text
null | inputs
dict | prediction
null | prediction_agent
null | annotation
list | annotation_agent
null | multi_label
bool 1
class | explanation
null | id
stringlengths 1
5
| metadata
null | status
stringclasses 2
values | event_timestamp
null | metrics
null |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
null |
{
"abstract": " In this paper, by using the idea of linearizing maximal op-erators originated\nby Charles Fefferman and the TT* method of Stein-Wainger, we establish a\nweighted inequality for vector valued maximal Carleson type operators with\nsingular kernels proposed by Andersen and John on the weighted Lorentz spaces\nwith vector-valued functions.\n",
"title": "Vector valued maximal Carleson type operators on the weighted Lorentz spaces"
}
| null | null |
[
"Mathematics"
] | null | true | null |
17501
| null |
Validated
| null | null |
null |
{
"abstract": " We look at interval exchange transformations defined as first return maps on\nthe set of diagonals of a flow of direction $\\theta$ on a square-tiled surface:\nusing a combinatorial approach, we show that, when the surface has at least one\ntrue singularity both the flow and the interval exchange are rigid if and only\nif tan $\\theta$ has bounded partial quotients. Moreover, if all vertices of the\nsquares are singularities of the flat metric, and tan $\\theta$ has bounded\npartial quotients, the square-tiled interval exchange transformation T is not\nof rank one. Finally, for another class of surfaces, those defined by the\nunfolding of billiards in Veech triangles, we build an uncountable set of rigid\ndirectional flows and an uncountable set of rigid interval exchange\ntransformations.\n",
"title": "Rigidity of square-tiled interval exchange transformations"
}
| null | null | null | null | true | null |
17502
| null |
Default
| null | null |
null |
{
"abstract": " Consider the parabolic equation with measure data \\begin{equation*} \\left\\{\n\\begin{aligned} &u_t-{\\rm div} \\mathbf{a}(D u,x,t)=\\mu&\\text{in}& \\quad\n\\Omega_T, &u=0 \\quad &\\text{on}& \\quad \\partial_p\\Omega_T, \\end{aligned}\\right.\n\\end{equation*} where $\\Omega$ is a bounded domain in $\\mathbb{R}^n$,\n$\\Omega_T=\\Omega\\times (0,T)$, $\\partial_p\\Omega_T=(\\partial\\Omega\\times\n(0,T))\\cup (\\Omega\\times\\{0\\})$, and $\\mu$ is a signed Borel measure with\nfinite total mass. Assume that the nonlinearity ${\\bf a}$ satisfies a small\nBMO-seminorm condition, and $\\Omega$ is a Reifenberg flat domain. This paper\nproves a global Marcinkiewicz estimate for the SOLA (Solution Obtained as\nLimits of Approximation) to the parabolic equation.\n",
"title": "Global Marcinkiewicz estimates for nonlinear parabolic equations with nonsmooth coefficients"
}
| null | null |
[
"Mathematics"
] | null | true | null |
17503
| null |
Validated
| null | null |
null |
{
"abstract": " As the number of contributors to online peer-production systems grows, it\nbecomes increasingly important to predict whether the edits that users make\nwill eventually be beneficial to the project. Existing solutions either rely on\na user reputation system or consist of a highly specialized predictor that is\ntailored to a specific peer-production system. In this work, we explore a\ndifferent point in the solution space that goes beyond user reputation but does\nnot involve any content-based feature of the edits. We view each edit as a game\nbetween the editor and the component of the project. We posit that the\nprobability that an edit is accepted is a function of the editor's skill, of\nthe difficulty of editing the component and of a user-component interaction\nterm. Our model is broadly applicable, as it only requires observing data about\nwho makes an edit, what the edit affects and whether the edit survives or not.\nWe apply our model on Wikipedia and the Linux kernel, two examples of\nlarge-scale peer-production systems, and we seek to understand whether it can\neffectively predict edit survival: in both cases, we provide a positive answer.\nOur approach significantly outperforms those based solely on user reputation\nand bridges the gap with specialized predictors that use content-based\nfeatures. It is simple to implement, computationally inexpensive, and in\naddition it enables us to discover interesting structure in the data.\n",
"title": "Can Who-Edits-What Predict Edit Survival?"
}
| null | null | null | null | true | null |
17504
| null |
Default
| null | null |
null |
{
"abstract": " This brief note highlights some basic concepts required toward understanding\nthe evolution of machine learning and deep learning models. The note starts\nwith an overview of artificial intelligence and its relationship to biological\nneuron that ultimately led to the evolution of todays intelligent models.\n",
"title": "Introduction to intelligent computing unit 1"
}
| null | null | null | null | true | null |
17505
| null |
Default
| null | null |
null |
{
"abstract": " We present novel experimental results on pattern formation of signaling\nDictyostelium discoideum amoeba in the presence of a periodic array of\nmillimeter-sized pillars. We observe concentric cAMP waves that initiate almost\nsynchronously at the pillars and propagate outwards. These waves have higher\nfrequency than the other firing centers and dominate the system dynamics. The\ncells respond chemotactically to these circular waves and stream towards the\npillars, forming periodic Voronoi domains that reflect the periodicity of the\nunderlying lattice. We performed comprehensive numerical simulations of a\nreaction-diffusion model to study the characteristics of the boundary\nconditions given by the obstacles. Our simulations show that, the obstacles can\nact as the wave source depending on the imposed boundary condition.\nInterestingly, a critical minimum accumulation of cAMP around the obstacles is\nneeded for the pillars to act as the wave source. This critical value is lower\nat smaller production rates of the intracellular cAMP which can be controlled\nin our experiments using caffeine. Experiments and simulations also show that\nin the presence of caffeine the number of firing centers is reduced which is\ncrucial in our system for circular waves emitted from the pillars to\nsuccessfully take over the dynamics. These results are crucial to understand\nthe signaling mechanism of Dictyostelium cells that experience spatial\nheterogeneities in its natural habitat.\n",
"title": "Spatial heterogeneities shape collective behavior of signaling amoeboid cells"
}
| null | null | null | null | true | null |
17506
| null |
Default
| null | null |
null |
{
"abstract": " The purpose of the paper is to study Yamabe solitons on three-dimensional\npara-Sasakian, paracosymplectic and para-Kenmotsu manifolds. Mainly, we proved\nthat *If the semi-Riemannian metric of a three-dimensional para-Sasakian\nmanifold is a Yamabe soliton, then it is of constant scalar curvature, and the\nflow vector field V is Killing. In the next step, we proved that either\nmanifold has constant curvature -1 and reduces to an Einstein manifold, or V is\nan infinitesimal automorphism of the paracontact metric structure on the\nmanifold. *If the semi-Riemannian metric of a three-dimensional\nparacosymplectic manifold is a Yamabe soliton, then it has constant scalar\ncurvature. Furthermore either manifold is $\\eta$-Einstein, or Ricci flat. *If\nthe semi-Riemannian metric on a three-dimensional para-Kenmotsu manifold is a\nYamabe soliton, then the manifold is of constant sectional curvature -1,\nreduces to an Einstein manifold. Furthermore, Yamabe soliton is expanding with\n$\\lambda$=-6 and the vector field V is Killing. Finally, we construct examples\nto illustrate the results obtained in previous sections.\n",
"title": "Yamabe Solitons on three-dimensional normal almost paracontact metric manifolds"
}
| null | null | null | null | true | null |
17507
| null |
Default
| null | null |
null |
{
"abstract": " We investigate exceedances of the process over a sufficiently high threshold.\nThe exceedances determine the risk of hazardous events like climate\ncatastrophes, huge insurance claims, the loss and delay in telecommunication\nnetworks.\nDue to dependence such exceedances tend to occur in clusters. The cluster\nstructure of social networks is caused by dependence (social relationships and\ninterests) between nodes and possibly heavy-tailed distributions of the node\ndegrees. A minimal time to reach a large node determines the first hitting\ntime. We derive an asymptotically equivalent distribution and a limit\nexpectation of the first hitting time to exceed the threshold $u_n$ as the\nsample size $n$ tends to infinity. The results can be extended to the second\nand, generally, to the $k$th ($k> 2$) hitting times. Applications in\nlarge-scale networks such as social, telecommunication and recommender systems\nare discussed.\n",
"title": "Clustering and Hitting Times of Threshold Exceedances and Applications"
}
| null | null | null | null | true | null |
17508
| null |
Default
| null | null |
null |
{
"abstract": " A problem of classification of local field potentials (LFPs), recorded from\nthe prefrontal cortex of a macaque monkey, is considered. An adult macaque\nmonkey is trained to perform a memory-based saccade. The objective is to decode\nthe eye movement goals from the LFP collected during a memory period. The LFP\nclassification problem is modeled as that of classification of smooth functions\nembedded in Gaussian noise. It is then argued that using minimax function\nestimators as features would lead to consistent LFP classifiers. The theory of\nGaussian sequence models allows us to represent minimax estimators as finite\ndimensional objects. The LFP classifier resulting from this mathematical\nendeavor is a spectrum based technique, where Fourier series coefficients of\nthe LFP data, followed by appropriate shrinkage and thresholding, are used as\nfeatures in a linear discriminant classifier. The classifier is then applied to\nthe LFP data to achieve high decoding accuracy. The function classification\napproach taken in the paper also provides a systematic justification for using\nFourier series, with shrinkage and thresholding, as features for the problem,\nas opposed to using the power spectrum. It also suggests that phase information\nis crucial to the decision making.\n",
"title": "Classification of Local Field Potentials using Gaussian Sequence Model"
}
| null | null |
[
"Statistics"
] | null | true | null |
17509
| null |
Validated
| null | null |
null |
{
"abstract": " We give a few explicit examples which answer an open minded question of\nProfessor Igor Dolgachev on complex dynamics of the inertia group of a smooth\nrational curve on a projective K3 surface and its variants for a rational\nsurface and a non-projective K3 surface.\n",
"title": "A few explicit examples of complex dynamics of inertia groups on surfaces - a question of Professor Igor Dolgachev"
}
| null | null |
[
"Mathematics"
] | null | true | null |
17510
| null |
Validated
| null | null |
null |
{
"abstract": " We consider inference about the history of a sample of DNA sequences,\nconditional upon the haplotype counts and the number of segregating sites\nobserved at the present time. After deriving some theoretical results in the\ncoalescent setting, we implement rejection sampling and importance sampling\nschemes to perform the inference. The importance sampling scheme addresses an\nextension of the Ewens Sampling Formula for a configuration of haplotypes and\nthe number of segregating sites in the sample. The implementations include both\nconstant and variable population size models. The methods are illustrated by\ntwo human Y chromosome data sets.\n",
"title": "Ancestral inference from haplotypes and mutations"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
17511
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, ellipsoid method for linear programming is derived using only\nminimal knowledge of algebra and matrices. Unfortunately, most authors first\ndescribe the algorithm, then later prove its correctness, which requires a good\nknowledge of linear algebra.\n",
"title": "Ellipsoid Method for Linear Programming made simple"
}
| null | null | null | null | true | null |
17512
| null |
Default
| null | null |
null |
{
"abstract": " We experimentally demonstrate a ring geometry all-fiber cavity system for\ncavity quantum electrodynamics with an ensemble of cold atoms. The fiber cavity\ncontains a nanofiber section which mediates atom-light interactions through an\nevanescent field. We observe well-resolved, vacuum Rabi splitting of the cavity\ntransmission spectrum in the weak driving limit due to a collective enhancement\nof the coupling rate by the ensemble of atoms within the evanescent field, and\nwe present a simple theoretical model to describe this. In addition, we\ndemonstrate a method to control and stabilize the resonant frequency of the\ncavity by utilizing the thermal properties of the nanofiber.\n",
"title": "Collective strong coupling of cold atoms to an all-fiber ring cavity"
}
| null | null | null | null | true | null |
17513
| null |
Default
| null | null |
null |
{
"abstract": " We explore the possibility of discovering extreme voting patterns in the U.S.\nCongressional voting records by drawing ideas from the mixture of contaminated\nnormal distributions. A mixture of latent trait models via contaminated normal\ndistributions is proposed. We assume that the low dimensional continuous latent\nvariable comes from a contaminated normal distribution and, therefore, picks up\nextreme patterns in the observed binary data while clustering. We consider in\nparticular such model for the analysis of voting records. The model is applied\nto a U.S. Congressional Voting data set on 16 issues. Note this approach is the\nfirst instance within the literature of a mixture model handling binary data\nwith possible extreme patterns.\n",
"title": "Robust Model-Based Clustering of Voting Records"
}
| null | null | null | null | true | null |
17514
| null |
Default
| null | null |
null |
{
"abstract": " Recently, integrability conditions (ICs) in mutistate Landau-Zener (MLZ)\ntheory were proposed [1]. They describe common properties of all known solved\nsystems with linearly time-dependent Hamiltonians. Here we show that ICs enable\nefficient computer assisted search for new solvable MLZ models that span\ncomplexity range from several interacting states to mesoscopic systems with\nmany-body dynamics and combinatorially large phase space. This diversity\nsuggests that nontrivial solvable MLZ models are numerous. In addition, we\nrefine the formulation of ICs and extend the class of solvable systems to\nmodels with points of multiple diabatic level crossing.\n",
