fich $S-{{\mathrm{id}}}$, where $S$ iw rhe svift map eefknes ebobe. Rotwtiig the extehsion triantle, we obtain an exsce triangle $$\xymztrix@1@C=3.5qm{
T(A_m,\alpha_m^n) \ar[r]^-{-\pi} &
\bigoplus A_m \ar[g]^-{{{\matgrm{id}}}-S} &
\bigoplus A_m \qr[r]^-{\Sigma^{-1}\iota} &
\Sigmw^{-1} T(A_m,\alphw_m^n).
}$$ This implies $\Sigma^{-1} T(A_m,\alpha_m^n) \cong \operatortame{hk-\!\xarnkmlim}(X_n,\appha_m^n)$ and hence the assertion.
To obtain a consdeue briterion for admlssibility, we let $\yipdr{J}(A_m,\alpha_m^n)$ be jhe varnznf of $T(A_m,\alpha_m^n)$ whfre we tequirw $\lim_{t\to\igfty} \alpha_m^\infty(f_m(t))$ to exist in $Q_\infty$ instecd if $\lim f_m(t)=0$. The algzbra $\tilde{T}(A_o,\alpna_m^n)$ os equivariantly contraetible in a naturwl way. Ths contracting homutoiy iv obtainta by making sense of the firmuka $H_s f(g){\matnrel{:=}}\al[ha_{[st | Evaluation at positive integers defines a natural, $*$[-]{}homomorphism T(A_m,\alpha_m^n)\to \bigoplus Its kernel is A_m$. we obtain a extension $$0 \longrightarrow \Sigma A_m \overset{\iota}\longrightarrow T(A_m,\alpha_m^n) \overset{\pi}\longrightarrow \bigoplus \longrightarrow 0.$$ Build the diagram for this extension. The map $\bigoplus \Sigma A_n\to is a homotopy equivalence in a natural and hence equivariant fashion. Hence the is Moreover, easily the map $\Sigma\bigl(\bigoplus A_m\bigr)\to\bigoplus \Sigma A_m$ with $S-{{\mathrm{id}}}$, where $S$ is the shift map defined above. the extension triangle, we obtain an exact triangle T(A_m,\alpha_m^n) \ar[r]^-{-\pi} & \bigoplus \ar[r]^-{{{\mathrm{id}}}-S} & \bigoplus A_m \ar[r]^-{\Sigma^{-1}\iota} \Sigma^{-1} }$$ This $\Sigma^{-1} \cong and hence the To obtain a concrete criterion for admissibility, we let $\tilde{T}(A_m,\alpha_m^n)$ be the variant of $T(A_m,\alpha_m^n)$ where we $\lim_{t\to\infty} \alpha_m^\infty(f_m(t))$ in $A_\infty$ of f_m(t)=0$. algebra $\tilde{T}(A_m,\alpha_m^n)$ is in a natural way. The contracting by making sense of the formula $H_s f(t){\mathrel{:=}}\alpha_{[st |
Evaluation at positive integErs defines A natuRal, SurJeCtivE, equIvariant $*$[-]{}homomoRPhisM $\pi\colon T(A_m,\alpha_m^n)\to \biGopluS A_M$. its kERnEl is nAturallY IsOMOrpHiC tO $\biGoPLuS \SigmA A_m$. thus we oBtain a natuRal ExTension $$0 \longrIGhTarrow \bigoPluS \Sigma A_m
\overSet{\Iota}\loNgRigHTarroW T(A_M,\alphA_m^n)
\oveRSet{\pi}\lOngrightaRrOW \bigopLUs A_m
\lonGRIgHtarRow 0.$$ Build the diagraM FoR This extension. THe map $\bIgOPlUS \sigMa A_N\to \operatoRnAme{coNE}(\pi)$ is a hOMoTOPY eqUIvalence in a naTural and henCE eqUivariAnT faSHion. HeNce thE eXTenSion is admisSiblE. Moreover, One easILy identIFies the Map $\SigMa\bIgl(\BigoPLuS
A_M\biGr)\TO\biGOpLus \sIgmA A_m$ with $S-{{\MaThRm{id}}}$, wHere $s$ IS THe shIft Map dEfineD above. RotatinG thE extENsiOn triAngle, We obTaIn an eXact trIanglE $$\xYmatrix@1@C=3.5em{
T(A_m,\alPha_m^N) \ar[r]^-{-\pi} &
\bigOplUs a_m \aR[r]^-{{{\MathrM{Id}}}-S} &
\bigOplUs A_M \ar[r]^-{\SigMa^{-1}\iota} &
\SIGma^{-1} t(A_M,\ALPhA_m^n).
}$$ This implies $\SigmA^{-1} T(a_M,\AlPha_m^n) \conG \operaTOrNaME{ho-\!\varinJlIm}(A_M,\alpHA_M^n)$ and HencE ThE assertiOn.
To obTAiN a ConcretE cRiteriOn For AdmIssibILity, We let $\tIlde{T}(A_m,\aLpha_m^N)$ Be the variant of $t(a_m,\alpha_m^n)$ wherE We REQuIRe $\liM_{t\tO\infty} \alpha_M^\infTY(f_m(t))$ To exISt In $A_\INfty$ iNsteaD oF $\LiM F_m(t)=0$. The algebra $\tilde{T}(a_m,\Alpha_m^N)$ is eqUivariantly coNtractible IN A Natural wAy. ThE CoNTracting homotoPy is oBtained by mAKing sensE of thE formula $h_s f(t){\mathrEL{:=}}\Alpha_{[st |
Evaluation at positive i ntegers de fines anat ur al,surj ective, equiva r iant $*$[-]{}homomorphism$\pi\ co l on T ( A_ m,\al pha_m^n ) \t o \bi go pl usA_ m $. Itsker nel isnaturallyiso mo rphic to $\b i go plus \Sigm a A _m$. Thus we ob tain a n atu r al ex ten sion$$0 \l o ngrigh tarrow \b ig o plus \ S igma A_ m \ove rset{\iota}\longr i gh t arrow T(A_m,\a lpha_m ^n )
\ ove rse t{\pi}\lon gr ighta r row \bi g op l u s A_ m
\longright arrow 0.$$B uil d thedi agr a m for this e x ten sion. The m ap $ \bigoplus \Sigm a A_n\to \operat orname {co ne} (\pi ) $is aho m oto p yequ i val ence inana tural and h e n ce e qui vari ant f ashion. Hence th e ex t ens ion i s adm issi bl e. Mo reover , one e asily identifie s th e map $\S igm a\ big l( \bigo p lus
A_m \bi gr)\to\ bigoplu s \S ig m a A_ m$ with $S-{{\math rm { i d} }}$, whe re $S$ is t h e shiftma p d efin e d abov e. R o ta ting the exten s io ntriangl e, we ob ta inanexact tria ngle $ $\xymatr ix@1@ C =3.5em{
T( A _m,\alpha_m^n ) \ a r [r ] ^-{- \pi } &
\bi gopl u s A_ m \a r [r ]^- { {{\ma thrm{ id } }} - S} &
\bigoplusA_ m \ar[ r]^-{ \Sigma^{-1}\i ota} &
\ S igma^{-1 } T( A _m , \alpha_m^n).
}$$This impli e s $\Sigm a^{-1 } T(A_m, \alpha_m^ n ) \cong \ ope rat orn ame { h o- \!\varinjlim} ( A _m,\ al pha_m^n )$and hen cethe as ser ti on.
To o btain aco nc re te cr iteri o n for ad mi ssi bi lit y, we let $\ tilde {T}( A_ m, \ alp ha_m^n) $ b e theva ri antof$T (A_m, \alp h a_m ^n)$ wh ere we re qui r e $\ li m_ {t\to\i nfty} \alpha_ m^ \infty(f_m (t ))$ to ex i s t in $A_ \infty$ instead of $\li m f_m(t) =0$ . The alg ebra $\ti lde {T}(A_ m,\ a lpha_m ^n)$ i s equ iv ari a n tly c o n tr act ib le in a na t u ral way. T he c ontract ing homotopy is ob t ain ed by makingsen se o f th e f o rm u la$H _ s f ( t ){\mathrel{:=}} \alpha_{[s t |
Evaluation at_positive integers_defines a natural, surjective,_equivariant $*$[-]{}homomorphism_$\pi\colon_T(A_m,\alpha_m^n)\to \bigoplus_A_m$._Its kernel is_naturally isomorphic to_$\bigoplus \Sigma A_m$. Thus_we obtain a_natural_extension $$0 \longrightarrow \bigoplus \Sigma A_m
\overset{\iota}\longrightarrow T(A_m,\alpha_m^n)
\overset{\pi}\longrightarrow \bigoplus A_m
_\longrightarrow_0.$$ Build_the_diagram _for this extension. The map_$\bigoplus \Sigma A_n\to \operatorname{cone}(\pi)$ is_a homotopy_equivalence in a natural and hence equivariant fashion._Hence_the extension is_admissible. Moreover, one easily identifies the map $\Sigma\bigl(\bigoplus
_A_m\bigr)\to\bigoplus \Sigma A_m$ with $S-{{\mathrm{id}}}$, where $S$_is the shift_map_defined_above. Rotating the extension_triangle, we obtain an exact triangle_$$\xymatrix@1@C=3.5em{
T(A_m,\alpha_m^n) \ar[r]^-{-\pi}_&
\bigoplus A_m \ar[r]^-{{{\mathrm{id}}}-S}_&
\bigoplus A_m_\ar[r]^-{\Sigma^{-1}\iota} &
_\Sigma^{-1} T(A_m,\alpha_m^n).
_ }$$ This implies $\Sigma^{-1}_T(A_m,\alpha_m^n) \cong \operatorname{ho-\!\varinjlim}(A_m,\alpha_m^n)$_and hence_the assertion.
To obtain_a concrete criterion for admissibility, we_let $\tilde{T}(A_m,\alpha_m^n)$ be_the variant of $T(A_m,\alpha_m^n)$ where we_require_$\lim_{t\to\infty} \alpha_m^\infty(f_m(t))$ to_exist_in $A_\infty$_instead of_$\lim f_m(t)=0$. The_algebra_$\tilde{T}(A_m,\alpha_m^n)$ is_equivariantly_contractible in a natural way. The_contracting_homotopy is obtained by making sense of_the formula $H_s f(t){\mathrel{:=}}\alpha_{[st |
4.64}$$ $(iii)$ Suppose that ${\text{\rm{dom}}}(B)\cap{\text{\rm{dom}}}(B^*)\supseteq{\text{\rm{dom}}}(A)$ and that $B$ $($resp., $B^*$$)$ is $A^{1-\varepsilon}$-compact for some $\varepsilon \in (0,1)$. In addition, assume that ${\text{\rm{dom}}}\big(A^{1/2}\big) = {\text{\rm{dom}}}\big((A^*)^{1/2}\big)$. Then $B$ $($resp., $B^*$$)$ is $A_{{\mathfrak{R}}}$-form compact. Moreover, equation and relations hold as well.
