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": " Independent component analysis (ICA) decomposes multivariate data into\nmutually independent components (ICs). The ICA model is subject to a constraint\nthat at most one of these components is Gaussian, which is required for model\nidentifiability. Linear non-Gaussian component analysis (LNGCA) generalizes the\nICA model to a linear latent factor model with any number of both non-Gaussian\ncomponents (signals) and Gaussian components (noise), where observations are\nlinear combinations of independent components. Although the individual Gaussian\ncomponents are not identifiable, the Gaussian subspace is identifiable. We\nintroduce an estimator along with its optimization approach in which\nnon-Gaussian and Gaussian components are estimated simultaneously, maximizing\nthe discrepancy of each non-Gaussian component from Gaussianity while\nminimizing the discrepancy of each Gaussian component from Gaussianity. When\nthe number of non-Gaussian components is unknown, we develop a statistical test\nto determine it based on resampling and the discrepancy of estimated\ncomponents. Through a variety of simulation studies, we demonstrate the\nimprovements of our estimator over competing estimators, and we illustrate the\neffectiveness of the test to determine the number of non-Gaussian components.\nFurther, we apply our method to real data examples and demonstrate its\npractical value.\n",
"title": "Optimization and Testing in Linear Non-Gaussian Component Analysis"
}
| null | null | null | null | true | null |
12001
| null |
Default
| null | null |
null |
{
"abstract": " The New Horizons spacecraft's nominal trajectory crosses the planet's\nsatellite plane at $\\sim 10,000\\ \\rm{km}$ from the barycenter, between the\norbits of Pluto and Charon. I have investigated the risk to the spacecraft\nbased on observational limits of rings and dust within this region, assuming\nvarious particle size distributions. The best limits are placed by 2011 and\n2012 HST observations, which significantly improve on the limits from stellar\noccultations, although they do not go as close to the planet. From the HST data\nand assuming a `reasonable worst case' for the size distribution, we place a\nlimit of $N < 20$ damaging impacts by grains of radius $> 0.2\\ \\textrm{mm}$\nonto the spacecraft during the encounter. The number of hits is $\\approx$\n200$\\times$ above the NH mission requirement, and $\\approx$ $2000\\times$ above\nthe mission's desired level. Stellar occultations remain valuable because they\nare able to measure $N$ closer to the Pluto surface than direct imaging,\nalthough with a sensitivity limit several orders of magnitude higher than that\nfrom HST imaging. Neither HST nor occultations are sensitive enough to place\nlimits on $N$ at or below the mission requirements.\n",
"title": "New Horizons Ring Collision Hazard: Constraints from Earth-based Observations"
}
| null | null | null | null | true | null |
12002
| null |
Default
| null | null |
null |
{
"abstract": " Maintenance is an important activity in industry. It is performed either to\nrevive a machine/component or to prevent it from breaking down. Different\nstrategies have evolved through time, bringing maintenance to its current\nstate: condition-based and predictive maintenances. This evolution was due to\nthe increasing demand of reliability in industry. The key process of\ncondition-based and predictive maintenances is prognostics and health\nmanagement, and it is a tool to predict the remaining useful life of\nengineering assets. Nowadays, plants are required to avoid shutdowns while\noffering safety and reliability. Nevertheless, planning a maintenance activity\nrequires accurate information about the system/component health state. Such\ninformation is usually gathered by means of independent sensor nodes. In this\nstudy, we consider the case where the nodes are interconnected and form a\nwireless sensor network. As far as we know, no research work has considered\nsuch a case of study for prognostics. Regarding the importance of data\naccuracy, a good prognostics requires reliable sources of information. This is\nwhy, in this paper, we will first discuss the dependability of wireless sensor\nnetworks, and then present a state of the art in prognostic and health\nmanagement activities.\n",
"title": "Dependability of Sensor Networks for Industrial Prognostics and Health Management"
}
| null | null | null | null | true | null |
12003
| null |
Default
| null | null |
null |
{