"title": "The Quest for Solvable Multistate Landau-Zener Models"
}
| null | null | null | null | true | null |
17515
| null |
Default
| null | null |
null |
{
"abstract": " We consider the estimation of the multi-period optimal portfolio obtained by\nmaximizing an exponential utility. Employing Jeffreys' non-informative prior\nand the conjugate informative prior, we derive stochastic representations for\nthe optimal portfolio weights at each time point of portfolio reallocation.\nThis provides a direct access not only to the posterior distribution of the\nportfolio weights but also to their point estimates together with uncertainties\nand their asymptotic distributions. Furthermore, we present the posterior\npredictive distribution for the investor's wealth at each time point of the\ninvestment period in terms of a stochastic representation for the future wealth\nrealization. This in turn makes it possible to use quantile-based risk measures\nor to calculate the probability of default. We apply the suggested Bayesian\napproach to assess the uncertainty in the multi-period optimal portfolio by\nconsidering assets from the FTSE 100 in the weeks after the British referendum\nto leave the European Union. The behaviour of the novel portfolio estimation\nmethod in a precarious market situation is illustrated by calculating the\npredictive wealth, the risk associated with the holding portfolio, and the\ndefault probability in each period.\n",
"title": "Bayesian Inference of the Multi-Period Optimal Portfolio for an Exponential Utility"
}
| null | null | null | null | true | null |
17516
| null |
Default
| null | null |
null |
{
"abstract": " Tests of gravity at the galaxy scale are in their infancy. As a first step to\nsystematically uncovering the gravitational significance of galaxies, we map\nthree fundamental gravitational variables -- the Newtonian potential,\nacceleration and curvature -- over the galaxy environments of the local\nuniverse to a distance of approximately 200 Mpc. Our method combines the\ncontributions from galaxies in an all-sky redshift survey, halos from an N-body\nsimulation hosting low-luminosity objects, and linear and quasi-linear modes of\nthe density field. We use the ranges of these variables to determine the extent\nto which galaxies expand the scope of generic tests of gravity and are capable\nof constraining specific classes of model for which they have special\nsignificance. Finally, we investigate the improvements afforded by upcoming\ngalaxy surveys.\n",
"title": "Reconstructing the gravitational field of the local universe"
}
| null | null | null | null | true | null |
17517
| null |
Default
| null | null |
null |
{
"abstract": " By mapping the most advanced elements of the contemporary social\ninteractions, the world scientific collaboration network develops an extremely\ninvolved and heterogeneous organization. Selected characteristics of this\nheterogeneity are studied here and identified by focusing on the scientific\ncollaboration community of H. Eugene Stanley - one of the most prolific world\nscholars at the present time. Based on the Web of Science records as of March\n28, 2016, several variants of networks are constructed. It is found that the\nStanley #1 network - this in analogy to the Erdős # - develops a largely\nconsistent hierarchical organization and Stanley himself obeys rules of the\nsame hierarchy. However, this is seen exclusively in the weighted network\nrepresentation. When such a weighted network is evolving, an existing relevant\nmodel indicates that the spread of weight gets stimulation to the\nmultiplicative bursts over the neighbouring nodes, which leads to a balanced\ngrowth of interconnections among them. While not exclusive to Stanley, such a\nbehaviour is not a rule, however. Networks of other outstanding scholars\nstudied here more often develop a star-like form and the central hubs\nconstitute the outliers. This study is complemented by a spectral analysis of\nthe normalised Laplacian matrices derived from the weighted variants of the\ncorresponding networks and, among others, it points to the efficiency of such a\nprocedure for identifying the component communities and relations among them in\nthe complex weighted networks.\n",
"title": "Hierarchical organization of H. Eugene Stanley scientific collaboration community in weighted network representation"
}
| null | null | null | null | true | null |
17518
| null |
Default
| null | null |
null |
{
"abstract": " We prove that the generating functions for the colored HOMFLY-PT polynomials\nof rational links are specializations of the generating functions of the\nmotivic Donaldson-Thomas invariants of appropriate quivers that we naturally\nassociate with these links. This shows that the conjectural links-quivers\ncorrespondence of Kucharski-Reineke-Stošić-Su{\\l}kowski as well as the\nLMOV conjecture hold for rational links. Along the way, we extend the\nlinks-quivers correspondence to tangles and, thus, explore elements of a skein\ntheory for motivic Donaldson-Thomas invariants.\n",
"title": "Rational links and DT invariants of quivers"
}
| null | null | null | null | true | null |
17519
| null |
Default
| null | null |
null |
{
"abstract": " We study the N=1 supersymmetric solutions of D=11 supergravity obtained as a\nwarped product of four-dimensional anti-de-Sitter space with a\nseven-dimensional Riemannian manifold M. Using the octonion bundle structure on\nM we reformulate the Killing spinor equations in terms of sections of the\noctonion bundle on M. The solutions then define a single complexified\nG2-structure on M or equivalently two real G2-structures. We then study the\ntorsion of these G2-structures and the relationships between them.\n",
"title": "G2-structures for N=1 supersymmetric AdS4 solutions of M-theory"
}
| null | null | null | null | true | null |
17520
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we show that the recent integration of statistical models with\ndeep recurrent neural networks provides a new way of formulating volatility\n(the degree of variation of time series) models that have been widely used in\ntime series analysis and prediction in finance. The model comprises a pair of\ncomplementary stochastic recurrent neural networks: the generative network\nmodels the joint distribution of the stochastic volatility process; the\ninference network approximates the conditional distribution of the latent\nvariables given the observables. Our focus here is on the formulation of\ntemporal dynamics of volatility over time under a stochastic recurrent neural\nnetwork framework. Experiments on real-world stock price datasets demonstrate\nthat the proposed model generates a better volatility estimation and prediction\nthat outperforms mainstream methods, e.g., deterministic models such as GARCH\nand its variants, and stochastic models namely the MCMC-based model\n\\emph{stochvol} as well as the Gaussian process volatility model \\emph{GPVol},\non average negative log-likelihood.\n",
"title": "A Neural Stochastic Volatility Model"
}
| null | null | null | null | true | null |
17521
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a segmentation-free, automatic and efficient procedure to\ndetect general geometric quadric forms in point clouds, where clutter and\nocclusions are inevitable. Our everyday world is dominated by man-made objects\nwhich are designed using 3D primitives (such as planes, cones, spheres,\ncylinders, etc.). These objects are also omnipresent in industrial\nenvironments. This gives rise to the possibility of abstracting 3D scenes\nthrough primitives, thereby positions these geometric forms as an integral part\nof perception and high level 3D scene understanding.\nAs opposed to state-of-the-art, where a tailored algorithm treats each\nprimitive type separately, we propose to encapsulate all types in a single\nrobust detection procedure. At the center of our approach lies a closed form 3D\nquadric fit, operating in both primal & dual spaces and requiring as low as 4\noriented-points. Around this fit, we design a novel, local null-space voting\nstrategy to reduce the 4-point case to 3. Voting is coupled with the famous\nRANSAC and makes our algorithm orders of magnitude faster than its conventional\ncounterparts. This is the first method capable of performing a generic\ncross-type multi-object primitive detection in difficult scenes. Results on\nsynthetic and real datasets support the validity of our method.\n",
"title": "A Minimalist Approach to Type-Agnostic Detection of Quadrics in Point Clouds"
}
| null | null | null | null | true | null |
17522
| null |
Default
| null | null |
null |
{
"abstract": " We have determined new relations between $UBV$ colors and mass-to-light\nratios ($M/L$) for dwarf irregular (dIrr) galaxies, as well as for transformed\n$g^\\prime - r^\\prime$. These $M/L$ to color relations (MLCRs) are based on\nstellar mass density profiles determined for 34 LITTLE THINGS dwarfs from\nspectral energy distribution fitting to multi-wavelength surface photometry in\npassbands from the FUV to the NIR. These relations can be used to determine\nstellar masses in dIrr galaxies for situations where other determinations of\nstellar mass are not possible. Our MLCRs are shallower than comparable MLCRs in\nthe literature determined for spiral galaxies. We divided our dwarf data into\nfour metallicity bins and found indications of a steepening of the MLCR with\nincreased oxygen abundance, perhaps due to more line blanketing occurring at\nhigher metallicity.\n",
"title": "Mass-to-Light versus Color Relations for Dwarf Irregular Galaxies"
}
| null | null | null | null | true | null |
17523
| null |
Default
| null | null |
null |
{
"abstract": " We study numerically the Bloch electron wavepacket dynamics in periodic\npotentials to simulate laser-solid interactions. We introduce a new perspective\nin the coordinate space combined with the motion of the Bloch electron\nwavepackets moving at group and phase velocities under the laser fields. This\nmodel interprets the origins of the two contributions (intra- and interband\ntransitions) of the high-order harmonic generation (HHG) by investigating the\nlocal and global behavior of the wavepackets. It also elucidates the underlying\nphysical picture of the HHG intensity enhancement by means of carrier-envelope\nphase (CEP), chirp and inhomogeneous fields. It provides a deep insight into\nthe emission of high-order harmonics from solids. This model is instructive for\nexperimental measurements and provides a new avenue to distinguish mechanisms\nof the HHG from solids in diffrent laser fields.\n",
"title": "Insight into High-order Harmonic Generation from Solids: The Contributions of the Bloch Wave-packets Moving on the Group and Phase Velocities"
}
| null | null | null | null | true | null |
17524
| null |
Default
| null | null |
null |
{
"abstract": " Sequential Monte Carlo has become a standard tool for Bayesian Inference of\ncomplex models. This approach can be computationally demanding, especially when\ninitialized from the prior distribution. On the other hand, deter-ministic\napproximations of the posterior distribution are often available with no\ntheoretical guaranties. We propose a bridge sampling scheme starting from such\na deterministic approximation of the posterior distribution and targeting the\ntrue one. The resulting Shortened Bridge Sampler (SBS) relies on a sequence of\ndistributions that is determined in an adaptive way. We illustrate the\nrobustness and the efficiency of the methodology on a large simulation study.\nWhen applied to network datasets, SBS inference leads to different statistical\nconclusions from the one supplied by the standard variational Bayes\napproximation.\n",
"title": "Using deterministic approximations to accelerate SMC for posterior sampling"
}
| null | null |
[
"Statistics"
] | null | true | null |
17525
| null |
Validated
| null | null |
null |
{
"abstract": " Decades of psychological research have been aimed at modeling how people\nlearn features and categories. The empirical validation of these theories is\noften based on artificial stimuli with simple representations. Recently, deep\nneural networks have reached or surpassed human accuracy on tasks such as\nidentifying objects in natural images. These networks learn representations of\nreal-world stimuli that can potentially be leveraged to capture psychological\nrepresentations. We find that state-of-the-art object classification networks\nprovide surprisingly accurate predictions of human similarity judgments for\nnatural images, but fail to capture some of the structure represented by\npeople. We show that a simple transformation that corrects these discrepancies\ncan be obtained through convex optimization. We use the resulting\nrepresentations to predict the difficulty of learning novel categories of\nnatural images. Our results extend the scope of psychological experiments and\ncomputational modeling by enabling tractable use of large natural stimulus\nsets.\n",
"title": "Evaluating (and improving) the correspondence between deep neural networks and human representations"
}
| null | null | null | null | true | null |
17526
| null |
Default
| null | null |
null |
{
"abstract": " We present a collection of conjectural trace identities and explain why they\nare equivalent to base change and descent of automorphic representations of\n$\\mathrm{GL}_n(\\mathbb{A}_F)$ along nonsolvable extensions (under some\nsimplifying hypotheses). The case $n=2$ is treated in more detail and\napplications towards the Artin conjecture for icosahedral Galois\nrepresentations are given.\n",
"title": "An approach to nonsolvable base change and descent"
}
| null | null | null | null | true | null |
17527
| null |
Default
| null | null |
null |
{
"abstract": " Targeted attacks against network infrastructure are notoriously difficult to\nguard against. In the case of communication networks, such attacks can leave\nusers vulnerable to censorship and surveillance, even when cryptography is\nused. Much of the existing work on network fault-tolerance focuses on random\nfaults and does not apply to adversarial faults (attacks). Centralized networks\nhave single points of failure by definition, leading to a growing popularity in\ndecentralized architectures and protocols for greater fault-tolerance. However,\ncentralized network structure can arise even when protocols are decentralized.\nDespite their decentralized protocols, the Internet and World-Wide Web have\nbeen shown both theoretically and historically to be highly susceptible to\nattack, in part due to emergent structural centralization. When single points\nof failure exist, they are potentially vulnerable to non-technological (i.e.,\ncoercive) attacks, suggesting the importance of a structural approach to\nattack-tolerance. We show how the assumption of partial trust transitivity,\nwhile more realistic than the assumption underlying webs of trust, can be used\nto quantify the effective redundancy of a network as a function of trust\ntransitivity. We also prove that the effective redundancy of the wrap-around\nbutterfly topology increases exponentially with trust transitivity and describe\na novel concurrent multipath routing algorithm for constructing paths to\nutilize that redundancy. When portions of network structure can be dictated our\nresults can be used to create scalable, attack-tolerant infrastructures. More\ngenerally, our results provide a theoretical formalism for evaluating the\neffects of network structure on adversarial fault-tolerance.\n",