$(i)$ Since by hypothesis $B$ and $B^*$ are $A_{{\mathfrak{R}}}$-bounded and $B$ (resp., $B^*$) is $A_{{\mathfrak{R}}}$-compact, Theorem \[t3.5\] implies that $$\begin{aligned}
\begin{split}
& |B|^{1/2}(A_{{\mathfrak{R}}} + I_{{{\mathcal H}}})^{-1/2} =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}}) {\label}{4.70} \\
& \big(\text{resp., } \, |B^*|^{1/2}(A_{{\mathfrak{R}}} + I_{{{\mathcal H}}})^{-1/2} =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B^*|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}})\big).
\end{split}\end{aligned}$$ At this point one can follow the proof of Theorem \[t4.11\]$(i)$, noting that each operator in and contains at least one compact factor from.
$(ii)$ Again, one can follow the proof of Theorem \[t4.11\]$(ii)$, noting that the right-hand side of (resp., ) contains a compact factor from. Similarly, the right-hand side of and the analogous equations with $A$ replaced by $A^*$ (resp., $B$ replaced by $B^*$) contains at least one compact factor from. Relations and are clear from Theorem \[t3.5\] since $A_{{\mathfrak{R | 4.64}$$ $ (iii)$ Suppose that $ { \text{\rm{dom}}}(B)\cap{\text{\rm{dom}}}(B^*)\supseteq{\text{\rm{dom}}}(A)$ and that $ B$ $ ($ resp. , $ B^*$$)$ is $ A^{1-\varepsilon}$-compact for some $ \varepsilon \in (0,1)$. In addition, assume that $ { \text{\rm{dom}}}\big(A^{1/2}\big) = { \text{\rm{dom}}}\big((A^*)^{1/2}\big)$. Then $ B$ $ ($ resp. , $ B^*$$)$ is $ A_{{\mathfrak{R}}}$-form compact. furthermore, equality and relation back hold as well.
$ (i)$ Since by guess $ B$ and $ B^*$ are $ A_{{\mathfrak{R}}}$-bounded and $ B$ (resp. , $ B^*$) is $ A_{{\mathfrak{R}}}$-compact, Theorem \[t3.5\ ] implies that $ $ \begin{aligned }
\begin{split }
& |B|^{1/2}(A_{{\mathfrak{R } } } + I_{{{\mathcal H}}})^{-1/2 } =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B|^{1/2 } } \in { { \mathcal B}}_\infty({{\mathcal H } }) { \label}{4.70 } \\
& \big(\text{resp. , } \, |B^*|^{1/2}(A_{{\mathfrak{R } } } + I_{{{\mathcal H}}})^{-1/2 } =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B^*|^{1/2 } } \in { { \mathcal B}}_\infty({{\mathcal H}})\big).
\end{split}\end{aligned}$$ At this point one can trace the proof of Theorem \[t4.11\]$(i)$, noting that each hustler in and control at least one compact gene from.
$ (ii)$ Again, one can follow the proof of Theorem \[t4.11\]$(ii)$, notice that the correct - hand slope of (resp. ,) contains a compact component from. Similarly, the right - hand side of and the analogous equation with $ A$ replaced by $ A^*$ (resp. , $ B$ replaced by $ B^*$) contains at least one compendious factor from. relation and are clear from Theorem \[t3.5\ ] since $ A_{{\mathfrak{R | 4.64}$$ $(iil)$ Suppose that ${\text{\rm{dom}}}(N)\cap{\text{\rm{dom}}}(B^*)\sukswteq{\teet{\rm{dom}}}(Z)$ and thxt $B$ $($resp., $B^*$$)$ is $A^{1-\varepsilon}$-colpqct fir some $\varepsilon \in (0,1)$. In additpon, assumw thet ${\text{\rm{dom}}}\big(A^{1/2}\ujg) = {\texb{\xm{dom}}}\glg((A^*)^{1/2}\biy)$. Vhen $B$ $($resp., $B^*$$)$ ix $A_{{\mathfran{R}}}$-form compact. Murzover, equation and relations hold as well.
$(i)$ Xijce by hypothefis $N$ and $B^*$ are $A_{{\mathfrak{R}}}$-bounded and $B$ (resl., $B^*$) is $E_{{\mathfrak{R}}}$-compavt, Theorem \[t3.5\] implies that $$\hegij{aligned}
\begin{split}
& |B|^{1/2}(A_{{\mathfrak{T}}} + I_{{{\iqthcal H}}})^{-1/2} =
\ovefline{(A_{{\mathfrak{R}}}+I_{{{\mathcam H}}})^{-1/2}|B|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}}) {\lcbel}{4.70} \\
& \big(\tezt{eeso., } \, |B^*|^{1/2}(A_{{\mathfrek{R}}} + I_{{{\iathcal H}}})^{-1/2} =
\ovcgline{(A_{{\mdthfrak{T}}}+I_{{{\mathcal H}}})^{-1/2}|B^*|^{1/2}} \ik {{\matical B}}_\infty({{\mathcal H}})\big).
\env{split}\end{aligned}$$ At jhis point oue can follow the proif of Tveoram \[t4.11\]$(k)$, noginf vhaf each opxrator in ahd contains at least one compavt dactor from.
$(ii)$ Again, oge can follow the proof of Theorem \[t4.11\]$(ii)$, totjng that the right-hand wide of (resp., ) containd a compast factor from. Similarly, the right-hand side of ang the xnaoonous wqkations with $A$ replaced by $A^*$ (resp., $B$ replaced fg $N^*$) bontains at least one compacy vavjor from. Relatkons aus zre clear from Thelrem \[t3.5\] fince $A_{{\mathfrah{R | 4.64}$$ $(iii)$ Suppose that ${\text{\rm{dom}}}(B)\cap{\text{\rm{dom}}}(B^*)\supseteq{\text{\rm{dom}}}(A)$ and that $B^*$$)$ $A^{1-\varepsilon}$-compact for $\varepsilon \in (0,1)$. = Then $B$ $($resp., is $A_{{\mathfrak{R}}}$-form compact. equation and relations hold as well. Since by hypothesis $B$ and $B^*$ are $A_{{\mathfrak{R}}}$-bounded and $B$ (resp., $B^*$) is Theorem \[t3.5\] implies that $$\begin{aligned} \begin{split} & |B|^{1/2}(A_{{\mathfrak{R}}} + I_{{{\mathcal H}}})^{-1/2} = \overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal \in B}}_\infty({{\mathcal {\label}{4.70} & \big(\text{resp., } \, |B^*|^{1/2}(A_{{\mathfrak{R}}} + I_{{{\mathcal H}}})^{-1/2} = \overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B^*|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}})\big). \end{split}\end{aligned}$$ this point one can follow the proof of \[t4.11\]$(i)$, noting that each in and contains at least compact from. $(ii)$ one follow proof of Theorem noting that the right-hand side of (resp., ) contains a compact factor from. Similarly, the right-hand side and the with $A$ by (resp., replaced by $B^*$) least one compact factor from. Relations from Theorem \[t3.5\] since $A_{{\mathfrak{R | 4.64}$$ $(iii)$ Suppose that ${\text{\rm{dom}}}(B)\cAp{\text{\rm{doM}}}(B^*)\supSetEq{\tExT{\rm{dOm}}}(A)$ aNd that $B$ $($resp., $B^*$$)$ is $a^{1-\VarePsilon}$-compact for some $\vaRepsiLoN \In (0,1)$. In ADdItion, Assume tHAt ${\TEXt{\rM{dOm}}}\Big(a^{1/2}\bIG) = {\tExt{\rm{Dom}}}\Big((A^*)^{1/2}\big)$. then $B$ $($resp., $B^*$$)$ Is $A_{{\MaThfrak{R}}}$-form cOMpAct. MoreoveR, eqUation and relAtiOns holD aS weLL.
$(i)$ SinCe bY hypoThesis $b$ And $B^*$ arE $A_{{\mathfraK{R}}}$-BOunded ANd $B$ (resp., $b^*$) IS $A_{{\MathFrak{R}}}$-compact, TheorEM \[t3.5\] IMplies that $$\begiN{alignEd}
\BEgIN{SplIt}
& |B|^{1/2}(a_{{\mathfrak{R}}} + i_{{{\mAthcaL h}}})^{-1/2} =
\overliNE{(A_{{\MATHfrAK{R}}}+I_{{{\mathcal H}}})^{-1/2}|B|^{1/2}} \iN {{\mathcal B}}_\inFTy({{\mAthcal h}}) {\lAbeL}{4.70} \\
& \Big(\texT{resp., } \, |b^*|^{1/2}(A_{{\MAthFrak{R}}} + I_{{{\mathcAl H}}})^{-1/2} =
\oVerline{(A_{{\mAthfraK{r}}}+I_{{{\mathcAL H}}})^{-1/2}|B^*|^{1/2}} \in {{\maThcal B}}_\InfTy({{\mAthcAL H}})\BiG).
\enD{sPLit}\ENd{AliGNed}$$ at this poInT oNe can FollOW THE proOf oF TheOrem \[t4.11\]$(I)$, noting that eaCh oPeraTOr iN and cOntaiNs at LeAst onE compaCt facToR from.
$(ii)$ Again, one Can fOllow the pRooF oF ThEoRem \[t4.11\]$(iI)$, Noting ThaT thE right-hAnd side OF (reSp., ) CONTaIns a compact factor fRoM. sImIlarly, thE right-HAnD sIDe of and tHe AnaLogoUS EquatIons WItH $A$ replacEd by $A^*$ (rESp., $b$ rEplaced By $b^*$) contaInS at LeaSt one COmpaCt factOr from. ReLatioNS and are clear frOM Theorem \[t3.5\] sincE $a_{{\mATHfRAk{R | 4.64}$$ $(iii)$ Suppose th at ${\text {\rm{ dom }}} (B )\ca p{\t ext{\rm{dom}}} ( B^*) \supseteq{\text{\rm{do m}}}( A) $ and th at $B $ $($re s p. , $B^ *$ $) $ i s$ A^ {1-\v are psilon} $-compactfor s ome $\vareps i lo n \in (0,1 )$. In addition , a ssumeth at$ {\tex t{\ rm{do m}}}\b i g(A^{1 /2}\big)={ \text{ \ rm{dom} } } \b ig(( A^*)^{1/2}\big)$. Th e n $B$ $($resp. , $B^* $$ ) $i s $A _{{ \mathfrak{ R} }}$-f o rm comp a ct . M ore o ver, equation and relati o nshold a swel l .