"abstract": " Stochastic gradient methods are the workhorse (algorithms) of large-scale\noptimization problems in machine learning, signal processing, and other\ncomputational sciences and engineering. This paper studies Markov chain\ngradient descent, a variant of stochastic gradient descent where the random\nsamples are taken on the trajectory of a Markov chain. Existing results of this\nmethod assume convex objectives and a reversible Markov chain and thus have\ntheir limitations. We establish new non-ergodic convergence under wider step\nsizes, for nonconvex problems, and for non-reversible finite-state Markov\nchains. Nonconvexity makes our method applicable to broader problem classes.\nNon-reversible finite-state Markov chains, on the other hand, can mix\nsubstatially faster. To obtain these results, we introduce a new technique that\nvaries the mixing levels of the Markov chains. The reported numerical results\nvalidate our contributions.\n",
"title": "On Markov Chain Gradient Descent"
}
| null | null | null | null | true | null |
12004
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a general framework for structure-preserving model\nreduction of a secondorder network system based on graph clustering. In this\napproach, vertex dynamics are captured by the transfer functions from inputs to\nindividual states, and the dissimilarities of vertices are quantified by the\nH2-norms of the transfer function discrepancies. A greedy hierarchical\nclustering algorithm is proposed to place those vertices with similar dynamics\ninto same clusters. Then, the reduced-order model is generated by the\nPetrov-Galerkin method, where the projection is formed by the characteristic\nmatrix of the resulting network clustering. It is shown that the simplified\nsystem preserves an interconnection structure, i.e., it can be again\ninterpreted as a second-order system evolving over a reduced graph.\nFurthermore, this paper generalizes the definition of network controllability\nGramian to second-order network systems. Based on it, we develop an efficient\nmethod to compute H2-norms and derive the approximation error between the\nfull-order and reduced-order models. Finally, the approach is illustrated by\nthe example of a small-world network.\n",
"title": "Reduction of Second-Order Network Systems with Structure Preservation"
}
| null | null | null | null | true | null |
12005
| null |
Default
| null | null |
null |
{
"abstract": " Given a finitely aligned $k$-graph $\\Lambda$, we let $\\Lambda^i$ denote the\n$(k-1)$-graph formed by removing all edges of degree $e_i$ from $\\Lambda$. We\nshow that the Toeplitz-Cuntz-Krieger algebra of $\\Lambda$, denoted by\n$\\mathcal{T}C^*(\\Lambda)$, may be realised as the Toeplitz algebra of a Hilbert\n$\\mathcal{T}C^*(\\Lambda^i)$-bimodule. When $\\Lambda$ is locally-convex, we show\nthat the Cuntz-Krieger algebra of $\\Lambda$, which we denote by $C^*(\\Lambda)$,\nmay be realised as the Cuntz-Pimsner algebra of a Hilbert\n$C^*(\\Lambda^i)$-bimodule. Consequently, $\\mathcal{T}C^*(\\Lambda)$ and\n$C^*(\\Lambda)$ may be viewed as iterated Toeplitz and iterated Cuntz-Pimsner\nalgebras over $c_0(\\Lambda^0)$ respectively.\n",
"title": "Higher-rank graph algebras are iterated Cuntz-Pimsner algebras"
}
| null | null | null | null | true | null |
12006
| null |
Default
| null | null |
null |
{
"abstract": " A phenomenon can hardly be found that accompanied physical paradigms and\ntheoretical concepts in a more reflecting way than magnetism. From the\nbeginnings of metaphysics and the first classical approaches to magnetic poles\nand streamlines of the field, it has inspired modern physics on its way to the\nclassical field description of electrodynamics, and further to the quantum\nmechanical description of internal degrees of freedom of elementary particles.\nMeanwhile, magnetic manifestations have posed and still do pose complex and\noften controversially debated questions. This regards so various and utterly\ndistinct topics as quantum spin systems and the grand unification theory. This\nmay be foremost caused by the fact that all of these effects are based on\ncorrelated structures, which are induced by the interplay of dynamics and\nelementary interactions. It is strongly correlated systems that certainly\nrepresent one of the most fascinating and universal fields of research. In\nparticular, low dimensional systems are in the focus of interest, as they\nreveal strongly pronounced correlations of counterintuitive nature. As regards\nthis framework, the quantum Hall effect must be seen as one of the most\nintriguing and complex problems of modern solid state physics. Even after two\ndecades and the same number of Nobel prizes, it still keeps researchers of\nnearly all fields of physics occupied. In spite of seminal progress, its\ninherent correlated order still lacks understanding on a microscopic level.\nDespite this, it is obvious that the phenomenon is thoroughly fundamental of\nnature. To resolve some puzzles of this nature is a key topic of this thesis.\n(excerpt from abstract)\n",