"title": "Towards Attack-Tolerant Networks: Concurrent Multipath Routing and the Butterfly Network"
}
| null | null |
[
"Computer Science"
] | null | true | null |
17528
| null |
Validated
| null | null |
null |
{
"abstract": " Reinforcement learning and symbolic planning have both been used to build\nintelligent autonomous agents. Reinforcement learning relies on learning from\ninteractions with real world, which often requires an unfeasibly large amount\nof experience. Symbolic planning relies on manually crafted symbolic knowledge,\nwhich may not be robust to domain uncertainties and changes. In this paper we\npresent a unified framework {\\em PEORL} that integrates symbolic planning with\nhierarchical reinforcement learning (HRL) to cope with decision-making in a\ndynamic environment with uncertainties.\nSymbolic plans are used to guide the agent's task execution and learning, and\nthe learned experience is fed back to symbolic knowledge to improve planning.\nThis method leads to rapid policy search and robust symbolic plans in complex\ndomains. The framework is tested on benchmark domains of HRL.\n",
"title": "PEORL: Integrating Symbolic Planning and Hierarchical Reinforcement Learning for Robust Decision-Making"
}
| null | null | null | null | true | null |
17529
| null |
Default
| null | null |
null |
{
"abstract": " When pristine material surfaces are exposed to air, highly reactive broken\nbonds can promote the formation of surface oxides with structures and\nproperties differing greatly from bulk. Determination of the oxide structure,\nhowever, is often elusive through the use of indirect diffraction methods or\ntechniques that probe only the outer most layer. As a result, surface oxides\nforming on widely used materials, such as group III-nitrides, have not been\nunambiguously resolved, even though critical properties can depend sensitively\non their presence. In this work, aberration corrected scanning transmission\nelectron microscopy reveals directly, and with depth dependence, the structure\nof native two--dimensional oxides that form on AlN and GaN surfaces. Through\natomic resolution imaging and spectroscopy, we show that the oxide layers are\ncomprised of tetrahedra--octahedra cation--oxygen units, similar to bulk\n$\\theta$--Al$_2$O$_3$ and $\\beta$--Ga$_2$O$_3$. By applying density functional\ntheory, we show that the observed structures are more stable than previously\nproposed surface oxide models. We place the impact of these observations in the\ncontext of key III-nitride growth, device issues, and the recent discovery of\ntwo-dimnesional nitrides.\n",
"title": "Structure of Native Two-dimensional Oxides on III--Nitride Surfaces"
}
| null | null | null | null | true | null |
17530
| null |
Default
| null | null |
null |
{
"abstract": " We develop the theory of hydrodynamic charge and heat transport in strongly\ninteracting quasi-relativistic systems on manifolds with inhomogeneous spatial\ncurvature. In solid-state physics, this is analogous to strain disorder in the\nunderlying lattice. In the hydrodynamic limit, we find that the thermal and\nelectrical conductivities are dominated by viscous effects, and that the\nthermal conductivity is most sensitive to this disorder. We compare the effects\nof inhomogeneity in the spatial metric to inhomogeneity in the chemical\npotential, and discuss the extent to which our hydrodynamic theory is relevant\nfor experimentally realizable condensed matter systems, including suspended\ngraphene at the Dirac point.\n",
"title": "Hydrodynamic charge and heat transport on inhomogeneous curved spaces"
}
| null | null | null | null | true | null |
17531
| null |
Default
| null | null |
null |
{
"abstract": " Predicting how a proposed cancer treatment will affect a given tumor can be\ncast as a machine learning problem, but the complexity of biological systems,\nthe number of potentially relevant genomic and clinical features, and the lack\nof very large scale patient data repositories make this a unique challenge.\n\"Pure data\" approaches to this problem are underpowered to detect\ncombinatorially complex interactions and are bound to uncover false\ncorrelations despite statistical precautions taken (1). To investigate this\nsetting, we propose a method to integrate simulations, a strong form of prior\nknowledge, into machine learning, a combination which to date has been largely\nunexplored. The results of multiple simulations (under various uncertainty\nscenarios) are used to compute similarity measures between every pair of\nsamples: sample pairs are given a high similarity score if they behave\nsimilarly under a wide range of simulation parameters. These similarity values,\nrather than the original high dimensional feature data, are used to train\nkernelized machine learning algorithms such as support vector machines, thus\nhandling the curse-of-dimensionality that typically affects genomic machine\nlearning. Using four synthetic datasets of complex systems--three biological\nmodels and one network flow optimization model--we demonstrate that when the\nnumber of training samples is small compared to the number of features, the\nsimulation kernel approach dominates over no-prior-knowledge methods. In\naddition to biology and medicine, this approach should be applicable to other\ndisciplines, such as weather forecasting, financial markets, and agricultural\nmanagement, where predictive models are sought and informative yet approximate\nsimulations are available. The Python SimKern software, the models (in MATLAB,\nOctave, and R), and the datasets are made freely available at\nthis https URL .\n",
"title": "Simulation assisted machine learning"
}
| null | null | null | null | true | null |
17532
| null |
Default
| null | null |
null |
{
"abstract": " Rechargeable redox flow batteries with serpentine flow field designs have\nbeen demonstrated to deliver higher current density and power density in medium\nand large-scale stationary energy storage applications. Nevertheless, the\nfundamental mechanisms involved with improved current density in flow batteries\nwith flow field designs have not been understood. Here we report a maximum\ncurrent density concept associated with stoichiometric availability of\nelectrolyte reactant flow penetration through the porous electrode that can be\nachieved in a flow battery system with a \"zero-gap\"serpentine flow field\narchitecture. This concept can explain a higher current density achieved within\nallowing reactions of all species soluble in the electrolyte. Further\nvalidations with experimental data are confirmed by an example of a vanadium\nflow battery with a serpentine flow structure over carbon paper electrode.\n",
"title": "Rechargeable redox flow batteries: Maximum current density with electrolyte flow reactant penetration in a serpentine flow structure"
}
| null | null | null | null | true | null |
17533
| null |
Default
| null | null |
null |
{
"abstract": " We present a list of open questions in mathematical physics. After a\nhistorical introduction, a number of problems in a variety of different fields\nare discussed, with the intention of giving an overall impression of the\ncurrent status of mathematical physics, particularly in the topical fields of\nclassical general relativity, cosmology and the quantum realm. This list is\nmotivated by the recent article proposing 42 fundamental questions (in physics)\nwhich must be answered on the road to full enlightenment. But paraphrasing a\nfamous quote by the British football manager Bill Shankly, in response to the\nquestion of whether mathematics can answer the Ultimate Question of Life, the\nUniverse, and Everything, mathematics is, of course, much more important than\nthat.\n",
"title": "Open problems in mathematical physics"
}
| null | null | null | null | true | null |
17534
| null |
Default
| null | null |
null |
{
"abstract": " Online advertising and product recommendation are important domains of\napplications for multi-armed bandit methods. In these fields, the reward that\nis immediately available is most often only a proxy for the actual outcome of\ninterest, which we refer to as a conversion. For instance, in web advertising,\nclicks can be observed within a few seconds after an ad display but the\ncorresponding sale --if any-- will take hours, if not days to happen. This\npaper proposes and investigates a new stochas-tic multi-armed bandit model in\nthe framework proposed by Chapelle (2014) --based on empirical studies in the\nfield of web advertising-- in which each action may trigger a future reward\nthat will then happen with a stochas-tic delay. We assume that the probability\nof conversion associated with each action is unknown while the distribution of\nthe conversion delay is known, distinguishing between the (idealized) case\nwhere the conversion events may be observed whatever their delay and the more\nrealistic setting in which late conversions are censored. We provide\nperformance lower bounds as well as two simple but efficient algorithms based\non the UCB and KLUCB frameworks. The latter algorithm, which is preferable when\nconversion rates are low, is based on a Poissonization argument, of independent\ninterest in other settings where aggregation of Bernoulli observations with\ndifferent success probabilities is required.\n",
"title": "Stochastic Bandit Models for Delayed Conversions"
}
| null | null | null | null | true | null |
17535
| null |
Default
| null | null |
null |
{
"abstract": " The goal of population spectral synthesis (PSS) is to decipher from the\nspectrum of a galaxy the mass, age and metallicity of its constituent stellar\npopulations. This technique has been established as a fundamental tool in\nextragalactic research. It has been extensively applied to large spectroscopic\ndata sets, notably the SDSS, leading to important insights into the galaxy\nassembly history. However, despite significant improvements over the past\ndecade, all current PSS codes suffer from two major deficiencies that inhibit\nus from gaining sharp insights into the star-formation history (SFH) of\ngalaxies and potentially introduce substantial biases in studies of their\nphysical properties (e.g., stellar mass, mass-weighted stellar age and specific\nstar formation rate). These are i) the neglect of nebular emission in spectral\nfits, consequently, ii) the lack of a mechanism that ensures consistency\nbetween the best-fitting SFH and the observed nebular emission characteristics\nof a star-forming (SF) galaxy. In this article, we present FADO (Fitting\nAnalysis using Differential evolution Optimization): a conceptually novel,\npublicly available PSS tool with the distinctive capability of permitting\nidentification of the SFH that reproduces the observed nebular characteristics\nof a SF galaxy. This so-far unique self-consistency concept allows us to\nsignificantly alleviate degeneracies in current spectral synthesis. The\ninnovative character of FADO is further augmented by its mathematical\nfoundation: FADO is the first PSS code employing genetic differential evolution\noptimization. This, in conjunction with other unique elements in its\nmathematical concept (e.g., optimization of the spectral library using\nartificial intelligence, convergence test, quasi-parallelization) results in\nkey improvements with respect to computational efficiency and uniqueness of the\nbest-fitting SFHs.\n",
"title": "Fitting Analysis using Differential Evolution Optimization (FADO): Spectral population synthesis through genetic optimization under self-consistency boundary conditions"
}
| null | null |
[
"Physics"
] | null | true | null |
17536
| null |
Validated
| null | null |
null |
{
"abstract": " In \\cite{Chen:2016fgi} we proposed a differential operator for the evaluation\nof the multi-dimensional residues on isolated (zero-dimensional) poles.In this\npaper we discuss some new insight on evaluating the (generalized)\nCachazo-He-Yuan (CHY) forms of the scattering amplitudes using this\ndifferential operator. We introduce a tableau representation for the\ncoefficients appearing in the proposed differential operator. Combining the\ntableaux with the polynomial forms of the scattering equations, the evaluation\nof the generalized CHY form becomes a simple combinatoric problem. It is thus\npossible to obtain the coefficients arising in the differential operator in a\nstraightforward way. We present the procedure for a complete solution of the\n$n$-gon amplitudes at one-loop level in a generalized CHY form. We also apply\nour method to fully evaluate the one-loop five-point amplitude in the maximally\nsupersymmetric Yang-Mills theory; the final result is identical to the one\nobtained by Q-Cut.\n",
"title": "A Combinatoric Shortcut to Evaluate CHY-forms"
}
| null | null | null | null | true | null |
17537
| null |
Default
| null | null |
null |
{
"abstract": " In this paper a new restarting method for Krylov subspace matrix exponential\nevaluations is proposed. Since our restarting technique essentially employs the\nresidual, some convergence results for the residual are given. We also discuss\nhow the restart length can be adjusted after each restart cycle, which leads to\nan adaptive restarting procedure. Numerical tests are presented to compare our\nrestarting with three other restarting methods. Some of the algorithms\ndescribed in this paper are a part of the Octave/Matlab package expmARPACK\navailable at this http URL.\n",
"title": "ART: adaptive residual--time restarting for Krylov subspace matrix exponential evaluations"
}
| null | null | null | null | true | null |
17538
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, nil extensions of some special type of ordered semigroups,\nsuch as, simple regular ordered semigroups, left simple and right regular\nordered semigroup. Moreover, we have characterized complete semilattice\ndecomposition of all ordered semigroups which are nil extension of ordered\nsemigroup.\n",