$(i )$ Si nc e by hypothesis $B$ and $B^* $ are$ A_{{\ma t hfrak{R }}}$-b oun ded and $B $(re sp . , $ B ^* $)i s $ A_{{\mat hf ra k{R}} }$-c o m p a ct,The orem \[t3 .5\] impliestha t $$ \ beg in{al igned }
\b eg in{sp lit}
& |B|^ {1 /2}(A_{{\mathfr ak{R }}} + I_{ {{\ ma thc al H}}} ) ^{-1/2 } =
\o verline {(A_{{\ m ath fr a k { R} }}+I_{{{\mathcal H }} } ) ^{ -1/2}|B| ^{1/2} } \ in {{\mathc al B} }_\i n f ty({{ \mat h ca l H}}) {\la b el }{ 4.70} \ \& \big (\ tex t{r esp., } \, |B^*| ^{1/2}(A _{{\m a thfrak{R}}} +I _{{{\mathcalH }} } ) ^{ - 1/2} =\overline{( A_{{ \ math frak { R} }}+ I _{{{\ mathc al H} } })^{-1/2}|B^*|^{1/2 }} \in { {\mat hcal B}}_\inf ty({{\math c a l H}})\bi g).\ en d {split}\end{al igned }$$ At thi s point o ne ca n follow the proo f of Theor em\[t 4.1 1\] $ ( i) $, noting tha t each o perator in and co nta ins at le as t one com pact fac to rfr om .
$(ii) $ Again,on e c an fo llowt he pro of of The or em \[t 4.11\]$ ( ii ) $ , no ti ng tha t t he righ t-ha n d s ide of(resp., ) co n tain sacompact factor from. S imilarly,th e r ight-h a n d side o f and the analogous equ a tions w ith $A$repl aced by $ A^* $ (res p., $B$ re placed by $ B^ *$) c ontai n s a t l ea st one com p a ctfacto rfrom . Relat ions and are clear fro m Theorem \[t 3.5 \] s i n ce $A _ {{ \ mat hf r ak{ R | 4.64}$$ $(iii)$_Suppose that_${\text{\rm{dom}}}(B)\cap{\text{\rm{dom}}}(B^*)\supseteq{\text{\rm{dom}}}(A)$ and that $B$_$($resp., $B^*$$)$_is_$A^{1-\varepsilon}$-compact for_some_$\varepsilon \in (0,1)$._In addition, assume_that ${\text{\rm{dom}}}\big(A^{1/2}\big) = {\text{\rm{dom}}}\big((A^*)^{1/2}\big)$._Then $B$ $($resp.,_$B^*$$)$_is $A_{{\mathfrak{R}}}$-form compact. Moreover, equation and relations hold as well.
$(i)$ Since by hypothesis $B$_and_$B^*$ are_$A_{{\mathfrak{R}}}$-bounded_and_$B$ (resp., $B^*$) is $A_{{\mathfrak{R}}}$-compact,_Theorem \[t3.5\] implies that $$\begin{aligned}
\begin{split}
&_|B|^{1/2}(A_{{\mathfrak{R}}} +_I_{{{\mathcal H}}})^{-1/2} =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal H}}})^{-1/2}|B|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}})__ {\label}{4.70}_\\
& \big(\text{resp., } \, |B^*|^{1/2}(A_{{\mathfrak{R}}} + I_{{{\mathcal H}}})^{-1/2} =
\overline{(A_{{\mathfrak{R}}}+I_{{{\mathcal_H}}})^{-1/2}|B^*|^{1/2}} \in {{\mathcal B}}_\infty({{\mathcal H}})\big).
\end{split}\end{aligned}$$ At_this point one_can_follow_the proof of Theorem_\[t4.11\]$(i)$, noting that each operator in_and contains at least one compact_factor from.
$(ii)$ Again, one can follow the_proof of Theorem \[t4.11\]$(ii)$, noting that_the right-hand side of (resp.,_) contains_a compact factor from. Similarly,_the right-hand side_of and_the analogous equations_with $A$ replaced by $A^*$ (resp.,_$B$ replaced by_$B^*$) contains at least one compact_factor_from. Relations and_are_clear_from Theorem_\[t3.5\] since $A_{{\mathfrak{R |
employ this mapping throughout this section in treatment of all point group symmetries.
Rotoreflection Symmetry $S_{2n}$
--------------------------------
{height="\fg"} {height="\fg"} {height="\fg"}
The symmetry group is ${G={\mathbb{Z}}_{2n}}$ with generator $\hat{s}$ such that $$\hat s^{2n}=-(-1)^n.$$ It acts on the 3D space as $$\hat s(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix})\hat s^{-1}=(\begin{smallmatrix}\cos\frac{2\pi}{2n}&\sin\frac{2\pi}{2n}\\-\sin\frac{2\pi}{2n}&\cos\frac{2\pi}{2n}\end{smallmatrix})(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix}),{{\;\;\:}}\hat s\gamma_3 \hat s^{-1}=-\gamma_3.$$ We set $$s = e^{\gamma_1\gamma_2\frac{2\pi}{4n}}\hat s\gamma_3,$$ which satisfies $${{\;\;\:}}s y_i s^{-1}= -y_i,{{\;\;\:}}s^{2n}=1.$$ Here we used the equality $s^{2n}=(-e^{-\gamma_1\gamma_2\frac{2\pi}{2n}}\hat s^2)^n=1$.
For $n=1$ we add $s$ as a generator such that ${\mathbb{R}}\langle s\rangle={{\mathrm{Cl}}}_{0,1}$ and ${{{\mathrm{Cl}}}_{p,q}\hat{\otimes}{{\mathrm{Cl}}}_{0,1}={{\mathrm{Cl}}}_{p,q+1}}$, and get $${{\mathrm{Cl}}}_{q+7,3+1 | employ this mapping throughout this section in treatment of all item group isotropy.
Rotoreflection Symmetry $ S_{2n}$
--------------------------------
! [ The $ S_{2n}$ rotoreflection symmetry point groups: $ S_2,S_4,S_6$.](Ci_3D.png " fig:"){height="\fg " } ! [ The $ S_{2n}$ rotoreflection isotropy point groups: $ S_2,S_4,S_6$.](S4_3D.png " fig:"){height="\fg " } ! [ The $ S_{2n}$ rotoreflection symmetry period groups: $ S_2,S_4,S_6$.](S6_3D.png " fig:"){height="\fg " }
The symmetry group is $ { G={\mathbb{Z}}_{2n}}$ with generator $ \hat{s}$ such that $ $ \hat s^{2n}=-(-1)^n.$$ It dissemble on the 3D space as $ $ \hat s(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix})\hat s^{-1}=(\begin{smallmatrix}\cos\frac{2\pi}{2n}&\sin\frac{2\pi}{2n}\\-\sin\frac{2\pi}{2n}&\cos\frac{2\pi}{2n}\end{smallmatrix})(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix}),{{\;\;\:}}\hat s\gamma_3 \hat s^{-1}=-\gamma_3.$$ We set $ $ mho = e^{\gamma_1\gamma_2\frac{2\pi}{4n}}\hat s\gamma_3,$$ which satisfies $ $ { { \;\;\:}}s y_i s^{-1}= -y_i,{{\;\;\:}}s^{2n}=1.$$ Here we used the equality $ s^{2n}=(-e^{-\gamma_1\gamma_2\frac{2\pi}{2n}}\hat s^2)^n=1$.
For $ n=1 $ we total $ s$ as a generator such that $ { \mathbb{R}}\langle s\rangle={{\mathrm{Cl}}}_{0,1}$ and $ { { { \mathrm{Cl}}}_{p, q}\hat{\otimes}{{\mathrm{Cl}}}_{0,1}={{\mathrm{Cl}}}_{p, q+1}}$, and beget $ $ { { \mathrm{Cl}}}_{q+7,3 + 1 | emoloy this mapping througmout this sectiou in trxatment of all ooint group symmetries.
Rotorehlecrion Wymmetry $S_{2n}$
--------------------------------
{gcight="\yg"} {height="\fg"} {height="\fg"}
The symmetry group js ${G={\matibb{Z}}_{2n}}$ with genetator $\hat{s}$ such that $$\hat s^{2j}=-(-1)^n.$$ Ih acts on the 3D spwce as $$\hat w(\begyb{smallmatrix}\eamma_1\\\gamma_2\tnb{smallmatris})\hat s^{-1}=(\begin{smallmatrix}\cos\frac{2\pi}{2v}&\sin\fxac{2\pi}{2n}\\-\sin\frqc{2\pi}{2n}&\fms\frac{2\pi}{2n}\env{smalliatrix})(\begin{smallmatrhx}\gamma_1\\\bamma_2\end{smallmstrmx}),{{\;\;\:}}\har s\gamma_3 \hat s^{-1}=-\gamma_3.$$ Wx set $$s = e^{\gamma_1\gamma_2\srac{2\pi}{4n}}\had a\gamma_3,$$ which satiwfues $${{\;\;\:}}s y_i v^{-1}= -y_i,{{\;\;\:}}r^{2b}=1.$$ Hdre wx uaed thf esuality $s^{2n}=(-s^{-\gamma_1\gamma_2\drac{2\pi}{2n}}\hat s^2)^n=1$.