"title": "Ultracold Atomic Gases in Artificial Magnetic Fields (PhD thesis)"
}
| null | null | null | null | true | null |
12007
| null |
Default
| null | null |
null |
{
"abstract": " Arrays of integers are often compressed in search engines. Though there are\nmany ways to compress integers, we are interested in the popular byte-oriented\ninteger compression techniques (e.g., VByte or Google's Varint-GB). They are\nappealing due to their simplicity and engineering convenience. Amazon's\nvarint-G8IU is one of the fastest byte-oriented compression technique published\nso far. It makes judicious use of the powerful single-instruction-multiple-data\n(SIMD) instructions available in commodity processors. To surpass varint-G8IU,\nwe present Stream VByte, a novel byte-oriented compression technique that\nseparates the control stream from the encoded data. Like varint-G8IU, Stream\nVByte is well suited for SIMD instructions. We show that Stream VByte decoding\ncan be up to twice as fast as varint-G8IU decoding over real data sets. In this\nsense, Stream VByte establishes new speed records for byte-oriented integer\ncompression, at times exceeding the speed of the memcpy function. On a 3.4GHz\nHaswell processor, it decodes more than 4 billion differentially-coded integers\nper second from RAM to L1 cache.\n",
"title": "Stream VByte: Faster Byte-Oriented Integer Compression"
}
| null | null | null | null | true | null |
12008
| null |
Default
| null | null |
null |
{
"abstract": " We study the Kondo physics of a quantum magnetic impurity in two-dimensional\ntopological superconductors (TSCs), either intrinsic or induced on the surface\nof a bulk topological insulator, using a numerical renormalization group\ntechnique. We show that, despite sharing the p + ip pairing symmetry, intrinsic\nand extrinsic TSCs host different physical processes that produce distinct\nKondo signatures. Extrinsic TSCs harbor an unusual screening mechanism\ninvolving both electron and orbital degrees of freedom that produces rich and\nprominent Kondo phenomena, especially an intriguing pseudospin Kondo singlet\nstate in the superconducting gap and a spatially anisotropic spin correlation.\nIn sharp contrast, intrinsic TSCs support a robust impurity spin doublet ground\nstate and an isotropic spin correlation. These findings advance fundamental\nknowledge of novel Kondo phenomena in TSCs and suggest experimental avenues for\ntheir detection and distinction.\n",
"title": "Kondo Signatures of a Quantum Magnetic Impurity in Topological Superconductors"
}
| null | null | null | null | true | null |
12009
| null |
Default
| null | null |
null |
{
"abstract": " In the present note we study Waldschmidt constants of Stanley-Reisner ideals\nof a hypergraph and a graph with vertices forming a bipyramid over a planar\nn-gon. The case of the hypergraph has been studied by Bocci and Franci. We\nreprove their main result. The case of the graph is new. Interestingly, both\ncases provide series of ideals with Waldschmidt constants descending to 1. It\nwould be interesting to known if there are bounded ascending sequences of\nWaldschmidt constants.\n",
"title": "Waldschmidt constants for Stanley-Reisner ideals of a class of graphs"
}
| null | null | null | null | true | null |
12010
| null |
Default
| null | null |
null |
{
"abstract": " Over the past two decades the main focus of research into first-order (FO)\nmodel checking algorithms has been on sparse relational structures -\nculminating in the FPT algorithm by Grohe, Kreutzer and Siebertz for FO model\nchecking of nowhere dense classes of graphs. On contrary to that, except the\ncase of locally bounded clique-width only little is currently known about FO\nmodel checking of dense classes of graphs or other structures. We study the FO\nmodel checking problem for dense graph classes definable by geometric means\n(intersection and visibility graphs). We obtain new nontrivial FPT results,\ne.g., for restricted subclasses of circular-arc, circle, box, disk, and\npolygon-visibility graphs. These results use the FPT algorithm by Gajarský et\nal. for FO model checking of posets of bounded width. We also complement the\ntractability results by related hardness reductions.\n",
"title": "FO model checking of geometric graphs"
}
| null | null |
[
"Computer Science"
] | null | true | null |
12011