"title": "Nil extensions of simple regular ordered semigroup"
}
| null | null | null | null | true | null |
17539
| null |
Default
| null | null |
null |
{
"abstract": " We examine a class of embeddings based on structured random matrices with\northogonal rows which can be applied in many machine learning applications\nincluding dimensionality reduction and kernel approximation. For both the\nJohnson-Lindenstrauss transform and the angular kernel, we show that we can\nselect matrices yielding guaranteed improved performance in accuracy and/or\nspeed compared to earlier methods. We introduce matrices with complex entries\nwhich give significant further accuracy improvement. We provide geometric and\nMarkov chain-based perspectives to help understand the benefits, and empirical\nresults which suggest that the approach is helpful in a wider range of\napplications.\n",
"title": "The Unreasonable Effectiveness of Structured Random Orthogonal Embeddings"
}
| null | null | null | null | true | null |
17540
| null |
Default
| null | null |
null |
{
"abstract": " Probabilistic modeling is fundamental to the statistical analysis of complex\ndata. In addition to forming a coherent description of the data-generating\nprocess, probabilistic models enable parameter inference about given data sets.\nThis procedure is well-developed in the Bayesian perspective, in which one\ninfers probability distributions describing to what extent various possible\nparameters agree with the data. In this paper we motivate and review\nprobabilistic modeling for adaptive immune receptor repertoire data then\ndescribe progress and prospects for future work, from germline haplotyping to\nadaptive immune system deployment across tissues. The relevant quantities in\nimmune sequence analysis include not only continuous parameters such as gene\nuse frequency, but also discrete objects such as B cell clusters and lineages.\nThroughout this review, we unravel the many opportunities for probabilistic\nmodeling in adaptive immune receptor analysis, including settings for which the\nBayesian approach holds substantial promise (especially if one is optimistic\nabout new computational methods). From our perspective the greatest prospects\nfor progress in probabilistic modeling for repertoires concern ancestral\nsequence estimation for B cell receptor lineages, including uncertainty from\ngermline genotype, rearrangement, and lineage development.\n",
"title": "The Bayesian optimist's guide to adaptive immune receptor repertoire analysis"
}
| null | null | null | null | true | null |
17541
| null |
Default
| null | null |
null |
{
"abstract": " There has been considerable research on automated index tuning in database\nmanagement systems (DBMSs). But the majority of these solutions tune the index\nconfiguration by retrospectively making computationally expensive physical\ndesign changes all at once. Such changes degrade the DBMS's performance during\nthe process, and have reduced utility during subsequent query processing due to\nthe delay between a workload shift and the associated change. A better approach\nis to generate small changes that tune the physical design over time, forecast\nthe utility of these changes, and apply them ahead of time to maximize their\nimpact.\nThis paper presents predictive indexing that continuously improves a\ndatabase's physical design using lightweight physical design changes. It uses a\nmachine learning model to forecast the utility of these changes, and\ncontinuously refines the index configuration of the database to handle evolving\nworkloads. We introduce a lightweight hybrid scan operator with which a DBMS\ncan make use of partially-built indexes for query processing. Our evaluation\nshows that predictive indexing improves the throughput of a DBMS by 3.5--5.2x\ncompared to other state-of-the-art indexing approaches. We demonstrate that\npredictive indexing works seamlessly with other lightweight automated physical\ndesign tuning methods.\n",
"title": "Predictive Indexing"
}
| null | null | null | null | true | null |
17542
| null |
Default
| null | null |
null |
{
"abstract": " Despite remarkable successes, Deep Reinforcement Learning (DRL) is not robust\nto hyperparameterization, implementation details, or small environment changes\n(Henderson et al. 2017, Zhang et al. 2018). Overcoming such sensitivity is key\nto making DRL applicable to real world problems. In this paper, we identify\nsensitivity to time discretization in near continuous-time environments as a\ncritical factor; this covers, e.g., changing the number of frames per second,\nor the action frequency of the controller. Empirically, we find that\nQ-learning-based approaches such as Deep Q- learning (Mnih et al., 2015) and\nDeep Deterministic Policy Gradient (Lillicrap et al., 2015) collapse with small\ntime steps. Formally, we prove that Q-learning does not exist in continuous\ntime. We detail a principled way to build an off-policy RL algorithm that\nyields similar performances over a wide range of time discretizations, and\nconfirm this robustness empirically.\n",
"title": "Making Deep Q-learning methods robust to time discretization"
}
| null | null | null | null | true | null |
17543
| null |
Default
| null | null |
null |
{
"abstract": " We demonstrate that in diffusive superconductor/ferromagnet/superconductor\n(S/F/S) junctions a finite, {\\it anomalous}, Josephson current can flow even at\nzero phase difference between the S electrodes. The conditions for the\nobservation of this effect are non-coplanar magnetization distribution and a\nbroken magnetization inversion symmetry of the superconducting current. The\nlatter symmetry is intrinsic for the widely used quasiclassical approximation\nand prevent previous works, based on this approximation, from obtaining the\nJosephson anomalous current. We show that this symmetry can be removed by\nintroducing spin-dependent boundary conditions for the quasiclassical equations\nat the superconducting/ferromagnet interfaces in diffusive systems. Using this\nrecipe we considered generic multilayer magnetic systems and determine the\nideal experimental conditions in order to maximize the anomalous current.\n",
"title": "Anomalous current in diffusive ferromagnetic Josephson junctions"
}
| null | null | null | null | true | null |
17544
| null |
Default
| null | null |
null |
{
"abstract": " Although a majority of the theoretical literature in high-dimensional\nstatistics has focused on settings which involve fully-observed data, settings\nwith missing values and corruptions are common in practice. We consider the\nproblems of estimation and of constructing component-wise confidence intervals\nin a sparse high-dimensional linear regression model when some covariates of\nthe design matrix are missing completely at random. We analyze a variant of the\nDantzig selector [9] for estimating the regression model and we use a\nde-biasing argument to construct component-wise confidence intervals. Our first\nmain result is to establish upper bounds on the estimation error as a function\nof the model parameters (the sparsity level s, the expected fraction of\nobserved covariates $\\rho_*$, and a measure of the signal strength\n$\\|\\beta^*\\|_2$). We find that even in an idealized setting where the\ncovariates are assumed to be missing completely at random, somewhat\nsurprisingly and in contrast to the fully-observed setting, there is a\ndichotomy in the dependence on model parameters and much faster rates are\nobtained if the covariance matrix of the random design is known. To study this\nissue further, our second main contribution is to provide lower bounds on the\nestimation error showing that this discrepancy in rates is unavoidable in a\nminimax sense. We then consider the problem of high-dimensional inference in\nthe presence of missing data. We construct and analyze confidence intervals\nusing a de-biased estimator. In the presence of missing data, inference is\ncomplicated by the fact that the de-biasing matrix is correlated with the pilot\nestimator and this necessitates the design of a new estimator and a novel\nanalysis. We also complement our mathematical study with extensive simulations\non synthetic and semi-synthetic data that show the accuracy of our asymptotic\npredictions for finite sample sizes.\n",
"title": "Rate Optimal Estimation and Confidence Intervals for High-dimensional Regression with Missing Covariates"
}
| null | null | null | null | true | null |
17545
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we use an iterative algorithm for solving Fredholm equations of\nthe first kind. The basic algorithm is known and is based on an EM algorithm\nwhen involved functions are non-negative and integrable. With this algorithm we\ndemonstrate two examples involving the estimation of a mixing density and a\nfirst passage time density function involving Brownian motion. We also develop\nthe basic algorithm to include functions which are not necessarily non-negative\nand again present illustrations under this scenario. A self contained proof of\nconvergence of all the algorithms employed is presented.\n",
"title": "Applications of an algorithm for solving Fredholm equations of the first kind"
}
| null | null | null | null | true | null |
17546
| null |
Default
| null | null |
null |
{
"abstract": " Kernel quadratures and other kernel-based approximation methods typically\nsuffer from prohibitive cubic time and quadratic space complexity in the number\nof function evaluations. The problem arises because a system of linear\nequations needs to be solved. In this article we show that the weights of a\nkernel quadrature rule can be computed efficiently and exactly for up to tens\nof millions of nodes if the kernel, integration domain, and measure are fully\nsymmetric and the node set is a union of fully symmetric sets. This is based on\nthe observations that in such a setting there are only as many distinct weights\nas there are fully symmetric sets and that these weights can be solved from a\nlinear system of equations constructed out of row sums of certain submatrices\nof the full kernel matrix. We present several numerical examples that show\nfeasibility, both for a large number of nodes and in high dimensions, of the\ndeveloped fully symmetric kernel quadrature rules. Most prominent of the fully\nsymmetric kernel quadrature rules we propose are those that use sparse grids.\n",
"title": "Fully symmetric kernel quadrature"
}
| null | null | null | null | true | null |
17547
| null |
Default
| null | null |
null |
{
"abstract": " In this work we introduce a conditional accelerated lazy stochastic gradient\ndescent algorithm with optimal number of calls to a stochastic first-order\noracle and convergence rate $O\\left(\\frac{1}{\\varepsilon^2}\\right)$ improving\nover the projection-free, Online Frank-Wolfe based stochastic gradient descent\nof Hazan and Kale [2012] with convergence rate\n$O\\left(\\frac{1}{\\varepsilon^4}\\right)$.\n",
"title": "Conditional Accelerated Lazy Stochastic Gradient Descent"
}
| null | null | null | null | true | null |
17548
| null |
Default
| null | null |
null |
{
"abstract": " Generative moment matching network (GMMN) is a deep generative model that\ndiffers from Generative Adversarial Network (GAN) by replacing the\ndiscriminator in GAN with a two-sample test based on kernel maximum mean\ndiscrepancy (MMD). Although some theoretical guarantees of MMD have been\nstudied, the empirical performance of GMMN is still not as competitive as that\nof GAN on challenging and large benchmark datasets. The computational\nefficiency of GMMN is also less desirable in comparison with GAN, partially due\nto its requirement for a rather large batch size during the training. In this\npaper, we propose to improve both the model expressiveness of GMMN and its\ncomputational efficiency by introducing adversarial kernel learning techniques,\nas the replacement of a fixed Gaussian kernel in the original GMMN. The new\napproach combines the key ideas in both GMMN and GAN, hence we name it MMD GAN.\nThe new distance measure in MMD GAN is a meaningful loss that enjoys the\nadvantage of weak topology and can be optimized via gradient descent with\nrelatively small batch sizes. In our evaluation on multiple benchmark datasets,\nincluding MNIST, CIFAR- 10, CelebA and LSUN, the performance of MMD-GAN\nsignificantly outperforms GMMN, and is competitive with other representative\nGAN works.\n",
"title": "MMD GAN: Towards Deeper Understanding of Moment Matching Network"
}
| null | null | null | null | true | null |
17549
| null |
Default
| null | null |
null |
{
"abstract": " We study the multipartite entanglement of a quantum many-body system\nundergoing a quantum quench. We quantify multipartite entanglement through the\nquantum Fisher information (QFI) density and we are able to express it after a\nquench in terms of a generalized response function. For pure state initial\nconditions and in the thermodynamic limit, we can express the QFI as the\nfluctuations of an observable computed in the so-called diagonal ensemble. We\napply the formalism to the dynamics of a quantum Ising chain after a quench in\nthe transverse field. In this model the asymptotic state is, in almost all\ncases, more than two-partite entangled. Moreover, starting from the\nferromagnetic phase, we find a divergence of multipartite entanglement for\nsmall quenches closely connected to a corresponding divergence of the\ncorrelation length.\n",
"title": "Multipartite entanglement after a quantum quench"
}
| null | null | null | null | true | null |
17550
| null |
Default
| null | null |
null |
{
"abstract": " Thermal properties of graphene monolayers are studied by path-integral\nmolecular dynamics (PIMD) simulations, which take into account the quantization\nof vibrational modes in the crystalline membrane, and allow one to consider\nanharmonic effects in these properties. This system was studied at temperatures\nin the range from 12 to 2000~K and zero external stress, by describing the\ninteratomic interactions through the LCBOPII effective potential. We analyze\nthe internal energy and specific heat and compare the results derived from the\nsimulations with those yielded by a harmonic approximation for the vibrational\nmodes. This approximation turns out to be rather precise up to temperatures of\nabout 400~K. At higher temperatures, we observe an influence of the elastic\nenergy, due to the thermal expansion of the graphene sheet. Zero-point and\nthermal effects on the in-plane and \"real\" surface of graphene are discussed.\nThe thermal expansion coefficient $\\alpha$ of the real area is found to be\npositive at all temperatures, in contrast to the expansion coefficient\n$\\alpha_p$ of the in-plane area, which is negative at low temperatures, and\nbecomes positive for $T \\gtrsim$ 1000~K.\n",