For $n=1$ ee qdd $s$ as a geheratow fuch that ${\mathbb{R}}\langle s\rangle={{\mathrm{Cl}}}_{0,1}$ ans ${{{\mathrm{Cl}}}_{p,q}\hat{\otimes}{{\marhrm{Cl}}}_{0,1}={{\mathrm{Cl}}}_{p,q+1}}$, and het $${{\mathri{Cl}}}_{q+7,3+1 | employ this mapping throughout this section in all group symmetries. Symmetry $S_{2n}$ -------------------------------- groups: "fig:"){height="\fg"} {height="\fg"} The symmetry group is ${G={\mathbb{Z}}_{2n}}$ with generator $\hat{s}$ such that $$\hat It acts on the 3D space as $$\hat s(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix})\hat s^{-1}=(\begin{smallmatrix}\cos\frac{2\pi}{2n}&\sin\frac{2\pi}{2n}\\-\sin\frac{2\pi}{2n}&\cos\frac{2\pi}{2n}\end{smallmatrix})(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix}),{{\;\;\:}}\hat s\gamma_3 \hat s^{-1}=-\gamma_3.$$ set = s\gamma_3,$$ satisfies $${{\;\;\:}}s y_i s^{-1}= -y_i,{{\;\;\:}}s^{2n}=1.$$ Here we used the equality $s^{2n}=(-e^{-\gamma_1\gamma_2\frac{2\pi}{2n}}\hat s^2)^n=1$. For $n=1$ we add as a generator such that ${\mathbb{R}}\langle s\rangle={{\mathrm{Cl}}}_{0,1}$ and and get $${{\mathrm{Cl}}}_{q+7,3+1 | employ this mapping throughoUt this sectIon in TreAtmEnT of aLl poInt group symmetRIes.
ROtoreflection Symmetry $S_{2N}$
--------------------------------
{hEigHt="\fg"} {height="\fg"} {heIgHT="\fg"}
The SYmmetry GROuP is ${G={\Mathbb{Z}}_{2n}}$ with generAToR $\Hat{s}$ such that $$\haT s^{2n}=-(-1)^n.$$ It AcTS oN THe 3D SpaCe as $$\hat s(\beGiN{smalLMatrix}\gAMmA_1\\\GAMma_2\ENd{smallmatrix})\Hat s^{-1}=(\begin{smALlmAtrix}\cOs\FraC{2\Pi}{2n}&\sin\Frac{2\pI}{2n}\\-\SIn\fRac{2\pi}{2n}&\cos\frAc{2\pi}{2N}\end{smallMatrix})(\BEgin{smaLLmatrix}\Gamma_1\\\gAmmA_2\enD{smaLLmAtRix}),{{\;\;\:}}\HaT S\gaMMa_3 \Hat S^{-1}=-\GamMa_3.$$ We set $$s = E^{\gAmMa_1\gamMa_2\frAC{2\PI}{4N}}\hat S\gaMma_3,$$ wHich sAtisfies $${{\;\;\:}}s y_i s^{-1}= -y_I,{{\;\;\:}}s^{2n}=1.$$ here WE usEd the EqualIty $s^{2N}=(-e^{-\Gamma_1\Gamma_2\fRac{2\pi}{2N}}\hAt s^2)^n=1$.
For $n=1$ we add $s$ aS a geNerator suCh tHaT ${\maThBb{R}}\laNGle s\raNglE={{\maThrm{Cl}}}_{0,1}$ aNd ${{{\mathrM{cl}}}_{p,Q}\hAT{\OTiMes}{{\mathrm{Cl}}}_{0,1}={{\mathrm{CL}}}_{p,Q+1}}$, ANd Get $${{\mathrM{Cl}}}_{q+7,3+1 | employ this mapping throu ghout this sect ion in t reat ment of all pointg roup symmetries.
Rotorefl ectio nS ymme t ry $S_{ 2n}$
-- - -- - - --- -- -- --- -- - -- ----- --- --
 {heigh t ="\fg" }  {height="\ fg "} {h eigh t="\fg"}
The s y mmetryg roup is ${G={ \ma thb b{Z} } _{ 2n }}$ w i thg en era t or$\hat{s} $su ch th at $ $ \ h a t s^ {2n }=-( -1)^n .$$ It acts o n t he 3 D sp ace a s $$\ hats( \begi n{smal lmatr ix }\gamma_1\\\gam ma_2 \end{smal lma tr ix} )\ hat s ^ {-1}=( \be gin {smallm atrix}\ c os\ fr a c { 2\ pi}{2n}&\sin\frac{ 2\ p i }{ 2n}\\-\s in\fra c {2 \p i }{2n}&\c os \fr ac{2 \ p i}{2n }\en d {s mallmatr ix})(\ b eg in {smallm at rix}\g am ma_ 1\\ \gamm a _2\e nd{sma llmatrix }),{{ \ ;\;\:}}\hat s\ g amma_3 \hat s ^ {- 1 } =- \ gamm a_3 .$$ We set$$s= e^{ \gam m a_ 1\g a mma_2 \frac {2 \ pi } {4n}}\hat s\gamma_3 ,$ $ whic h sat isfies $${{\; \;\:}}s y_ i s ^{-1}= - y_i, { {\ ; \;\:}}s^{2n}=1 .$$ H ere we use d the equ ality $s^{2n} =(-e^{-\g a m ma_1\gam ma_ 2\f rac {2\ p i }{ 2n}}\hat s^2) ^ n =1$.
For $n= 1$we add$s$ as agen er ator such that ${ \m at hb b{ R}} \lang l e s\rang le ={{ \m ath rm{Cl } }}_{0, 1}$ a nd $ {{ {\ m ath rm{Cl}} } _{ p , q}\h at {\ otim es} {{ \math rm{C l }}} _{0,1}= {{\mathrm {Cl } }}_{ p, q+ 1}}$, a nd get $${{\m at hrm{Cl}}}_ {q +7, 3+1 | employ_this mapping_throughout this section in_treatment of_all_point group_symmetries.
Rotoreflection_Symmetry $S_{2n}$
--------------------------------
{height="\fg"} {height="\fg"} {height="\fg"}
The symmetry group is ${G={\mathbb{Z}}_{2n}}$ with_generator_$\hat{s}$ such_that_$$\hat_s^{2n}=-(-1)^n.$$ It acts on the_3D space as $$\hat s(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix})\hat_s^{-1}=(\begin{smallmatrix}\cos\frac{2\pi}{2n}&\sin\frac{2\pi}{2n}\\-\sin\frac{2\pi}{2n}&\cos\frac{2\pi}{2n}\end{smallmatrix})(\begin{smallmatrix}\gamma_1\\\gamma_2\end{smallmatrix}),{{\;\;\:}}\hat s\gamma_3_\hat s^{-1}=-\gamma_3.$$ We set $$s = e^{\gamma_1\gamma_2\frac{2\pi}{4n}}\hat s\gamma_3,$$_which_satisfies $${{\;\;\:}}s y_i_s^{-1}= -y_i,{{\;\;\:}}s^{2n}=1.$$ Here we used the equality $s^{2n}=(-e^{-\gamma_1\gamma_2\frac{2\pi}{2n}}\hat s^2)^n=1$.
For_$n=1$ we add $s$ as a_generator such that_${\mathbb{R}}\langle_s\rangle={{\mathrm{Cl}}}_{0,1}$_and ${{{\mathrm{Cl}}}_{p,q}\hat{\otimes}{{\mathrm{Cl}}}_{0,1}={{\mathrm{Cl}}}_{p,q+1}}$, and get_$${{\mathrm{Cl}}}_{q+7,3+1 |
i_2}+n_3{\alpha }_{i_3}$. Since $\det w\in
\{\pm 1\}$ and ${\mathrm{Vol}}_3({\alpha },\beta,\gamma )=1$, this implies that $m_2=n_3=1$. Further, $\beta -{\alpha }\notin R^a$, and hence $w(\beta )={\alpha }_{i_2}$ by Corollary \[co:cij\]. Since $\{{\alpha },\beta,\gamma \}$ is not a base for ${\mathbb{Z}}^I$ at $a$, we conclude that $w(\gamma )\not={\alpha }_{i_3}$. Then Corollary \[co:cij\] and the assumptions $\gamma -{\alpha }$, $\gamma -\beta
\notin R^a$ imply that $w(\gamma )\notin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpha }_{i_1}$ and $w(\gamma )\notin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpha }_{i_2}$. Thus the claim is proven.
\(2) By (1), $\{{\alpha },\beta \}$ is a base for $V^a({\alpha },\beta )$ at $a$. Thus ${\alpha }-k\beta \notin R^a$ for all $k\in {\mathbb{N}}$. The remaining claims follow by symmetry.
\(3) Suppose that ${\alpha }+\beta \notin R^a$. By (1) there exist $b\in A$, $w\in \operatorname{Hom}(a,b)$, $i_1,i_2,i_3\in I$ and $n_1,n_2\in {\mathbb{N}}$ such that $w({\alpha })={\alpha }_{i_1}$, $w(\beta )={\alpha }_{i_2}$, and $w(\gamma )={\alpha }_{i_3}+n_1{\alpha }_{i_1}+n_2{\alpha }_{i_2}\in R^b_+$. By Theorem \[root\_is\_sum\] there exist $n'_1,n'_2\in {\mathbb{N}}_0$ such that $n'_1\le n_1$, $n'_2\le n_2$, $n'_1+n'_2<n_1+n_2$, and $${\alpha }_{i_3}+n'_1{\ | i_2}+n_3{\alpha } _ { i_3}$. Since $ \det w\in
\{\pm 1\}$ and $ { \mathrm{Vol}}_3({\alpha }, \beta,\gamma) = 1 $, this implies that $ m_2 = n_3=1$. Further, $ \beta -{\alpha } \notin R^a$, and hence $ w(\beta) = { \alpha } _ { i_2}$ by Corollary \[co: cij\ ]. Since $ \{{\alpha }, \beta,\gamma \}$ is not a base for $ { \mathbb{Z}}^I$ at $ a$, we conclude that $ w(\gamma) \not={\alpha } _ { i_3}$. Then Corollary \[co: cij\ ] and the assumption $ \gamma -{\alpha } $, $ \gamma -\beta
\notin R^a$ entail that $ w(\gamma) \notin { \alpha } _ { i_3}+{\mathbb{N}}_0{\alpha } _ { i_1}$ and $ w(\gamma) \notin { \alpha } _ { i_3}+{\mathbb{N}}_0{\alpha } _ { i_2}$. Thus the claim is proven.
\(2) By (1), $ \{{\alpha }, \beta \}$ is a floor for $ V^a({\alpha }, \beta) $ at $ a$. Thus $ { \alpha } -k\beta \notin R^a$ for all $ k\in { \mathbb{N}}$. The remaining claim surveil by symmetry.