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper we generalize three identification recursive algorithms\nbelonging to the pseudo-linear class, by introducing a predictor established on\na generalized orthonormal function basis. Contrary to the existing\nidentification schemes that use such functions, no constraint on the model\npoles is imposed. Not only this predictor parameterization offers the\nopportunity to relax the convergence conditions of the associated recursive\nschemes, but it entails a modification of the bias distribution linked to the\nbasis poles. This result is specific to pseudo-linear regression properties,\nand cannot be transposed to most of prediction error method algorithms. A\ndetailed bias distribution is provided, using the concept of equivalent\nprediction error, which reveals strong analogies between the three proposed\nschemes, corresponding to ARMAX, Output Error and a generalization of ARX\nmodels. That leads to introduce an indicator of the basis poles location effect\non the bias distribution in the frequency domain. As shown by the simulations,\nthe said basis poles play the role of tuning parameters, allowing to manage the\nmodel fit in the frequency domain, and allowing efficient identification of\nfast sampled or stiff discrete-time systems.\n",
"title": "Pseudo-linear regression identification based on generalized orthonormal transfer functions: Convergence conditions and bias distribution"
}
| null | null | null | null | true | null |
12012
| null |
Default
| null | null |
null |
{
"abstract": " Although for a number of semilinear stochastic wave equations existence and\nuniqueness results for corresponding solution processes are known from the\nliterature, these solution processes are typically not explicitly known and\nnumerical approximation methods are needed in order for mathematical modelling\nwith stochastic wave equations to become relevant for real world applications.\nThis, in turn, requires the numerical analysis of convergence rates for such\nnumerical approximation processes. A recent article by the authors proves upper\nbounds for weak errors for spatial spectral Galerkin approximations of a class\nof semilinear stochastic wave equations. The findings there are complemented by\nthe main result of this work, that provides lower bounds for weak errors which\nshow that in the general framework considered the established upper bounds can\nessentially not be improved.\n",
"title": "Lower bounds for weak approximation errors for spatial spectral Galerkin approximations of stochastic wave equations"
}
| null | null | null | null | true | null |
12013
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of learning a low-rank matrix, constrained to lie in\na linear subspace, and introduce a novel factorization for modeling such\nmatrices. A salient feature of the proposed factorization scheme is it\ndecouples the low-rank and the structural constraints onto separate factors. We\nformulate the optimization problem on the Riemannian spectrahedron manifold,\nwhere the Riemannian framework allows to develop computationally efficient\nconjugate gradient and trust-region algorithms. Experiments on problems such as\nstandard/robust/non-negative matrix completion, Hankel matrix learning and\nmulti-task learning demonstrate the efficacy of our approach. A shorter version\nof this work has been published in ICML'18.\n",
"title": "Structured low-rank matrix learning: algorithms and applications"
}
| null | null |
[
"Statistics"
] | null | true | null |
12014
| null |
Validated
| null | null |
null |
{
"abstract": " This paper establishes the first performance guarantees for a combinatorial\nonline algorithm that schedules stochastic, nonpreemptive jobs on unrelated\nmachines to minimize the expected total weighted completion time. Prior work on\nunrelated machine scheduling with stochastic jobs was restricted to the offline\ncase, and required sophisticated linear or convex programming relaxations for\nthe assignment of jobs to machines. The algorithm introduced in this paper is\nbased on a purely combinatorial assignment of jobs to machines, hence it also\nworks online. The performance bounds are of the same order of magnitude as\nthose of earlier work, and depend linearly on an upper bound $\\Delta$ on the\nsquared coefficient of variation of the jobs' processing times. They are\n$4+2\\Delta$ when there are no release dates, and $12+6\\Delta$ when jobs are\nreleased over time. For the special case of deterministic processing times,\nwithout and with release times, this paper shows that the same combinatorial\ngreedy algorithm has a competitive ratio of 4 and 6, respectively. As to the\ntechnical contribution, the paper shows for the first time how dual fitting\ntechniques can be used for stochastic and nonpreemptive scheduling problems.\n",
"title": "Greed Works - Online Algorithms For Unrelated Machine Stochastic Scheduling"
}