"title": "Thermal properties of graphene from path-integral simulations"
}
| null | null | null | null | true | null |
17551
| null |
Default
| null | null |
null |
{
"abstract": " The summary presented in this paper highlights the results obtained in a\nfour-years project aiming at analyzing the development process of software\nartifacts from two points of view: Effectiveness and Affectiveness. The first\nattribute is meant to analyze the productivity of the Open Source Communities\nby measuring the time required to resolve an issue, while the latter provides a\nnovel approach for studying the development process by analyzing the\naffectiveness ex-pressed by developers in their comments posted during the\nissue resolution phase. Affectivenes is obtained by measuring Sentiment,\nPoliteness and Emotions. All the study presented in this summary are based on\nJira, one of the most used software repositories.\n",
"title": "Measuring Affectiveness and Effectiveness in Software Systems"
}
| null | null | null | null | true | null |
17552
| null |
Default
| null | null |
null |
{
"abstract": " A stochastic model of excitatory and inhibitory interactions which bears\nuniversality traits is introduced and studied. The endogenous component of\nnoise, stemming from finite size corrections, drives robust inter-nodes\ncorrelations, that persist at large large distances. Anti-phase synchrony at\nsmall frequencies is resolved on adjacent nodes and found to promote the\nspontaneous generation of long-ranged stochastic patterns, that invade the\nnetwork as a whole. These patterns are lacking under the idealized\ndeterministic scenario, and could provide novel hints on how living systems\nimplement and handle a large gallery of delicate computational tasks.\n",
"title": "Intertangled stochastic motifs in networks of excitatory-inhibitory units"
}
| null | null | null | null | true | null |
17553
| null |
Default
| null | null |
null |
{
"abstract": " We present a new Monte Carlo methodology for the accurate estimation of the\ndistribution of the sum of dependent log-normal random variables. The\nmethodology delivers statistically unbiased estimators for three distributional\nquantities of significant interest in finance and risk management: the left\ntail, or cumulative distribution function, the probability density function,\nand the right tail, or complementary distribution function of the sum of\ndependent log-normal factors. In all of these three cases our methodology\ndelivers fast and highly accurate estimators in settings for which existing\nmethodology delivers estimators with large variance that tend to underestimate\nthe true quantity of interest. We provide insight into the computational\nchallenges using theory and numerical experiments, and explain their much wider\nimplications for Monte Carlo statistical estimators of rare-event\nprobabilities. In particular, we find that theoretically strongly-efficient\nestimators should be used with great caution in practice, because they may\nyield inaccurate results in the pre-limit. Further, this inaccuracy may not be\ndetectable from the output of the Monte Carlo simulation, because the\nsimulation output may severely underestimate the true variance of the\nestimator.\n",
"title": "Accurate Computation of the Distribution of Sums of Dependent Log-Normals with Applications to the Black-Scholes Model"
}
| null | null | null | null | true | null |
17554
| null |
Default
| null | null |
null |
{
"abstract": " We study the unitary representations of the non-compact real forms of the\ncomplex Lie superalgebra sl(n|m). Among them, only the real form su(p,q|m)\n(p+q=n) admits nontrivial unitary representations, and all such representations\nare of the highest-weight type (or the lowest-weight type). We extend the\nstandard oscillator construction of the unitary representations of non-compact\nLie superalgebras over standard Fock spaces to generalised Fock spaces which\nallows us to define the action of oscillator determinants raised to non-integer\npowers. We prove that the proposed construction yields all the unitary\nrepresentations including those with continuous labels. The unitary\nrepresentations can be diagrammatically represented by non-compact Young\ndiagrams. We apply our general results to the physically important case of\nfour-dimensional conformal superalgebra su(2,2|4) and show how it yields\nreadily its unitary representations including those corresponding to\nsupermultiplets of conformal fields with continuous (anomalous) scaling\ndimensions.\n",
"title": "The complete unitary dual of non-compact Lie superalgebra su(p,q|m) via the generalised oscillator formalism, and non-compact Young diagrams"
}
| null | null |
[
"Mathematics"
] | null | true | null |
17555
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper we describe the problem of painter classification, and propose\na novel approach based on deep convolutional autoencoder neural networks. While\nprevious approaches relied on image processing and manual feature extraction\nfrom paintings, our approach operates on the raw pixel level, without any\npreprocessing or manual feature extraction. We first train a deep convolutional\nautoencoder on a dataset of paintings, and subsequently use it to initialize a\nsupervised convolutional neural network for the classification phase.\nThe proposed approach substantially outperforms previous methods, improving\nthe previous state-of-the-art for the 3-painter classification problem from\n90.44% accuracy (previous state-of-the-art) to 96.52% accuracy, i.e., a 63%\nreduction in error rate.\n",
"title": "DeepPainter: Painter Classification Using Deep Convolutional Autoencoders"
}
| null | null | null | null | true | null |
17556
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we will establish an elliptic local Li-Yau gradient estimate\nfor weak solutions of the heat equation on metric measure spaces with\ngeneralized Ricci curvature bounded from below. One of its main applications is\na sharp gradient estimate for the logarithm of heat kernels. These results seem\nnew even for smooth Riemannian manifolds.\n",
"title": "Sharp gradient estimate for heat kernels on $RCD^*(K,N)$ metric measure spaces"
}
| null | null | null | null | true | null |
17557
| null |
Default
| null | null |
null |
{
"abstract": " Due to one of the most representative contributions to the energy in diatomic\nmolecules being the vibrational, we consider the generalized Morse potential\n(GMP) as one of the typical potential of interaction for one-dimensional\nmicroscopic systems, which describes local anharmonic effects. From Eckart\npotential (EP) model, it is possible to find a connection with the GMP model,\nas well as obtain the analytical expression for the energy spectrum because it\nis based on $S\\,O\\left(2,1\\right)$ algebras. In this work we find the\nmacroscopic properties such as vibrational mean energy $U$, specific heat $C$,\nHelmholtz free energy $F$ and entropy $S$ for a heteronuclear diatomic system,\nalong with the exact partition function and its approximation for the high\ntemperature region. Finally, we make a comparison between the graphs of some\nthermodynamic functions obtained with the GMP and the Morse potential (MP) for\n$H\\,Cl$ molecules.\n",
"title": "Thermodynamic properties of diatomic molecules systems under anharmonic Eckart potential"
}
| null | null |
[
"Physics"
] | null | true | null |
17558
| null |
Validated
| null | null |
null |
{
"abstract": " Environmental pollutants, such as colors from the textile industry, affect\nwater quality indicators like color, smell, and taste. These substances in the\nwater cause the obstruction of filters and membranes and thereby reduce the\nefficiency of advanced water treatment processes. In addition, they are harmful\nto human health because of reaction with disinfectants and production of\nby-products. Iron oxide nanoparticles are considered effective absorbents for\nthe removal of pollutants from aqueous environments. In order to increase the\nstability and dispersion, nanospheres with iron oxide core and titanium dioxide\ncoating were used in this research and their ability to absorb Congo red color\nwas evaluated. Iron oxide-titanium oxide nanospheres were prepared based on the\ncoprecipitation method and then their physical properties were determined using\na tunneling electron microscope (TEM) and an X-ray diffraction device.\nMorphological investigation of the absorbent surface showed that iron\noxide-titanium oxide nanospheres sized about 5 to 10 nm. X-ray dispersion\nsurvey also suggested the high purity of the sample. In addition, the\nabsorption rate was measured in the presence of ultrasound waves and the\nresults indicated that the capacity of the synthesized sample to absorb Congo\nred is greatly dependent on the intensity power of ultrasound waves, as the\nabsorption rate reaches 100% at powers above 30 watts.\n",
"title": "Effects of ultrasound waves intensity on the removal of Congo red color from the textile industry wastewater by Fe3O4@TiO2 core-shell nanospheres"
}
| null | null |
[
"Physics"
] | null | true | null |
17559
| null |
Validated
| null | null |
null |
{
"abstract": " This paper provides the generating series for the embedding of tree-like\ngraphs of arbitrary number of vertices, accourding to their genus. It applies\nand extends the techniques of Chan, where it was used to give an alternate\nproof of the Goulden and Slofstra formula. Furthermore, this greatly\ngeneralizes the famous Harer-Zagier formula, which computes the Euler\ncharacteristic of the moduli space of curves, and is equivalent to the\ncomputation of one vertex maps.\n",
"title": "Enumeration of Tree-like Maps with Arbitrary Number of Vertices"
}
| null | null | null | null | true | null |
17560
| null |
Default
| null | null |
null |
{
"abstract": " We report results of simultaneous x-ray reflectivity and grazing incidence\nx-ray fluorescence measurements in combination with x-ray standing wave\nassisted depth resolved near edge x-ray absorption measurements to reveal new\ninsights on chemical speciation of W in a W-B4C superlattice structure.\nInterestingly, our results show existence of various unusual electronic states\nfor the W atoms especially those sitting at the surface and interface boundary\nof a thin film medium as compared to that of the bulk. These observations are\nfound to be consistent with the results obtained using first principles\ncalculations. Unlike the conventional x-ray absorption measurements the present\napproach has an advantage that it permits the determination of depth resolved\nchemical nature of an element in the thin layered materials at atomic length\nscale resolutions.\n",
"title": "Depth resolved chemical speciation of a superlattice structure"
}
| null | null | null | null | true | null |
17561
| null |
Default
| null | null |
null |
{
"abstract": " The observation of micron size spin relaxation makes graphene a promising\nmaterial for applications in spintronics requiring long distance spin\ncommunication. However, spin dependent scatterings at the contact/graphene\ninterfaces affect the spin injection efficiencies and hence prevent the\nmaterial from achieving its full potential. While this major issue could be\neliminated by nondestructive direct optical spin injection schemes, graphenes\nintrinsically low spin orbit coupling strength and optical absorption place an\nobstacle in their realization. We overcome this challenge by creating sharp\nartificial interfaces between graphene and WSe2 monolayers. Application of a\ncircularly polarized light activates the spin polarized charge carriers in the\nWSe2 layer due to its spin coupled valley selective absorption. These carriers\ndiffuse into the superjacent graphene layer, transport over a 3.5 um distance,\nand are finally detected electrically using BN/Co contacts in a non local\ngeometry. Polarization dependent measurements confirm the spin origin of the\nnon local signal.\n",
"title": "Optospintronics in graphene via proximity coupling"
}
| null | null | null | null | true | null |
17562
| null |
Default
| null | null |
null |
{
"abstract": " Advanced gravitational-wave detectors such as the Laser Interferometer\nGravitational-Wave Observatories (LIGO) require an unprecedented level of\nisolation from the ground. When in operation, they are expected to observe\nchanges in the space-time continuum of less than one thousandth of the diameter\nof a proton. Strong teleseismic events like earthquakes disrupt the proper\nfunctioning of the detectors, and result in a loss of data until the detectors\ncan be returned to their operating states. An earthquake early-warning system,\nas well as a prediction model have been developed to help understand the impact\nof earthquakes on LIGO. This paper describes a control strategy to use this\nearly-warning system to reduce the LIGO downtime by 30%. It also presents a\nplan to implement this new earthquake configuration in the LIGO automation\nsystem.\n",
"title": "Control strategy to limit duty cycle impact of earthquakes on the LIGO gravitational-wave detectors"
}
| null | null | null | null | true | null |
17563
| null |
Default
| null | null |
null |
{
"abstract": " The design of spacecraft trajectories for missions visiting multiple\ncelestial bodies is here framed as a multi-objective bilevel optimization\nproblem. A comparative study is performed to assess the performance of\ndifferent Beam Search algorithms at tackling the combinatorial problem of\nfinding the ideal sequence of bodies. Special focus is placed on the\ndevelopment of a new hybridization between Beam Search and the Population-based\nAnt Colony Optimization algorithm. An experimental evaluation shows all\nalgorithms achieving exceptional performance on a hard benchmark problem. It is\nfound that a properly tuned deterministic Beam Search always outperforms the\nremaining variants. Beam P-ACO, however, demonstrates lower parameter\nsensitivity, while offering superior worst-case performance. Being an anytime\nalgorithm, it is then found to be the preferable choice for certain practical\napplications.\n",
"title": "Multi-rendezvous Spacecraft Trajectory Optimization with Beam P-ACO"
}
| null | null | null | null | true | null |
17564
| null |
Default
| null | null |
null |
{
"abstract": " We prove that for every Bushnell-Kutzko type that satisfies a certain\nrigidity assumption, the equivalence of categories between the corresponding\nBernstein component and the category of modules for the Hecke algebra of the\ntype induces a bijection between irreducible unitary representations in the two\ncategories. This is a generalization of the unitarity criterion of Barbasch and\nMoy for representations with Iwahori fixed vectors.\n",