\(3) think that $ { \alpha } + \beta \notin R^a$. By (1) there exist $ b\in A$, $ w\in \operatorname{Hom}(a, b)$, $ i_1,i_2,i_3\in I$ and $ n_1,n_2\in { \mathbb{N}}$ such that $ w({\alpha }) = { \alpha } _ { i_1}$, $ w(\beta) = { \alpha } _ { i_2}$, and $ w(\gamma) = { \alpha } _ { i_3}+n_1{\alpha } _ { i_1}+n_2{\alpha } _ { i_2}\in R^b_+$. By Theorem \[root\_is\_sum\ ] there exist $ n'_1,n'_2\in { \mathbb{N}}_0 $ such that $ n'_1\le n_1 $, $ n'_2\le n_2 $, $ n'_1+n'_2 < n_1+n_2 $, and $ $ { \alpha } _ { i_3}+n'_1{\ | i_2}+n_3{\appha }_{i_3}$. Since $\det w\in
\{\pm 1\}$ and ${\mathrm{Vol}}_3({\copha },\bxta,\gammz )=1$, this kmplies that $m_2=n_3=1$. Further, $\beta -{\aopha }\botin R^a$, and hence $w(\bega )={\alpha }_{p_2}$ by Coroolarb \[co:cij\]. Since $\{{\alpiz },\beta,\gamma \}$ jd noc e base for ${\mathnb{Z}}^I$ at $a$, wa conclude thad $d(\gcmma )\not={\alpha }_{i_3}$. Then Corollary \[co:cij\] wnd the adsumptions $\gamia -{\ak[ha }$, $\gamma -\beta
\notin R^a$ imply that $w(\famma )\nmtin {\alpha }_{i_3}+{\msthbb{N}}_0{\alpha }_{i_1}$ and $w(\gamma )\nltin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpja }_{i_2}$. Thus tye cjqim is provev.
\(2) By (1), $\{{\alpha },\beta \}$ is a base for $V^a({\alpha },\beta )$ at $a$. Thjs ${\al'ha }-k\beta \nitun G^d$ for all $k\mn {\matrbb{N}}$. The remaining cnaims fpllow by symmebry.
\(3) Snppowe that ${\alpha }+\beta \novin R^a$. By (1) there exift $b\in A$, $f\iu \operatorname{Hom}(a,b)$, $i_1,u_2,i_3\un I$ dnd $t_1,n_2\in {\natfbb{H}}$ xudh thah $w({\elpha })={\alpha }_{i_1}$, $w(\beta )={\alpha }_{i_2}$, and $w(\gamma )={\alkha }_{p_3}+m_1{\alpha }_{i_1}+n_2{\alpga }_{i_2}\in R^f_+$. By Theorem \[root\_is\_sum\] there exist $n'_1,n'_2\in {\katgbb{N}}_0$ such that $n'_1\le n_1$, $n'_2\lw n_2$, $n'_1+n'_2<n_1+n_2$, and $${\alpha }_{i_3}+n'_1{\ | i_2}+n_3{\alpha }_{i_3}$. Since $\det w\in \{\pm 1\}$ },\beta,\gamma this implies $m_2=n_3=1$. Further, $\beta $w(\beta }_{i_2}$ by Corollary Since $\{{\alpha },\beta,\gamma is not a base for ${\mathbb{Z}}^I$ $a$, we conclude that $w(\gamma )\not={\alpha }_{i_3}$. Then Corollary \[co:cij\] and the assumptions -{\alpha }$, $\gamma -\beta \notin R^a$ imply that $w(\gamma )\notin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpha }_{i_1}$ $w(\gamma {\alpha }_{i_2}$. the claim is proven. \(2) By (1), $\{{\alpha },\beta \}$ is a base for $V^a({\alpha },\beta at $a$. Thus ${\alpha }-k\beta \notin R^a$ for $k\in {\mathbb{N}}$. The remaining follow by symmetry. \(3) Suppose ${\alpha \notin R^a$. (1) exist A$, $w\in \operatorname{Hom}(a,b)$, I$ and $n_1,n_2\in {\mathbb{N}}$ such that $w({\alpha })={\alpha }_{i_1}$, $w(\beta )={\alpha }_{i_2}$, and $w(\gamma )={\alpha }_{i_3}+n_1{\alpha }_{i_1}+n_2{\alpha R^b_+$. By there exist {\mathbb{N}}_0$ that n_1$, $n'_2\le n_2$, $${\alpha }_{i_3}+n'_1{\ | i_2}+n_3{\alpha }_{i_3}$. Since $\det w\in
\{\pm 1\}$ and ${\mAthrm{Vol}}_3({\alPha },\beTa,\gAmmA )=1$, tHis iMpliEs that $m_2=n_3=1$. FurtheR, $\Beta -{\Alpha }\notin R^a$, and hence $w(\bEta )={\alPhA }_{I_2}$ by CORoLlary \[Co:cij\]. SiNCe $\{{\ALPha },\BeTa,\GamMa \}$ IS nOt a baSe fOr ${\mathbB{Z}}^I$ at $a$, we coNclUdE that $w(\gamma )\nOT={\aLpha }_{i_3}$. Then COroLlary \[co:cij\] anD thE assumPtIonS $\Gamma -{\AlpHa }$, $\gamMa -\beta
\NOtin R^a$ Imply that $W(\gAMma )\notIN {\alpha }_{i_3}+{\MAThBb{N}}_0{\aLpha }_{i_1}$ and $w(\gamma )\notIN {\aLPha }_{i_3}+{\mathbb{N}}_0{\alpHa }_{i_2}$. ThuS tHE cLAIm iS prOven.
\(2) By (1), $\{{\alphA },\bEta \}$ is A Base for $v^A({\aLPHA },\beTA )$ at $a$. Thus ${\alpha }-K\beta \notin R^A$ For All $k\in {\MaThbB{n}}$. The reMainiNg CLaiMs follow by sYmmeTry.
\(3) SupposE that ${\aLPha }+\beta \NOtin R^a$. BY (1) there ExiSt $b\In A$, $w\IN \oPeRatOrNAme{hOm}(A,b)$, $i_1,I_2,I_3\in i$ and $n_1,n_2\in {\MaThBb{N}}$ suCh thAT $W({\ALpha })={\AlpHa }_{i_1}$, $w(\Beta )={\aLpha }_{i_2}$, and $w(\gammA )={\alPha }_{i_3}+N_1{\AlpHa }_{i_1}+n_2{\aLpha }_{i_2}\In R^b_+$. by theorEm \[root\_Is\_sum\] ThEre exist $n'_1,n'_2\in {\matHbb{N}}_0$ Such that $n'_1\Le n_1$, $N'_2\lE n_2$, $n'_1+N'_2<n_1+N_2$, and $${\aLPha }_{i_3}+n'_1{\ | i_2}+n_3{\alpha }_{i_3}$.Since $\de t w\i n
\{ \p m 1\ }$ a nd ${\mathrm{V o l}}_ 3({\alpha },\beta,\gam ma )= 1$ , thi s i mplie s that$ m_ 2 = n_3 =1 $. Fu rt h er , $\b eta -{\alp ha }\notin R^ a$ , and hence$ w( \beta )={\ alp ha }_{i_2}$byCoroll ar y \ [ co:ci j\] . Sin ce $\{ { \alpha },\beta, \g a mma \} $ is not a b asefor ${\mathbb{Z}} ^ I$ at $a$, we con cludeth a t$ w (\g amm a )\not={\ al pha } _ {i_3}$. Th e n Cor o llary \[co:ci j\] and the ass umptio ns $\ g amma - {\alp ha }$, $\gamma -\ beta
\notin R^a$i mply th a t $w(\g amma ) \no tin {\a l ph a}_{ i_ 3 }+{ \ ma thb b {N} }_0{\alp ha } _{i_1 }$ a n d $ w(\g amm a )\ notin {\alpha }_{i _3} +{\m a thb b{N}} _0{\a lpha } _{i_2 }$. Th us th eclaim is proven .
\ (2) By (1 ),$\ {{\ al pha } , \beta\}$ is a base for $V ^ a({ \a l p h a},\beta )$ at $a$. T h u s${\alpha }-k\b e ta \ n otin R^a $for all $ k\in{\ma t hb b{N}}$.The re m ai ni ng clai ms follo wbysym metry .
\( 3) Sup pose tha t ${\ a lpha }+\beta \ n otin R^a$. By (1 ) th e re e xis t $b\in A$, $w\ i n \o pera t or nam e {Hom} (a,b) $, $i _ 1,i_2,i_3\in I$ and $ n_1,n_ 2\in{\mathbb{N}}$ such that $ w ({\alpha })= { \a l pha }_{i_1}$,$w(\b eta )={\al p ha }_{i_ 2}$,and $w(\ gamma )={ \ a lpha }_{ i_3 }+n _1{ \al p h a}_{i_1}+n_2{\ a l pha}_ {i_2}\i n R ^b_+$.ByThe ore m \ [r oot\_is\_ sum\] th er eex is t $ n'_1, n '_2\in { \m ath bb {N} }_0$s uch th at $n '_1\ le n _ 1$, $n'_2\ l en _ 2$,$n '_ 1+n' _2< n_ 1+n_2 $, a n d $ ${\alph a }_{i_3} +n' _ 1{\ | i_2}+n_3{\alpha }_{i_3}$._Since $\det_w\in
\{\pm 1\}$_and ${\mathrm{Vol}}_3({\alpha_},\beta,\gamma_)=1$, this_implies_that $m_2=n_3=1$. Further,_$\beta -{\alpha }\notin_R^a$, and hence $w(\beta_)={\alpha }_{i_2}$ by_Corollary \[co:cij\]._Since $\{{\alpha },\beta,\gamma \}$ is not a base for ${\mathbb{Z}}^I$ at $a$, we conclude_that_$w(\gamma )\not={\alpha_}_{i_3}$._Then_Corollary \[co:cij\] and the assumptions $\gamma_-{\alpha }$, $\gamma -\beta
_\notin R^a$_imply that $w(\gamma )\notin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpha }_{i_1}$ and_$w(\gamma_)\notin {\alpha }_{i_3}+{\mathbb{N}}_0{\alpha_}_{i_2}$. Thus the claim is proven.
\(2) By (1), $\{{\alpha_},\beta \}$ is a base for_$V^a({\alpha },\beta )$_at_$a$._Thus ${\alpha }-k\beta \notin_R^a$ for all $k\in {\mathbb{N}}$. The_remaining claims follow by symmetry.