| null | null | null | null | true | null |
12015
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we introduce a new combinatorial curvature on triangulated\nsurfaces with inversive distance circle packing metrics. Then we prove that\nthis combinatorial curvature has global rigidity. To study the Yamabe problem\nof the new curvature, we introduce a combinatorial Ricci flow, along which the\ncurvature evolves almost in the same way as that of scalar curvature along the\nsurface Ricci flow obtained by Hamilton \\cite{Ham1}. Then we study the long\ntime behavior of the combinatorial Ricci flow and obtain that the existence of\na constant curvature metric is equivalent to the convergence of the flow on\ntriangulated surfaces with nonpositive Euler number. We further generalize the\ncombinatorial curvature to $\\alpha$-curvature and prove that it is also\nglobally rigid, which is in fact a generalized Bower-Stephenson conjecture\n\\cite{BS}. We also use the combinatorial Ricci flow to study the corresponding\n$\\alpha$-Yamabe problem.\n",
"title": "On a combinatorial curvature for surfaces with inversive distance circle packing metrics"
}
| null | null | null | null | true | null |
12016
| null |
Default
| null | null |
null |
{
"abstract": " Connectionist temporal classification (CTC) is widely used for maximum\nlikelihood learning in end-to-end speech recognition models. However, there is\nusually a disparity between the negative maximum likelihood and the performance\nmetric used in speech recognition, e.g., word error rate (WER). This results in\na mismatch between the objective function and metric during training. We show\nthat the above problem can be mitigated by jointly training with maximum\nlikelihood and policy gradient. In particular, with policy learning we are able\nto directly optimize on the (otherwise non-differentiable) performance metric.\nWe show that joint training improves relative performance by 4% to 13% for our\nend-to-end model as compared to the same model learned through maximum\nlikelihood. The model achieves 5.53% WER on Wall Street Journal dataset, and\n5.42% and 14.70% on Librispeech test-clean and test-other set, respectively.\n",
"title": "Improving End-to-End Speech Recognition with Policy Learning"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
12017
| null |
Validated
| null | null |
null |
{
"abstract": " Buhrman showed that an efficient communication protocol implies a reliable\nXOR game protocol. This idea rederives Linial and Shraibman's lower bounds of\ncommunication complexity, which was derived by using factorization norms, with\nworse constant factor in much more intuitive way. In this work, we improve and\ngeneralize Buhrman's idea, and obtain a class of lower bounds for classical\ncommunication complexity including an exact Linial and Shraibman's lower bound\nas a special case. In the proof, we explicitly construct a protocol for XOR\ngame from a classical communication protocol by using a concept of nonlocal\nboxes and Paw{\\l}owski et al.'s elegant protocol, which was used for showing\nthe violation of information causality in superquantum theories.\n",
"title": "Better Protocol for XOR Game using Communication Protocol and Nonlocal Boxes"
}
| null | null | null | null | true | null |
12018
| null |
Default
| null | null |
null |
{
"abstract": " We propose a novel diminishing learning rate scheme, coined\nDecreasing-Trend-Nature (DTN), which allows us to prove fast convergence of the\nStochastic Gradient Descent (SGD) algorithm to a first-order stationary point\nfor smooth general convex and some class of nonconvex including neural network\napplications for classification problems. We are the first to prove that SGD\nwith diminishing learning rate achieves a convergence rate of\n$\\mathcal{O}(1/t)$ for these problems. Our theory applies to neural network\napplications for classification problems in a straightforward way.\n",
"title": "DTN: A Learning Rate Scheme with Convergence Rate of $\\mathcal{O}(1/t)$ for SGD"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
12019
| null |
Validated
| null | null |
null |
{
"abstract": " The Wasserstein metric is an important measure of distance between\nprobability distributions, with applications in machine learning, statistics,\nprobability theory, and data analysis. This paper provides upper and lower\nbounds on statistical minimax rates for the problem of estimating a probability\ndistribution under Wasserstein loss, using only metric properties, such as\ncovering and packing numbers, of the sample space, and weak moment assumptions\non the probability distributions.\n",
"title": "Minimax Distribution Estimation in Wasserstein Distance"
}
| null | null | null | null | true | null |