"title": "Types and unitary representations of reductive p-adic groups"
}
| null | null | null | null | true | null |
17565
| null |
Default
| null | null |
null |
{
"abstract": " Let $k = \\mathbb{F}_{q}(T)$ be the rational function field over a finite\nfield $\\mathbb{F}_{q}$, where $q$ is a power of $2$. In this paper we solve the\nproblem of averaging the quadratic $L$-functions $L(s, \\chi_{u})$ over\nfundamental discriminants. Any separable quadratic extension $K$ of $k$ is of\nthe form $K = k(x_{u})$, where $x_{u}$ is a zero of $X^2+X+u=0$ for some $u\\in\nk$. We characterize the family $\\mathcal I$ (resp. $\\mathcal F$, $\\mathcal F'$)\nof rational functions $u\\in k$ such that any separable quadratic extension $K$\nof $k$ in which the infinite prime $\\infty = (1/T)$ of $k$ ramifies (resp.\nsplits, is inert) can be written as $K = k(x_{u})$ with a unique $u\\in\\mathcal\nI$ (resp. $u\\in\\mathcal F$, $u\\in\\mathcal F'$). For almost all $s\\in\\mathbb C$\nwith ${\\rm Re}(s)\\ge \\frac{1}2$, we obtain the asymptotic formulas for the\nsummation of $L(s,\\chi_{u})$ over all $k(x_{u})$ with $u\\in \\mathcal I$, all\n$k(x_{u})$ with $u\\in \\mathcal F$ or all $k(x_{u})$ with $u\\in \\mathcal F'$ of\ngiven genus. As applications, we obtain the asymptotic mean value formulas of\n$L$-functions at $s=\\frac{1}2$ and $s=1$ and the asymptotic mean value formulas\nof the class number $h_{u}$ or the class number times regulator $h_{u} R_{u}$.\n",
"title": "Average values of L-functions in even characteristic"
}
| null | null | null | null | true | null |
17566
| null |
Default
| null | null |
null |
{
"abstract": " We present a result on the number of decoupled molecules for systems binding\ntwo different types of ligands. In the case of $n$ and $2$ binding sites\nrespectively, we show that, generically, there are $2(n!)^{2}$ decoupled\nmolecules with the same binding polynomial. For molecules with more binding\nsites for the second ligand, we provide computational results.\n",
"title": "Decoupled molecules with binding polynomials of bidegree (n,2)"
}
| null | null |
[
"Computer Science",
"Physics"
] | null | true | null |
17567
| null |
Validated
| null | null |
null |
{
"abstract": " Learning to remember long sequences remains a challenging task for recurrent\nneural networks. Register memory and attention mechanisms were both proposed to\nresolve the issue with either high computational cost to retain memory\ndifferentiability, or by discounting the RNN representation learning towards\nencoding shorter local contexts than encouraging long sequence encoding.\nAssociative memory, which studies the compression of multiple patterns in a\nfixed size memory, were rarely considered in recent years. Although some recent\nwork tries to introduce associative memory in RNN and mimic the energy decay\nprocess in Hopfield nets, it inherits the shortcoming of rule-based memory\nupdates, and the memory capacity is limited. This paper proposes a method to\nlearn the memory update rule jointly with task objective to improve memory\ncapacity for remembering long sequences. Also, we propose an architecture that\nuses multiple such associative memory for more complex input encoding. We\nobserved some interesting facts when compared to other RNN architectures on\nsome well-studied sequence learning tasks.\n",
"title": "Learning to update Auto-associative Memory in Recurrent Neural Networks for Improving Sequence Memorization"
}
| null | null | null | null | true | null |
17568
| null |
Default
| null | null |
null |
{
"abstract": " Increasingly, Internet of Things (IoT) domains, such as sensor networks,\nsmart cities, and social networks, generate vast amounts of data. Such data are\nnot only unbounded and rapidly evolving. Rather, the content thereof\ndynamically evolves over time, often in unforeseen ways. These variations are\ndue to so-called concept drifts, caused by changes in the underlying data\ngeneration mechanisms. In a classification setting, concept drift causes the\npreviously learned models to become inaccurate, unsafe and even unusable.\nAccordingly, concept drifts need to be detected, and handled, as soon as\npossible. In medical applications and emergency response settings, for example,\nchange in behaviours should be detected in near real-time, to avoid potential\nloss of life. To this end, we introduce the McDiarmid Drift Detection Method\n(MDDM), which utilizes McDiarmid's inequality in order to detect concept drift.\nThe MDDM approach proceeds by sliding a window over prediction results, and\nassociate window entries with weights. Higher weights are assigned to the most\nrecent entries, in order to emphasize their importance. As instances are\nprocessed, the detection algorithm compares a weighted mean of elements inside\nthe sliding window with the maximum weighted mean observed so far. A\nsignificant difference between the two weighted means, upper-bounded by the\nMcDiarmid inequality, implies a concept drift. Our extensive experimentation\nagainst synthetic and real-world data streams show that our novel method\noutperforms the state-of-the-art. Specifically, MDDM yields shorter detection\ndelays as well as lower false negative rates, while maintaining high\nclassification accuracies.\n",
"title": "McDiarmid Drift Detection Methods for Evolving Data Streams"
}
| null | null | null | null | true | null |
17569
| null |
Default
| null | null |
null |
{
"abstract": " It has recently been shown that yield in amorphous solids under oscillatory\nshear is a dynamical transition from asymptotically periodic to asymptotically\nchaotic, diffusive dynamics. However, the type and universality class of this\ntransition are still undecided. Here we show that the diffusive behavior of the\nvector of coordinates of the particles comprising an amorphous solid when\nsubject to oscillatory shear, is analogous to that of a particle diffusing in a\npercolating lattice, the so-called \"ant in the labyrinth\" problem, and that\nyield corresponds to a percolation transition in the lattice. We explain this\nas a transition in the connectivity of the energy landscape, which affects the\nphase-space regions accessible to the coordinate vector for a given maximal\nstrain amplitude. This transition provides a natural explanation to the\nobserved limit-cycles, periods larger than one and diverging time-scales at\nyield.\n",
"title": "Yield in Amorphous Solids: The Ant in the Energy Landscape Labyrinth"
}
| null | null | null | null | true | null |
17570
| null |
Default
| null | null |
null |
{
"abstract": " We present a review of data types and statistical methods often encountered\nin astronomy. The aim is to provide an introduction to statistical applications\nin astronomy for statisticians and computer scientists. We highlight the\ncomplex, often hierarchical, nature of many astronomy inference problems and\nadvocate for cross-disciplinary collaborations to address these challenges.\n",
"title": "Statistical methods in astronomy"
}
| null | null | null | null | true | null |
17571
| null |
Default
| null | null |
null |
{
"abstract": " We provide sufficient conditions to guarantee that a translation based cipher\nis not vulnerable with respect to the partition-based trapdoor. This trapdoor\nhas been introduced, recently, by Bannier et al. (2016) and it generalizes that\nintroduced by Paterson in 1999. Moreover, we discuss the fact that studying the\ngroup generated by the round functions of a block cipher may not be sufficient\nto guarantee security against these trapdoors for the cipher.\n",
"title": "A note on some algebraic trapdoors for block ciphers"
}
| null | null | null | null | true | null |
17572
| null |
Default
| null | null |
null |
{
"abstract": " The paper, as a new contribution, aims to explore the impacts of\nbi-demographic structure on the current account and growth. By using a SVAR\nmodeling, we track the dynamic impacts between the underlying variables of the\nSaudi economy. New insights have been developed to study the interrelations\nbetween population growth, current account and economic growth inside the\nneoclassical theory of population. The long-run net impact on economic growth\nof the bi-population growth is negative, due to the typically lower skill sets\nof the immigrant labor population. Besides, the negative long-run contribution\nof immigrant workers to the current account growth largely exceeds that of\ncontributions from the native population, because of the increasing levels of\nremittance outflows from the country. We find that a positive shock in\nimmigration leads to a negative impact on native active age ratio. Thus, the\nimmigrants appear to be more substitutes than complements for native workers.\n",
"title": "Bi-Demographic Changes and Current Account using SVAR Modeling"
}
| null | null | null | null | true | null |
17573
| null |
Default
| null | null |
null |
{
"abstract": " Classification performances of the supervised machine learning techniques\nsuch as support vector machines, neural networks and logistic regression are\ncompared for modulation recognition purposes. The simple and robust features\nare used to distinguish continuous-phase FSK from QAM-PSK signals. Signals\nhaving root-raised-cosine shaped pulses are simulated in extreme noisy\nconditions having joint impurities of block fading, lack of symbol and sampling\nsynchronization, carrier offset, and additive white Gaussian noise. The\nfeatures are based on sample mean and sample variance of the imaginary part of\nthe product of two consecutive complex signal values.\n",
"title": "Supervised Machine Learning for Signals Having RRC Shaped Pulses"
}
| null | null | null | null | true | null |
17574
| null |
Default
| null | null |
null |
{
"abstract": " We propose an autoencoding sequence-based transceiver for communication over\ndispersive channels with intensity modulation and direct detection (IM/DD),\ndesigned as a bidirectional deep recurrent neural network (BRNN). The receiver\nuses a sliding window technique to allow for efficient data stream estimation.\nWe find that this sliding window BRNN (SBRNN), based on end-to-end deep\nlearning of the communication system, achieves a significant bit-error-rate\nreduction at all examined distances in comparison to previous block-based\nautoencoders implemented as feed-forward neural networks (FFNNs), leading to an\nincrease of the transmission distance. We also compare the end-to-end SBRNN\nwith a state-of-the-art IM/DD solution based on two level pulse amplitude\nmodulation with an FFNN receiver, simultaneously processing multiple received\nsymbols and approximating nonlinear Volterra equalization. Our results show\nthat the SBRNN outperforms such systems at both 42 and 84\\,Gb/s, while training\nfewer parameters. Our novel SBRNN design aims at tailoring the end-to-end deep\nlearning-based systems for communication over nonlinear channels with memory,\nsuch as the optical IM/DD fiber channel.\n",
"title": "End-to-End Optimized Transmission over Dispersive Intensity-Modulated Channels Using Bidirectional Recurrent Neural Networks"
}
| null | null | null | null | true | null |
17575
| null |
Default
| null | null |
null |
{
"abstract": " We prove effective Nullstellensatz and elimination theorems for difference\nequations in sequence rings. More precisely, we compute an explicit function of\ngeometric quantities associated to a system of difference equations (and these\ngeometric quantities may themselves be bounded by a function of the number of\nvariables, the order of the equations, and the degrees of the equations) so\nthat for any system of difference equations in variables $\\mathbf{x} = (x_1,\n\\ldots, x_m)$ and $\\mathbf{u} = (u_1, \\ldots, u_r)$, if these equations have\nany nontrivial consequences in the $\\mathbf{x}$ variables, then such a\nconsequence may be seen algebraically considering transforms up to the order of\nour bound. Specializing to the case of $m = 0$, we obtain an effective method\nto test whether a given system of difference equations is consistent.\n",
"title": "Effective difference elimination and Nullstellensatz"
}
| null | null |
[
"Mathematics"
] | null | true | null |
17576
| null |
Validated
| null | null |
null |
{
"abstract": " Let $q$ be a prime power of a prime $p$, $n$ a positive integer and $\\mathbb\nF_{q^n}$ the finite field with $q^n$ elements. The $k-$normal elements over\nfinite fields were introduced and characterized by Huczynska et al (2013).\nUnder the condition that $n$ is not divisible by $p$, they obtained an\nexistence result on primitive $1-$normal elements of $\\mathbb F_{q^n}$ over\n$\\mathbb F_q$ for $q>2$. In this note, we extend their result to the excluded\ncase $q=2$.\n",
"title": "A note on primitive $1-$normal elements over finite fields"
}
| null | null | null | null | true | null |
17577
| null |
Default
| null | null |
null |
{
"abstract": " Zebrafish pretectal neurons exhibit specificities for large-field optic flow\npatterns associated with rotatory or translatory body motion. We investigate\nthe hypothesis that these specificities reflect the input statistics of natural\noptic flow. Realistic motion sequences were generated using computer graphics\nsimulating self-motion in an underwater scene. Local retinal motion was\nestimated with a motion detector and encoded in four populations of\ndirectionally tuned retinal ganglion cells, represented as two signed input\nvariables. This activity was then used as input into one of two learning\nnetworks: a sparse coding network (competitive learning) and backpropagation\nnetwork (supervised learning). Both simulations develop specificities for optic\nflow which are comparable to those found in a neurophysiological study (Kubo et\nal. 2014), and relative frequencies of the various neuronal responses are best\nmodeled by the sparse coding approach. We conclude that the optic flow neurons\nin the zebrafish pretectum do reflect the optic flow statistics. The predicted\nvectorial receptive fields show typical optic flow fields but also \"Gabor\" and\ndipole-shaped patterns that likely reflect difference fields needed for\nreconstruction by linear superposition.\n",
"title": "Sparse Coding Predicts Optic Flow Specificities of Zebrafish Pretectal Neurons"
}
| null | null | null | null | true | null |
17578
| null |
Default
| null | null |
null |
{