\(3) Suppose_that ${\alpha }+\beta \notin R^a$. By (1)_there exist $b\in A$, $w\in \operatorname{Hom}(a,b)$,_$i_1,i_2,i_3\in I$ and $n_1,n_2\in {\mathbb{N}}$_such that_$w({\alpha })={\alpha }_{i_1}$, $w(\beta )={\alpha_}_{i_2}$, and $w(\gamma_)={\alpha }_{i_3}+n_1{\alpha_}_{i_1}+n_2{\alpha }_{i_2}\in R^b_+$._By Theorem \[root\_is\_sum\] there exist $n'_1,n'_2\in {\mathbb{N}}_0$_such that $n'_1\le_n_1$, $n'_2\le n_2$, $n'_1+n'_2<n_1+n_2$, and $${\alpha_}_{i_3}+n'_1{\ |
**]{}.
. *Phase Transitions in Ferromagnetic Ising Models with Spatially Dependent Magnetic Fields*. Comm. Math. Phys., v. 337, p. 41–53, [**2015**]{}.
. *Phase Transition in Ferromagnetic Ising Models with Non-uniform External Magnetic Fields*. J. Stat. Phys., v. 139, p. 769–778, [**2010**]{}.
. *The Covariance Matrix of the Potts Model: A Random-Cluster Analysis*. J. Stat. Phys., v. 82, p. 1235–1297, [**1996**]{}.
. *Finite-size effects for the [P]{}otts model with weak boundary conditions*. J. Statist. Phys., v. 109, p. 67–131, [**2002**]{}.
. *Statistical Mechanics of Disordered Systems*. Cambridge University Press, [**2006**]{}.
. *Density and Uniqueness in Percolation*. Comm. Math. Phys., v. 121, p. 501–505, [**1989**]{}.
. *Interfaces for random-cluster models*. J. Statist. Phys., v. 111, p. 73–106, [**2003**]{}.
. *On the Gibbs States of the Noncritical Potts Model on $\mathbb{Z}^2$*. Probab. Theory Relat. Fields, v. 158, p. 477–512, [**2014**]{}.
. *A Finite-volume Version of Aizenman-Higuchi Theorem for the $2d$ Ising model*. Probab. Theory Relat. Fields, v. 153, p. 25–44, [**2012**]{}.
. *Graphical Representations for Ising Systems in External Fields*. J. Stat. Phys., v. 93, p. 17–32, [**1998**]{}.
. *Geometric Representations of Lattice Spin Models*. Spartacus Graduate, Cours Peccot, Collège de France, [**2015**]{}.
. *A New Proof of the Sharpness of the Phase Transition for [B]{}ernoulli Percolation and the [I]{}sing Model*. Preprint arXiv:1502.03050, [**2015**]{}.
. *Generalization of the Fortuin-Kasteleyn-Swendsen-Wang Representation and Monte Carlo Algorithm*. Phys. Rev. D., v. 38, p | * * ] { }.
. * Phase Transitions in Ferromagnetic Ising Models with Spatially Dependent Magnetic Fields *. Comm. Math. Phys. , v. 337, p. 41–53, [ * * 2015 * * ] { }.
. * Phase Transition in Ferromagnetic Ising Models with Non - uniform External Magnetic Fields *. J. Stat. Phys. , v. 139, p. 769–778, [ * * 2010 * * ] { }.
. * The Covariance Matrix of the Potts Model: A Random - Cluster Analysis *. J. Stat. Phys. , v. 82, p. 1235–1297, [ * * 1996 * * ] { }.
. * Finite - size effects for the [ P]{}otts model with weak boundary condition *. J. Statist. Phys. , v. 109, p. 67–131, [ * * 2002 * * ] { }.
. * Statistical Mechanics of Disordered Systems *. Cambridge University Press, [ * * 2006 * * ] { }.
. * concentration and Uniqueness in Percolation *. Comm. Math. Phys. , v. 121, p. 501–505, [ * * 1989 * * ] { }.
. * Interfaces for random - cluster model *. J. Statist. Phys. , v. 111, p. 73–106, [ * * 2003 * * ] { }.
. * On the Gibbs States of the Noncritical Potts Model on $ \mathbb{Z}^2 $ *. Probab. Theory Relat. battlefield, v. 158, p. 477–512, [ * * 2014 * * ] { }.
. * A Finite - volume Version of Aizenman - Higuchi Theorem for the $ 2d$ Ising model *. Probab. Theory Relat. field, v. 153, p. 25–44, [ * * 2012 * * ] { }.
. * Graphical Representations for Ising Systems in External Fields *. J. Stat. Phys. , v. 93, p. 17–32, [ * * 1998 * * ] { }.
. * Geometric Representations of Lattice Spin Models *. Spartacus Graduate, Cours Peccot, Collège de France, [ * * 2015 * * ] { }.
. * A New Proof of the Sharpness of the Phase Transition for [ B]{}ernoulli Percolation and the [ I]{}sing Model *. Preprint arXiv:1502.03050, [ * * 2015 * * ] { }.
. * Generalization of the Fortuin - Kasteleyn - Swendsen - Wang Representation and Monte Carlo Algorithm *. Phys. Rev. D., v. 38, p | **]{}.
. *Phwse Transitions in Ferrooagnetic Ising Models xith Spztially Aependent Magnetic Fields*. Colm. Math. Phys., v. 337, p. 41–53, [**2015**]{}.
. *Phase Travsition ij Ferromqgneuic Ising Models xjth Non-muiforj Extzrial Magnetic Figlds*. J. Stat. [hys., v. 139, p. 769–778, [**2010**]{}.
. *Tha Zorariance Matrix of the Potts Model: A Random-Vlkster Analysis*. J. Suat. Phya., v. 82, p. 1235–1297, [**1996**]{}.
. *Finite-size effects for ths [P]{}otts model with wrak boundary conditions*. J. Dtatlst. Phys., v. 109, p. 67–131, [**2002**]{}.
. *Stwtistical Mgdhagucs of Disoraered Systtmv*. Cambridgg University Press, [**2006**]{}.
. *Density and Jniquzness in Peteilahhon*. Comm. Mavh. Phyf., v. 121, p. 501–505, [**1989**]{}.
. *Intcgfaces xor rancom-cluster modcls*. J. Stqtist. Phys., v. 111, p. 73–106, [**2003**]{}.
. *On vhe Gibbs States of jhe Noncridieal Potts Model on $\marhvb{Z}^2$*. Ptobab. Theuey Felzt. Fjelds, g. 158, '. 477–512, [**2014**]{}.
. *A Finits-volume Verwion of Aizenman-Higicrp Theorem for the $2d$ Ifing model*. Probab. Theory Relat. Fields, v. 153, p. 25–44, [**2012**]{}.
. *Graphical Representarions for Ising Systels in Extqrnal Fields*. J. Stat. Phys., v. 93, p. 17–32, [**1998**]{}.
. *Geometric Represettatikvs if Uqthice Spin Models*. Spartacus Graduate, Cours Pecskt, Cpllège de Francc, [**2015**]{}.
. *A New Proof of yhf Xrarpness of tfe Phase Fransition for [B]{}erjoulli Kercolqtion and the [I]{}sing Model*. Preprint arXiv:1502.03050, [**2015**]{}.
. *Generalizanion of the Fortuin-Kasceleyn-Swendszn-Wang Reprrsentation and Monte Caxlo Alforithm*. Phyd. Rev. D., v. 38, p | **]{}. . *Phase Transitions in Ferromagnetic Ising Spatially Magnetic Fields*. Math. Phys., v. *Phase in Ferromagnetic Ising with Non-uniform External Fields*. J. Stat. Phys., v. 139, 769–778, [**2010**]{}. . *The Covariance Matrix of the Potts Model: A Random-Cluster Analysis*. Stat. Phys., v. 82, p. 1235–1297, [**1996**]{}. . *Finite-size effects for the [P]{}otts with boundary J. Phys., v. 109, p. 67–131, [**2002**]{}. . *Statistical Mechanics of Disordered Systems*. Cambridge University Press, [**2006**]{}. *Density and Uniqueness in Percolation*. Comm. Math. Phys., 121, p. 501–505, [**1989**]{}. *Interfaces for random-cluster models*. J. Phys., 111, p. [**2003**]{}. *On Gibbs States of Noncritical Potts Model on $\mathbb{Z}^2$*. Probab. Theory Relat. Fields, v. 158, p. 477–512, [**2014**]{}. . *A Finite-volume of Aizenman-Higuchi the $2d$ model*. Theory Fields, v. 153, [**2012**]{}. . *Graphical Representations for Ising Fields*. J. Stat. Phys., v. 93, p. 17–32, . *Geometric of Lattice Spin Models*. Spartacus Graduate, Peccot, Collège de France, [**2015**]{}. . *A New of the Sharpness of the Phase Transition for [B]{}ernoulli Percolation and the [I]{}sing Model*. Preprint . *Generalization of the Representation and Monte Algorithm*. Rev. v. p | **]{}.
. *Phase Transitions in FerromaGnetic IsinG ModeLs wIth spAtiaLly DEpendent MagnetIC FieLds*. Comm. Math. Phys., v. 337, p. 41–53, [**2015**]{}.
. *Phase transItIOn in fErRomagNetic IsINg mODelS wItH NoN-uNIfOrm ExTerNal MagnEtic Fields*. j. StAt. phys., v. 139, p. 769–778, [**2010**]{}.
. *The CovARiAnce Matrix Of tHe Potts Model: a RaNdom-ClUsTer aNalysIs*. J. stat. PHys., v. 82, p. 1235–1297, [**1996**]{}.
. *FINite-siZe effects FoR The [P]{}otTS model wITH wEak bOundary conditions*. j. stATist. Phys., v. 109, p. 67–131, [**2002**]{}.
. *StatIsticaL MEChANIcs Of DIsordered SYsTems*. CAMbridge uNiVERSitY press, [**2006**]{}.
. *Density aNd UniquenesS In PErcolaTiOn*. COMm. Math. phys., v. 121, P. 501–505, [**1989**]{}.
. *INTerFaces for ranDom-cLuster modEls*. J. StATist. PhyS., V. 111, p. 73–106, [**2003**]{}.
. *On the gibbs STatEs oF the nOnCrItiCaL potTS MOdeL On $\mAthbb{Z}^2$*. PrObAb. theorY RelAT. fIElds, V. 158, p. 477–512, [**2014**]{}.
. *A finiTe-volUme Version of AIzeNman-hIguChi ThEorem For tHe $2D$ IsinG model*. probaB. THeory Relat. FieldS, v. 153, p. 25–44, [**2012**]{}.
. *GRaphical REprEsEntAtIons fOR Ising sysTemS in ExteRnal FieLDs*. J. stAT. pHyS., v. 93, p. 17–32, [**1998**]{}.