"abstract": " Being able to predict whether a song can be a hit has impor- tant\napplications in the music industry. Although it is true that the popularity of\na song can be greatly affected by exter- nal factors such as social and\ncommercial influences, to which degree audio features computed from musical\nsignals (whom we regard as internal factors) can predict song popularity is an\ninteresting research question on its own. Motivated by the recent success of\ndeep learning techniques, we attempt to ex- tend previous work on hit song\nprediction by jointly learning the audio features and prediction models using\ndeep learning. Specifically, we experiment with a convolutional neural net-\nwork model that takes the primitive mel-spectrogram as the input for feature\nlearning, a more advanced JYnet model that uses an external song dataset for\nsupervised pre-training and auto-tagging, and the combination of these two\nmodels. We also consider the inception model to characterize audio infor-\nmation in different scales. Our experiments suggest that deep structures are\nindeed more accurate than shallow structures in predicting the popularity of\neither Chinese or Western Pop songs in Taiwan. We also use the tags predicted\nby JYnet to gain insights into the result of different models.\n",
"title": "Revisiting the problem of audio-based hit song prediction using convolutional neural networks"
}
| null | null | null | null | true | null |
17579
| null |
Default
| null | null |
null |
{
"abstract": " We train multi-task autoencoders on linguistic tasks and analyze the learned\nhidden sentence representations. The representations change significantly when\ntranslation and part-of-speech decoders are added. The more decoders a model\nemploys, the better it clusters sentences according to their syntactic\nsimilarity, as the representation space becomes less entangled. We explore the\nstructure of the representation space by interpolating between sentences, which\nyields interesting pseudo-English sentences, many of which have recognizable\nsyntactic structure. Lastly, we point out an interesting property of our\nmodels: The difference-vector between two sentences can be added to change a\nthird sentence with similar features in a meaningful way.\n",
"title": "Natural Language Multitasking: Analyzing and Improving Syntactic Saliency of Hidden Representations"
}
| null | null |
[
"Statistics"
] | null | true | null |
17580
| null |
Validated
| null | null |
null |
{
"abstract": " Time series forecasting is a crucial component of many important\napplications, ranging from forecasting the stock markets to energy load\nprediction. The high-dimensionality, velocity and variety of the data collected\nin these applications pose significant and unique challenges that must be\ncarefully addressed for each of them. In this work, a novel Temporal Logistic\nNeural Bag-of-Features approach, that can be used to tackle these challenges,\nis proposed. The proposed method can be effectively combined with deep neural\nnetworks, leading to powerful deep learning models for time series analysis.\nHowever, combining existing BoF formulations with deep feature extractors pose\nsignificant challenges: the distribution of the input features is not\nstationary, tuning the hyper-parameters of the model can be especially\ndifficult and the normalizations involved in the BoF model can cause\nsignificant instabilities during the training process. The proposed method is\ncapable of overcoming these limitations by a employing a novel adaptive scaling\nmechanism and replacing the classical Gaussian-based density estimation\ninvolved in the regular BoF model with a logistic kernel. The effectiveness of\nthe proposed approach is demonstrated using extensive experiments on a\nlarge-scale financial time series dataset that consists of more than 4 million\nlimit orders.\n",
"title": "Temporal Logistic Neural Bag-of-Features for Financial Time series Forecasting leveraging Limit Order Book Data"
}
| null | null | null | null | true | null |
17581
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the ground state properties of ultracold atoms with long range\ninteractions trapped in a two leg ladder configuration in the presence of an\nartificial magnetic field. Using a Gross-Pitaevskii approach and a mean field\nGutzwiller variational method, we explore both the weakly interacting and\nstrongly interacting regime, respectively. We calculate the boundaries between\nthe density-wave/supersolid and the Mott-insulator/superfluid phases as a\nfunction of magnetic flux and uncover regions of supersolidity. The mean-field\nresults are confirmed by numerical simulations using a cluster mean field\napproach.\n",
"title": "Extended Bose Hubbard model for two leg ladder systems in artificial magnetic fields"
}
| null | null | null | null | true | null |
17582
| null |
Default
| null | null |
null |
{
"abstract": " Recent determination of the Hubble constant via Cepheid-calibrated supernovae\nby \\citet{riess_2.4_2016} (R16) find $\\sim 3\\sigma$ tension with inferences\nbased on cosmic microwave background temperature and polarization measurements\nfrom $Planck$. This tension could be an indication of inadequacies in the\nconcordance $\\Lambda$CDM model. Here we investigate the possibility that the\ndiscrepancy could instead be due to systematic bias or uncertainty in the\nCepheid calibration step of the distance ladder measurement by R16. We consider\nvariations in total-to-selective extinction of Cepheid flux as a function of\nline-of-sight, hidden structure in the period-luminosity relationship, and\npotentially different intrinsic color distributions of Cepheids as a function\nof host galaxy. Considering all potential sources of error, our final\ndetermination of $H_0 = 73.3 \\pm 1.7~{\\rm km/s/Mpc}$ (not including systematic\nerrors from the treatment of geometric distances or Type Ia Supernovae) shows\nremarkable robustness and agreement with R16. We conclude systematics from the\nmodeling of Cepheid photometry, including Cepheid selection criteria, cannot\nexplain the observed tension between Cepheid-variable and CMB-based inferences\nof the Hubble constant. Considering a `model-independent' approach to relating\nCepheids in galaxies with known distances to Cepheids in galaxies hosting a\nType Ia supernova and finding agreement with the R16 result, we conclude no\ngeneralization of the model relating anchor and host Cepheid magnitude\nmeasurements can introduce significant bias in the $H_0$ inference.\n",
"title": "Insensitivity of The Distance Ladder Hubble Constant Determination to Cepheid Calibration Modeling Choices"
}
| null | null | null | null | true | null |
17583
| null |
Default
| null | null |
null |
{
"abstract": " Automatic generation of caption to describe the content of an image has been\ngaining a lot of research interests recently, where most of the existing works\ntreat the image caption as pure sequential data. Natural language, however\npossess a temporal hierarchy structure, with complex dependencies between each\nsubsequence. In this paper, we propose a phrase-based hierarchical Long\nShort-Term Memory (phi-LSTM) model to generate image description. In contrast\nto the conventional solutions that generate caption in a pure sequential\nmanner, our proposed model decodes image caption from phrase to sentence. It\nconsists of a phrase decoder at the bottom hierarchy to decode noun phrases of\nvariable length, and an abbreviated sentence decoder at the upper hierarchy to\ndecode an abbreviated form of the image description. A complete image caption\nis formed by combining the generated phrases with sentence during the inference\nstage. Empirically, our proposed model shows a better or competitive result on\nthe Flickr8k, Flickr30k and MS-COCO datasets in comparison to the state-of-the\nart models. We also show that our proposed model is able to generate more novel\ncaptions (not seen in the training data) which are richer in word contents in\nall these three datasets.\n",
"title": "Phrase-based Image Captioning with Hierarchical LSTM Model"
}
| null | null | null | null | true | null |
17584
| null |
Default
| null | null |
null |
{
"abstract": " Swirl-switching is a low-frequency oscillatory phenomenon which affects the\nDean vortices in bent pipes and may cause fatigue in piping systems. Despite\nthirty years worth of research, the mechanism that causes these oscillations\nand the frequencies that characterise them remain unclear. Here we show that a\nthree-dimensional wave-like structure is responsible for the low-frequency\nswitching of the dominant Dean vortex. The present study, performed via direct\nnumerical simulation, focuses on the turbulent flow through a 90 \\degree pipe\nbend preceded and followed by straight pipe segments. A pipe with curvature 0.3\n(defined as ratio between pipe radius and bend radius) is studied for a bulk\nReynolds number Re = 11 700, corresponding to a friction Reynolds number\nRe_\\tau \\approx 360. Synthetic turbulence is generated at the inflow section\nand used instead of the classical recycling method in order to avoid the\ninterference between recycling and swirl-switching frequencies. The flow field\nis analysed by three-dimensional proper orthogonal decomposition (POD) which\nfor the first time allows the identification of the source of swirl-switching:\na wave-like structure that originates in the pipe bend. Contrary to some\nprevious studies, the flow in the upstream pipe does not show any direct\ninfluence on the swirl-switching modes. Our analysis further shows that a\nthree- dimensional characterisation of the modes is crucial to understand the\nmechanism, and that reconstructions based on 2D POD modes are incomplete.\n",
"title": "The three-dimensional structure of swirl-switching in bent pipe flow"
}
| null | null | null | null | true | null |
17585
| null |
Default
| null | null |
null |
{
"abstract": " A key advance in learning generative models is the use of amortized inference\ndistributions that are jointly trained with the models. We find that existing\ntraining objectives for variational autoencoders can lead to inaccurate\namortized inference distributions and, in some cases, improving the objective\nprovably degrades the inference quality. In addition, it has been observed that\nvariational autoencoders tend to ignore the latent variables when combined with\na decoding distribution that is too flexible. We again identify the cause in\nexisting training criteria and propose a new class of objectives (InfoVAE) that\nmitigate these problems. We show that our model can significantly improve the\nquality of the variational posterior and can make effective use of the latent\nfeatures regardless of the flexibility of the decoding distribution. Through\nextensive qualitative and quantitative analyses, we demonstrate that our models\noutperform competing approaches on multiple performance metrics.\n",
"title": "InfoVAE: Information Maximizing Variational Autoencoders"
}
| null | null | null | null | true | null |
17586
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we propose a novel learning method for image classification\ncalled Between-Class learning (BC learning). We generate between-class images\nby mixing two images belonging to different classes with a random ratio. We\nthen input the mixed image to the model and train the model to output the\nmixing ratio. BC learning has the ability to impose constraints on the shape of\nthe feature distributions, and thus the generalization ability is improved. BC\nlearning is originally a method developed for sounds, which can be digitally\nmixed. Mixing two image data does not appear to make sense; however, we argue\nthat because convolutional neural networks have an aspect of treating input\ndata as waveforms, what works on sounds must also work on images. First, we\npropose a simple mixing method using internal divisions, which surprisingly\nproves to significantly improve performance. Second, we propose a mixing method\nthat treats the images as waveforms, which leads to a further improvement in\nperformance. As a result, we achieved 19.4% and 2.26% top-1 errors on\nImageNet-1K and CIFAR-10, respectively.\n",
"title": "Between-class Learning for Image Classification"
}
| null | null | null | null | true | null |
17587
| null |
Default
| null | null |
null |
{
"abstract": " The sensitivity of molecular dynamics on changes in the potential energy\nfunction plays an important role in understanding the dynamics and function of\ncomplex molecules.We present a method to obtain path ensemble averages of a\nperturbed dynamics from a set of paths generated by a reference dynamics. It is\nbased on the concept of path probability measure and the Girsanov theorem, a\nresult from stochastic analysis to estimate a change of measure of a path\nensemble. Since Markov state models (MSM) of the molecular dynamics can be\nformulated as a combined phase-space and path ensemble average, the method can\nbe extended toreweight MSMs by combining it with a reweighting of the Boltzmann\ndistribution. We demonstrate how to efficiently implement the Girsanov\nreweighting in a molecular dynamics simulation program by calculating parts of\nthe reweighting factor \"on the fly\" during the simulation, and we benchmark the\nmethod on test systems ranging from a two-dimensional diffusion process to an\nartificial many-body system and alanine dipeptide and valine dipeptide in\nimplicit and explicit water. The method can be used to study the sensitivity of\nmolecular dynamics on external perturbations as well as to reweight\ntrajectories generated by enhanced sampling schemes to the original dynamics.\n",
"title": "Girsanov reweighting for path ensembles and Markov state models"
}
| null | null | null | null | true | null |
17588
| null |
Default
| null | null |
null |
{
"abstract": " In this work we study a multi-agent coordination problem in which agents are\nonly able to communicate with each other intermittently through a cloud server.\nTo reduce the amount of required communication, we develop a self-triggered\nalgorithm that allows agents to communicate with the cloud only when necessary\nrather than at some fixed period. Unlike the vast majority of similar works\nthat propose distributed event- and/or self-triggered control laws, this work\ndoesn't assume agents can be \"listening\" continuously. In other words, when an\nevent is triggered by one agent, neighboring agents will not be aware of this\nuntil the next time they establish communication with the cloud themselves.\nUsing a notion of \"promises\" about future control inputs, agents are able to\nkeep track of higher quality estimates about their neighbors allowing them to\nstay disconnected from the cloud for longer periods of time while still\nguaranteeing a positive contribution to the global task. We prove that our\nself-triggered coordination algorithm guarantees that the system asymptotically\nreaches the set of desired states. Simulations illustrate our results.\n",