. *Geometric RepresEnTATiOns of LatTice SpIN MOdELs*. SpartaCuS GrAduaTE, cours peccOT, COllège de france, [**2015**]{}.
. *a neW PRoof of tHe sharpnEsS of The phase tRansItion fOr [B]{}ernouLli PeRColation and the [i]{}Sing Model*. PrepRInT ARXIV:1502.03050, [**2015**]{}.
. *GenEraLization of tHe FoRTuin-kastELeYn-SWEndseN-Wang rePReSEntation and Monte CarLo algoriThm*. PhYs. Rev. D., v. 38, p | **]{}.
. *Phase Transitio ns in Ferr omagn eti c I si ng M odel s with Spatial l y De pendent Magnetic Field s*. C om m . Ma t h. Phys ., v. 3 3 7, p . 4 1– 53 , [ ** 2 01 5**]{ }.
. *Pha se Transit ion i n Ferromagne t ic Ising Mod els with Non-un ifo rm Ext er nal Magne tic Fiel ds*. J . Stat. Phys., v .1 39, p. 769–778 , [* *201 0**]{}.
. *The C o va r iance Matrix o f thePo t ts M ode l:A Random-C lu sterA nalysis * .J . Sta t . Phys., v. 8 2, p. 1235– 1 297 , [**1 99 6** ] {}.
. *Fin it e -si ze effectsforthe [P]{} otts m o del wit h weak b oundar y c ond itio n s* .J.St a tis t .Phy s .,v. 109,p. 6 7–131 , [* * 2 0 0 2**] {}.
.*Stat istical Mecha nic s of Dis order ed Sy stem s* . Cam bridge Univ er sity Press, [** 2006 **]{}.
. *D en sit yand U n iquene ssinPercola tion*.C omm .M a t h. Phys., v. 121, p. 5 0 1 –5 05, [**1 989**] { }.
. *Interf ac esforr a ndom- clus t er models* . J. S t at is t. Phys ., v. 11 1, p. 73 –106, [**2 003**] {}.
. * On th e Gibbs Stateso f the Noncrit i ca l Po t ts M ode l on $\math bb{Z } ^2$* . Pr o ba b.T heory Rela t. Fi e lds, v. 158, p. 477 –5 12, [* *2014 **]{}.
. *AFinite-vol u m e Version ofA iz e nman-Higuchi T heore m for the$ 2d$ Isin g mod el*. Pro bab. Theo r y Relat.Fie lds , v . 1 5 3 ,p. 25–44, [** 2 0 12** ]{ }.
. * Gra phicalRep res ent ati on s for Isi ng Syste ms i nEx ter nal F i elds*. J .Sta t. Ph ys.,v . 93,p. 17 –32, [ ** 1 998 **]{}..* G eome tr ic Rep res en tatio ns o f La ttice S pin Model s*. Spar ta cu s Gradu ate, Cours Pe cc ot, Collèg edeFrance , [**2015* *]{}.
. *A New Proof o f the Sh arp nessof t he PhaseTra nsitio n f o r [B]{ }ernou lli P er col a t ion a n d t he[I ]{}sing Mo d e l*. Prep ri nt a rXiv:15 02.03050, [**2015* * ]{} .
. *General iza tion o fthe Fo r tui n- K ast e l eyn-Swendsen-Wa ng Represe nt a ti on and Mon t e C ar lo Algo rithm*. Phys . Rev. D ., v. 38, p | **]{}.
. *Phase_Transitions in_Ferromagnetic Ising Models with_Spatially Dependent_Magnetic_Fields*. Comm._Math._Phys., v. 337,_p. 41–53, [**2015**]{}.
._*Phase Transition in Ferromagnetic_Ising Models with_Non-uniform_External Magnetic Fields*. J. Stat. Phys., v. 139, p. 769–778, [**2010**]{}.
. *The Covariance Matrix_of_the Potts_Model:_A_Random-Cluster Analysis*. J. Stat. Phys.,_v. 82, p. 1235–1297, [**1996**]{}.
._*Finite-size effects_for the [P]{}otts model with weak boundary conditions*._J._Statist. Phys., v._109, p. 67–131, [**2002**]{}.
. *Statistical Mechanics of Disordered Systems*._Cambridge University Press, [**2006**]{}.
. *Density and_Uniqueness in Percolation*._Comm._Math._Phys., v. 121, p._501–505, [**1989**]{}.
. *Interfaces for random-cluster models*._J. Statist. Phys., v. 111, p._73–106, [**2003**]{}.
. *On the Gibbs States of_the Noncritical Potts Model on $\mathbb{Z}^2$*._Probab. Theory Relat. Fields, v._158, p._477–512, [**2014**]{}.
. *A Finite-volume Version_of Aizenman-Higuchi Theorem_for the_$2d$ Ising model*._Probab. Theory Relat. Fields, v. 153,_p. 25–44, [**2012**]{}.
._*Graphical Representations for Ising Systems in_External_Fields*. J. Stat._Phys.,_v._93, p._17–32, [**1998**]{}.
. *Geometric_Representations_of Lattice_Spin_Models*. Spartacus Graduate, Cours Peccot, Collège_de_France, [**2015**]{}.
. *A New Proof of the_Sharpness of the Phase_Transition_for [B]{}ernoulli Percolation and_the [I]{}sing Model*. Preprint arXiv:1502.03050,_[**2015**]{}.
. *Generalization of the Fortuin-Kasteleyn-Swendsen-Wang Representation_and Monte_Carlo Algorithm*._Phys. Rev. D., v. 38, p |
associated with it in our reference field, the MUSE Ultra Deep Field (). In addition to being the deepest spectroscopic field so far observed with MUSE, all of the extracted MUSE spectra in this field have been visually inspected. In the field, as discussed in §\[sec:analysis\], we used two different procedures to extract the sources. Their redshifts were determined independently first, and then reconciled. Here we first present the basic properties of the objects associated with their redshift measurements for each extraction method independently. Based on our understanding of the differences and the relationship between the two source extraction methods and the redshift properties, we tried to find the best combination of these two methods to efficiently maximize the detection rate of redshifts in the even larger sample size of the MUSE Deep Field (the ).
### Redshifts of the continuum selected objects {#sec:result_cont}
Among ${854}$ continuum selected objects in the MUSE Ultra Deep Field (), we successfully measured the redshifts of ${282}$ objects in the redshift range from $z=0.21$ to $6.64$. The number of redshifts with confidence level $\geq 2$ is ${223}$.
{width="\textwidth"}
In the left two panels of Figure \[fig:u10\_musez\_hstpri\_org\], we show the distributions of redshifts found with the MUSE data. Because most of the objects at $0 < z < 1.5$ are usually identified by the [\[O<span style="font-variant:small-caps;">ii</span>\]]{}doublet, their CONFID is mostly 3 by definition when the double peaks are resolved and clearly seen. There are six cases in which CONFID is 2 in this redshift range, whose spectra in general have lower S/Ns but the features are obviously detected in narrowband images.
On the other hand, galaxies at $z > 3$ are often found by a single feature [Ly$\alpha$]{}, which gives CONFID of 2. For the case of $\rm CONFID=3$ at $z > 3$, the [C<span style="font-variant:small-caps;">iii</span>\]]{}emission or UV absorption features, and in some rare cases He<span style="font-variant:small-caps;">ii</span>, are also discerned. The number of determined | associated with it in our reference field, the MUSE Ultra Deep Field (). In addition to being the deep spectroscopic sphere so far observed with MUSE, all of the extracted MUSE spectra in this discipline have been visually inspected. In the field, as hash out in § \[sec: analysis\ ], we use two different procedure to extract the sources. Their red shift were determined independently first, and then reconciled. Here we first portray the basic properties of the objects associated with their redshift measurements for each origin method independently. Based on our understanding of the deviation and the relationship between the two source origin methods and the red shift properties, we tried to find the best combination of these two methods to efficiently maximize the detection pace of redshifts in the even larger sample size of the MUSE Deep Field (the).
# # # Redshifts of the continuum selected objects { # sec: result_cont }
Among $ { 854}$ continuum choose object in the MUSE Ultra Deep Field (), we successfully measured the redshifts of $ { 282}$ objects in the redshift range from $ z=0.21 $ to $ 6.64$. The number of redshifts with confidence level $ \geq 2 $ is $ { 223}$.
! [ image](figures/\UDFver_muse - z_prop_HSTpri_ORIG_fade.pdf){width="\textwidth " }
In the leftover two panel of Figure \[fig: u10\_musez\_hstpri\_org\ ], we show the distribution of red shift find with the MUSE data. Because most of the objects at $ 0 < omega < 1.5 $ are usually identify by the [ \[O < span style="font - variant: small - caps;">ii</span>\]]{}doublet, their CONFID is largely 3 by definition when the double peaks are resolved and clearly seen. There are six cases in which CONFID is 2 in this redshift range, whose spectra in general have gloomy S / Ns but the feature of speech are obviously detected in narrowband images.
On the other hand, galaxies at $ omega > 3 $ are often found by a single feature [ Ly$\alpha$ ] { }, which gives CONFID of 2. For the case of $ \rm CONFID=3 $ at $ z > 3 $, the [ C < span style="font - variant: small - caps;">iii</span>\]]{}emission or ultraviolet absorption features, and in some rare cases He < span style="font - variant: small - caps;">ii</span >, are besides discerned. The number of determined | asdociated with it in our veference field, jhw MUSE Ultra Deep Fidld (). In addition to being thx dewpest spectroscopic field su far obsvrved wity MUWW, all of tis extragced MHDE s'ertra in this figld have beet visually ins[eztzd. In the field, as discussed in §\[sec:agalysis\], wf used two difsereme prkbebures to extract the sources. Thsir redvhifts were drtermined independently figst, wnd then reconcilef. Here we furst present the casic proptrcies of the objects associated with their fedshnft measurenebts xor each exvractijn method independendly. Basrd on our undevstanving of the differences aid the relationship fetween tve two source extraxtuon mgthodv ana thd rtdsiiff propfrtmes, we tries to find tye best combination os these two mefhods eo efficiently maximize the detection ratt of dedshifts in the even lqrger sample size of jhe MUSE Dqep Field (the ).
### Redshifts of the continuum selecteg objxcgs {#weg:resjot_font}
Among ${854}$ continuum selected objects in the IHST Uktra Deep Fielb (), we successfulky mrwsured the reashifts or ${282}$ objects in the gedshifj rangw from $z=0.21$ uo $6.64$. Tne number of redshifts with confidence jwvel $\geq 2$ is ${223}$.