"title": "Coordination of multi-agent systems via asynchronous cloud communication"
}
| null | null | null | null | true | null |
17589
| null |
Default
| null | null |
null |
{
"abstract": " Pattern lock has been widely used for authentication to protect user privacy\non mobile devices (e.g., smartphones and tablets). Given its pervasive usage,\nthe compromise of pattern lock could lead to serious consequences. Several\nattacks have been constructed to crack the lock. However, these approaches\nrequire the attackers to either be physically close to the target device or be\nable to manipulate the network facilities (e.g., WiFi hotspots) used by the\nvictims. Therefore, the effectiveness of the attacks is significantly impacted\nby the environment of mobile devices. Also, these attacks are not scalable\nsince they cannot easily infer unlock patterns of a large number of devices.\nMotivated by an observation that fingertip motions on the screen of a mobile\ndevice can be captured by analyzing surrounding acoustic signals on it, we\npropose PatternListener, a novel acoustic attack that cracks pattern lock by\nanalyzing imperceptible acoustic signals reflected by the fingertip. It\nleverages speakers and microphones of the victim's device to play imperceptible\naudio and record the acoustic signals reflected by the fingertip. In\nparticular, it infers each unlock pattern by analyzing individual lines that\ncompose the pattern and are the trajectories of the fingertip. We propose\nseveral algorithms to construct signal segments according to the captured\nsignals for each line and infer possible candidates of each individual line\naccording to the signal segments. Finally, we map all line candidates into grid\npatterns and thereby obtain the candidates of the entire unlock pattern. We\nimplement a PatternListener prototype by using off-the-shelf smartphones and\nthoroughly evaluate it using 130 unique patterns. The real experimental results\ndemonstrate that PatternListener can successfully exploit over 90% patterns\nwithin five attempts.\n",
"title": "PatternListener: Cracking Android Pattern Lock Using Acoustic Signals"
}
| null | null | null | null | true | null |
17590
| null |
Default
| null | null |
null |
{
"abstract": " Many analysis and machine learning tasks require the availability of marginal\nstatistics on multidimensional datasets while providing strong privacy\nguarantees for the data subjects. Applications for these statistics range from\nfinding correlations in the data to fitting sophisticated prediction models. In\nthis paper, we provide a set of algorithms for materializing marginal\nstatistics under the strong model of local differential privacy. We prove the\nfirst tight theoretical bounds on the accuracy of marginals compiled under each\napproach, perform empirical evaluation to confirm these bounds, and evaluate\nthem for tasks such as modeling and correlation testing. Our results show that\nreleasing information based on (local) Fourier transformations of the input is\npreferable to alternatives based directly on (local) marginals.\n",
"title": "Marginal Release Under Local Differential Privacy"
}
| null | null | null | null | true | null |
17591
| null |
Default
| null | null |
null |
{
"abstract": " In the last decades there have been an increasing interest in improving the\naccuracy of spacecraft navigation and trajectory data. In the course of this\nplan some anomalies have been found that cannot, in principle, be explained in\nthe context of the most accurate orbital models including all known effects\nfrom classical dynamics and general relativity. Of particular interest for its\npuzzling nature, and the lack of any accepted explanation for the moment, is\nthe flyby anomaly discovered in some spacecraft flybys of the Earth over the\ncourse of twenty years. This anomaly manifest itself as the impossibility of\nmatching the pre and post-encounter Doppler tracking and ranging data within a\nsingle orbit but, on the contrary, a difference of a few mm$/$s in the\nasymptotic velocities is required to perform the fitting.\nNevertheless, no dedicated missions have been carried out to elucidate the\norigin of this phenomenon with the objective either of revising our\nunderstanding of gravity or to improve the accuracy of spacecraft Doppler\ntracking by revealing a conventional origin.\nWith the occasion of the Juno mission arrival at Jupiter and the close flybys\nof this planet, that are currently been performed, we have developed an orbital\nmodel suited to the time window close to the perijove. This model shows that an\nanomalous acceleration of a few mm$/$s$^2$ is also present in this case. The\nchance for overlooked conventional or possible unconventional explanations is\ndiscussed.\n",
"title": "A possible flyby anomaly for Juno at Jupiter"
}
| null | null |
[
"Physics"
] | null | true | null |
17592
| null |
Validated
| null | null |
null |
{
"abstract": " We conjecture the universal probability distribution at large time for the\none-point height in the 1D Kardar-Parisi-Zhang (KPZ) stochastic growth\nuniversality class, with initial conditions interpolating from any one of the\nthree main classes (droplet, flat, stationary) on the left, to another on the\nright, allowing for drifts and also for a step near the origin. The result is\nobtained from a replica Bethe ansatz calculation starting from the KPZ\ncontinuum equation, together with a \"decoupling assumption\" in the large time\nlimit. Some cases are checked to be equivalent to previously known results from\nother models in the same class, which provides a test of the method, others\nappear to be new. In particular we obtain the crossover distribution between\nflat and stationary initial conditions (crossover from Airy$_1$ to Airy$_{\\rm\nstat}$) in a simple compact form.\n",
"title": "Crossover between various initial conditions in KPZ growth: flat to stationary"
}
| null | null | null | null | true | null |
17593
| null |
Default
| null | null |
null |
{
"abstract": " We carry out a study of the statistical distribution of rainfall\nprecipitation data for 20 cites in India. We have determined the best-fit\nprobability distribution for these cities from the monthly precipitation data\nspanning 100 years of observations from 1901 to 2002. To fit the observed data,\nwe considered 10 different distributions. The efficacy of the fits for these\ndistributions was evaluated using four empirical non-parametric goodness-of-fit\ntests namely Kolmogorov-Smirnov, Anderson-Darling, Chi-Square, Akaike\ninformation criterion, and Bayesian Information criterion. Finally, the\nbest-fit distribution using each of these tests were reported, by combining the\nresults from the model comparison tests. We then find that for most of the\ncities, Generalized Extreme-Value Distribution or Inverse Gaussian Distribution\nmost adequately fits the observed data.\n",
"title": "Multimodel Response Assessment for Monthly Rainfall Distribution in Some Selected Indian Cities Using Best Fit Probability as a Tool"
}
| null | null | null | null | true | null |
17594
| null |
Default
| null | null |
null |
{
"abstract": " Bin Packing problems have been widely studied because of their broad\napplications in different domains. Known as a set of NP-hard problems, they\nhave different vari- ations and many heuristics have been proposed for\nobtaining approximate solutions. Specifically, for the 1D variable sized bin\npacking problem, the two key sets of optimization heuristics are the bin\nassignment and the bin allocation. Usually the performance of a single static\noptimization heuristic can not beat that of a dynamic one which is tailored for\neach bin packing instance. Building such an adaptive system requires modeling\nthe relationship between bin features and packing perform profiles. The primary\ndrawbacks of traditional AI machine learnings for this task are the natural\nlimitations of feature engineering, such as the curse of dimensionality and\nfeature selection quality. We introduce a deep learning approach to overcome\nthe drawbacks by applying a large training data set, auto feature selection and\nfast, accurate labeling. We show in this paper how to build such a system by\nboth theoretical formulation and engineering practices. Our prediction system\nachieves up to 89% training accuracy and 72% validation accuracy to select the\nbest heuristic that can generate a better quality bin packing solution.\n",
"title": "Small Boxes Big Data: A Deep Learning Approach to Optimize Variable Sized Bin Packing"
}
| null | null | null | null | true | null |
17595
| null |
Default
| null | null |
null |
{
"abstract": " Human populations exhibit complex behaviors---characterized by long-range\ncorrelations and surges in activity---across a range of social, political, and\ntechnological contexts. Yet it remains unclear where these collective behaviors\ncome from, or if there even exists a set of unifying principles. Indeed,\nexisting explanations typically rely on context-specific mechanisms, such as\ntraffic jams driven by work schedules or spikes in online traffic induced by\nsignificant events. However, analogies with statistical mechanics suggest a\nmore general mechanism: that collective patterns can emerge organically from\nfine-scale interactions within a population. Here, across four different modes\nof human activity, we show that the simplest correlations in a\npopulation---those between pairs of individuals---can yield accurate\nquantitative predictions for the large-scale behavior of the entire population.\nTo quantify the minimal consequences of pairwise correlations, we employ the\nprinciple of maximum entropy, making our description equivalent to an Ising\nmodel whose interactions and external fields are notably calculated from past\nobservations of population activity. In addition to providing accurate\nquantitative predictions, we show that the topology of learned Ising\ninteractions resembles the network of inter-human communication within a\npopulation. Together, these results demonstrate that fine-scale correlations\ncan be used to predict large-scale social behaviors, a perspective that has\ncritical implications for modeling and resource allocation in human\npopulations.\n",
"title": "Surges of collective human activity emerge from simple pairwise correlations"
}
| null | null |
[
"Computer Science"
] | null | true | null |
17596
| null |
Validated
| null | null |
null |
{
"abstract": " Although timely sepsis diagnosis and prompt interventions in Intensive Care\nUnit (ICU) patients are associated with reduced mortality, early clinical\nrecognition is frequently impeded by non-specific signs of infection and\nfailure to detect signs of sepsis-induced organ dysfunction in a constellation\nof dynamically changing physiological data. The goal of this work is to\nidentify patient at risk of life-threatening sepsis utilizing a data-centered\nand machine learning-driven approach. We derive a mortality risk predictive\ndynamic Bayesian network (DBN) guided by a customized sepsis knowledgebase and\ncompare the predictive accuracy of the derived DBN with the Sepsis-related\nOrgan Failure Assessment (SOFA) score, the Quick SOFA (qSOFA) score, the\nSimplified Acute Physiological Score (SAPS-II) and the Modified Early Warning\nScore (MEWS) tools.\nA customized sepsis ontology was used to derive the DBN node structure and\nsemantically characterize temporal features derived from both structured\nphysiological data and unstructured clinical notes. We assessed the performance\nin predicting mortality risk of the DBN predictive model and compared\nperformance to other models using Receiver Operating Characteristic (ROC)\ncurves, area under curve (AUROC), calibration curves, and risk distributions.\nThe derived dataset consists of 24,506 ICU stays from 19,623 patients with\nevidence of suspected infection, with 2,829 patients deceased at discharge. The\nDBN AUROC was found to be 0.91, which outperformed the SOFA (0.843), qSOFA\n(0.66), MEWS (0.73), and SAPS-II (0.77) scoring tools. Continuous Net\nReclassification Index and Integrated Discrimination Improvement analysis\nsupported the superiority DBN. Compared with conventional rule-based risk\nscoring tools, the sepsis knowledgebase-driven DBN algorithm offers improved\nperformance for predicting mortality of infected patients in ICUs.\n",
"title": "Semantically Enhanced Dynamic Bayesian Network for Detecting Sepsis Mortality Risk in ICU Patients with Infection"
}
| null | null | null | null | true | null |
17597
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we prove the Hölder regularity of bounded, uniformly\ncontinuous, viscosity solutions of some degenerate fully nonlinear equations in\nthe Heisenberg group.\n",
"title": "Hölder regularity of viscosity solutions of some fully nonlinear equations in the Heisenberg group"
}
| null | null | null | null | true | null |
17598
| null |
Default
| null | null |
null |
{
"abstract": " There exists a critical speed of propagation of the line solitons in the\nZakharov-Kuznetsov (ZK) equation such that small transversely periodic\nperturbations are unstable for line solitons with larger-than-critical speeds\nand orbitally stable for those with smaller-than-critical speeds. The normal\nform for transverse instability of the line soliton with a nearly critical\nspeed of propagation is derived by means of symplectic projections and\nnear-identity transformations. Justification of this normal form is provided\nwith the energy method. The normal form predicts a transformation of the\nunstable line solitons with larger-than-critical speeds to the orbitally stable\ntransversely modulated solitary waves.\n",
"title": "Normal form for transverse instability of the line soliton with a nearly critical speed of propagation"
}
| null | null |
[
"Physics",
"Mathematics"
] | null | true | null |
17599
| null |
Validated
| null | null |
null |
{
"abstract": " This paper describes our submission \"CLaC\" to the CoNLL-2016 shared task on\nshallow discourse parsing. We used two complementary approaches for the task. A\nstandard machine learning approach for the parsing of explicit relations, and a\ndeep learning approach for non-explicit relations. Overall, our parser achieves\nan F1-score of 0.2106 on the identification of discourse relations (0.3110 for\nexplicit relations and 0.1219 for non-explicit relations) on the blind\nCoNLL-2016 test set.\n",
"title": "The CLaC Discourse Parser at CoNLL-2016"
}
| null | null | null | null | true | null |
17600
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.