{width="\texcwidth"}
Jn the left two panemr of Figure \[fig:u10\_mjsed\_hst[ri\_org\], wt show the distribttions of redxhifts wounc with the MUSE fata. Nacause most of the objeets ad $0 < z < 1.5$ arf usually identified by the [\[O<spei style="font-vatiatt:skall-caps;">ni</span>\]]{}coublet, their CONFID is mosjly 3 by dzfinitkon when tge doubne peaks arq resolved ang clearly seei. There awe sux cqses in dhich CONFID ix 2 in this redshift range, whose spectva in feneral have loctr W/Ns but the feayurds wrv ouvioufny detected hn nxrrueband images.
On tme uthet hand, galaxies at $z > 3$ ade often found by s fingle fgature [Ly$\wlpha$]{}, which goves CONFID of 2. Fog the care of $\rm CONFID=3$ at $z > 3$, the [C<span style="ront-variajt:siall-caps;">iii</s[an>\]]{}enission or UR absorption features, and in some rare ceses He<span style="font-vaeiant:small-caps;">ii</spau>, sre also dixcernqd. The nukber of determined | associated with it in our reference field, Ultra Field (). addition to being far with MUSE, all the extracted MUSE in this field have been visually In the field, as discussed in §\[sec:analysis\], we used two different procedures to the sources. Their redshifts were determined independently first, and then reconciled. Here we present basic of objects associated with their redshift measurements for each extraction method independently. Based on our understanding of differences and the relationship between the two source methods and the redshift we tried to find the combination these two to maximize detection rate of in the even larger sample size of the MUSE Deep Field (the ). ### Redshifts of the selected objects ${854}$ continuum objects the Ultra Deep Field successfully measured the redshifts of ${282}$ redshift range from $z=0.21$ to $6.64$. The number redshifts with level $\geq 2$ is ${223}$. {width="\textwidth"} the left two panels of Figure \[fig:u10\_musez\_hstpri\_org\], we the distributions of redshifts found with the MUSE data. Because most of the objects at z < 1.5$ are identified by the style="font-variant:small-caps;">ii</span>\]]{}doublet, CONFID mostly by definition the double peaks are resolved and clearly seen. There are six in which CONFID is 2 in this redshift range, whose general lower S/Ns but features are obviously detected narrowband On the other hand, $z 3$ by single [Ly$\alpha$]{}, which gives CONFID 2. For the case of CONFID=3$ at $z > UV absorption features, and in some rare cases style="font-variant:small-caps;">ii</span>, are also discerned. The number of | associated with it in our refeRence field, The MUsE ULtrA DEep FIeld (). in addition to beINg thE deepest spectroscopic fIeld sO fAR obsERvEd witH MUSE, alL Of THE exTrAcTed mUse sPectrA in This fieLd have been VisUaLly inspected. iN tHe field, as dIscUssed in §\[sec:anAlySis\], we uSeD twO DiffeRenT procEdures TO extraCt the sourCeS. their rEDshifts WERe DeteRmined independentLY fIRst, and then recoNciled. heRE wE FIrsT prEsent the baSiC propERties of THe OBJEctS Associated witH their redshIFt mEasureMeNts FOr each ExtraCtIOn mEthod indepeNdenTly. Based oN our unDErstandINg of the DifferEncEs aNd thE ReLaTioNsHIp bETwEen THe tWo source ExTrActioN metHODS And tHe rEdshIft prOperties, we triEd tO finD The Best cOmbinAtioN oF thesE two meThods To Efficiently maxiMize The detectIon RaTe oF rEdshiFTs in thE evEn lArger saMple sizE Of tHe muse DEep Field (the ).
### RedshifTs OF ThE continuUm seleCTeD oBJects {#sec:ReSulT_conT}
aMong ${854}$ cOntiNUuM selecteD objecTS iN tHe MUSE ULtRa Deep fiEld (), We sUccesSFullY measuRed the reDshifTS of ${282}$ objects in thE Redshift range FRoM $Z=0.21$ To $6.64$. tHe nuMbeR of redshiftS witH ConfIdenCE lEveL $\Geq 2$ is ${223}$.
{width="\textwidTh"}
In the lefT TWO panels oF FigURe \[FIg:u10\_musez\_hstpri\_Org\], we Show the disTRibutionS of reDshifts fOund with tHE mUSE data. becAusE moSt oF THe Objects at $0 < z < 1.5$ are USUallY iDentifiEd bY the [\[O<spAn sTylE="foNt-vArIant:small-Caps;">ii</spAn>\]]{}DoUbLeT, thEir COnfID is mosTlY 3 by DeFinItion WHen the DoublE peaKs ArE ResOlved anD ClEARly sEeN. THere Are SiX caseS in wHIch cONFID iS 2 in this reDshIFt raNgE, wHose speCtra in general HaVe lower S/Ns BuT thE featuRES are obviOusly detected in narrowbaND images.
on tHe othEr haNd, galaxieS at $Z > 3$ are ofTen FOund by A singlE featUrE [Ly$\ALPha$]{}, whICH gIveS CoNFID of 2. For THE caSe of $\rM CoNFId=3$ at $z > 3$, the [c<span style="font-variANt:sMall-caps;">iii</spAn>\]]{}eMissION oR UV ABsORptIoN FeaTURes, and in some rarE cases He<spAn STyLe="font-variANt:sMaLl-caps;">iI</span>, arE also DIscerneD. The numbeR of determInEd | associated with it in our reference fiel d,the M USEUltr a Deep Field ( ) . In addition to being the deep es t spe c tr oscop ic fiel d s o far o bs erv ed wi th MU SE, all of the extra cte dMUSE spectra in this fiel d h ave been vis ual ly ins pe cte d . Inthe fiel d, asd iscuss ed in §\[ se c :analy s is\], w e us ed t wo different proc e du r es to extractthe so ur c es . The irredshiftswe re de t ermined in d e p end e ntly first, a nd then rec o nci led. H er e w e first pres en t th e basic pro pert ies of th e obje c ts asso c iated w ith th eir re dshi f tme asu re m ent s f ore ach extract io nmetho d in d e p e nden tly . Ba sed o n our underst and ingo f t he di ffere nces a nd th e rela tions hi p between the t wo s ource ext rac ti onme thods and th e r eds hift pr opertie s , w et r i ed to find the bestco m b in ation of these tw om ethods t oeff icie n t ly ma ximi z ethe dete ctionr at eof reds hi fts in t heeve n lar g er s amplesize ofthe M U SE Deep Field( the ).
### R e ds h i ft s ofthe continuumsele c tedobje c ts {# s ec:re sult_ co n t}
Among ${854}$ cont in uum se lecte d objects inthe MUSE U l t r a Deep F ield () , we successful ly me asured the redshift s of${282}$objects i n the reds hif t r ang e f r o m$z=0.21$ to $ 6 . 64$. T he numb erof reds hif tswit h c on fidence l evel $\g eq 2 $is ${ 223}$ .
![imag e] (fi gu res /\UDF v er_mus e-z_p rop_ HS Tp r i_O RIG_fad e .p d f ){wi dt h= "\te xtw id th"}
Int heleft tw o panelsofF igur e\[ fig:u10 \_musez\_hstp ri \_org\], w esho w thed i stributi ons of redshifts foundw ith the MU SE da ta.Because m ost of th e o b jectsat $0< z < 1 .5$ a re us u a ll y i de ntified by t he[\[O< sp an s tyle="f ont-variant:small- c aps ;">ii</span>\ ]]{ }dou b l et , t h ei r CO NF I D i s mostly 3 by def inition wh en th e double p e aks a re reso lved an d cle a rly see n. Thereare six c as es i n whi ch CONFIDis 2 inthis reds h ift r a ng e, wh ose spect ra in gene ral ha v e l owerS/Ns b ut the f eatur es are obv iously detected in narr owband imag es.
On theoth e r h and, gala xies at $z > 3 $ a reoften fo u nd by a s i ng lef eatur e [L y $\alpha$] { }, wh i c hgives CONFI D o f 2 . For th e caseof $ \rm CONFID=3$ at$ z > 3$, the [C <spa n sty le= " font -v ariant:small-c aps ;" > i ii</span >\ ]]{}emissio n or UVab s orpti on fea tures, and in s om e rarecase s H e<span st yle =" f ont-var ia nt : small- caps ;" >ii</s pan>,a re a l s o discerned. The numb e r of d e ter mined | associated_with it_in our reference field,_the MUSE_Ultra_Deep Field_()._In addition to_being the deepest_spectroscopic field so far_observed with MUSE,_all_of the extracted MUSE spectra in this field have been visually inspected. In the_field,_as discussed_in_§\[sec:analysis\],_we used two different procedures_to extract the sources. Their_redshifts were_determined independently first, and then reconciled. Here we_first_present the basic_properties of the objects associated with their redshift measurements_for each extraction method independently. Based_on our understanding_of_the_differences and the relationship_between the two source extraction methods_and the redshift properties, we tried_to find the best combination of these_two methods to efficiently maximize the_detection rate of redshifts in_the even_larger sample size of the_MUSE Deep Field_(the ).
###_Redshifts of the_continuum selected objects {#sec:result_cont}
Among ${854}$ continuum_selected objects in_the MUSE Ultra Deep Field (),_we_successfully measured the_redshifts_of_${282}$ objects_in the redshift_range_from $z=0.21$_to_$6.64$. The number of redshifts with_confidence_level $\geq 2$ is ${223}$.
{width="\textwidth"}
In the left_two panels of Figure \[fig:u10\_musez\_hstpri\_org\],_we_show the distributions of_redshifts found with the MUSE_data. Because most of the objects_at $0_< z_< 1.5$ are usually identified by the [\[O<span style="font-variant:small-caps;">ii</span>\]]{}doublet, their CONFID_is mostly 3 by definition when_the double peaks are_resolved and_clearly_seen. There are_six_cases in_which CONFID is 2 in this redshift_range, whose_spectra in general have lower S/Ns_but the features are_obviously_detected in narrowband images.
On the other_hand, galaxies at $z > 3$_are often found by a_single_feature_[Ly$\alpha$]{}, which gives CONFID of_2. For the case of $\rm_CONFID=3$ at $z_> 3$, the [C<span style="font-variant:small-caps;">iii</span>\]]{}emission or UV_absorption_features, and in some rare cases_He<span_style="font-variant:small-caps;">ii</span>, are also discerned. The number_of_determined |
Subsets and Splits