12020
| null |
Default
| null | null |
null |
{
"abstract": " It is shown that the unit ball in ${\\mathbb C}^n$ is the only complex\nmanifold that can universally cover both Stein and non-Stein strictly\npseudoconvex domains.\n",
"title": "Uniformization and Steinness"
}
| null | null | null | null | true | null |
12021
| null |
Default
| null | null |
null |
{
"abstract": " In this article, recent progress on ML-randomness with respect to conditional\nprobabilities is reviewed. In particular a new result of conditional randomness\nwith respect to mutually singular probabilities are shown, which is a\ngeneralization of Hanssen's result (2010) for Bernoulli processes.\n",
"title": "Recent progress on conditional randomness"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
12022
| null |
Validated
| null | null |
null |
{
"abstract": " With red supergiants (RSGs) predicted to end their lives as Type IIP core\ncollapse supernova (CCSN), their behaviour before explosion needs to be fully\nunderstood. Mass loss rates govern RSG evolution towards SN and have strong\nimplications on the appearance of the resulting explosion. To study how the\nmass-loss rates change with the evolution of the star, we have measured the\namount of circumstellar material around 19 RSGs in a coeval cluster. Our study\nhas shown that mass loss rates ramp up throughout the lifetime of an RSG, with\nmore evolved stars having mass loss rates a factor of 40 higher than early\nstage RSGs. Interestingly, we have also found evidence for an increase in\ncircumstellar extinction throughout the RSG lifetime, meaning the most evolved\nstars are most severely affected. We find that, were the most evolved RSGs in\nNGC2100 to go SN, this extra extinction would cause the progenitor's initial\nmass to be underestimated by up to 9M$_\\odot$.\n",
"title": "The evolution of red supergiants to supernovae"
}
| null | null |
[
"Physics"
] | null | true | null |
12023
| null |
Validated
| null | null |
null |
{
"abstract": " The symmetry algebra of the real elliptic Liouville equation is an\ninfinite-dimensional loop algebra with the simple Lie algebra $o(3,1)$ as its\nmaximal finite-dimensional subalgebra. The entire algebra generates the\nconformal group of the Euclidean plane $E_2$. This infinite-dimensional algebra\ndistinguishes the elliptic Liouville equation from the hyperbolic one with its\nsymmetry algebra that is the direct sum of two Virasoro algebras. Following a\ndiscretisation procedure developed earlier, we present a difference scheme that\nis invariant under the group $O(3,1)$ and has the elliptic Liouville equation\nin polar coordinates as its continuous limit. The lattice is a solution of an\nequation invariant under $O(3,1)$ and is itself invariant under a subgroup of\n$O(3,1)$, namely the $O(2)$ rotations of the Euclidean plane.\n",
"title": "Conformally invariant elliptic Liouville equation and its symmetry preserving discretization"
}
| null | null | null | null | true | null |
12024
| null |
Default
| null | null |
null |
{
"abstract": " Due to its accuracy and generality, Monte Carlo radiative transfer (MCRT) has\nemerged as the prevalent method for Ly$\\alpha$ radiative transfer in arbitrary\ngeometries. The standard MCRT encounters a significant efficiency barrier in\nthe high optical depth, diffusion regime. Multiple acceleration schemes have\nbeen developed to improve the efficiency of MCRT but the noise from photon\npacket discretization remains a challenge. The discrete diffusion Monte Carlo\n(DDMC) scheme has been successfully applied in state-of-the-art radiation\nhydrodynamics (RHD) simulations. Still, the established framework is not\noptimal for resonant line transfer. Inspired by the DDMC paradigm, we present a\nnovel extension to resonant DDMC (rDDMC) in which diffusion in space and\nfrequency are treated on equal footing. We explore the robustness of our new\nmethod and demonstrate a level of performance that justifies incorporating the\nmethod into existing Ly$\\alpha$ codes. We present computational speedups of\n$\\sim 10^2$-$10^6$ relative to contemporary MCRT implementations with schemes\nthat skip scattering in the core of the line profile. This is because the rDDMC\nruntime scales with the spatial and frequency resolution rather than the number\nof scatterings - the latter is typically $\\propto \\tau_0$ for static media, or\n$\\propto (a \\tau_0)^{2/3}$ with core-skipping. We anticipate new frontiers in\nwhich on-the-fly Ly$\\alpha$ radiative transfer calculations are feasible in 3D\nRHD. More generally, rDDMC is transferable to any computationally demanding\nproblem amenable to a Fokker-Planck approximation of frequency redistribution.\n",
"title": "Discrete diffusion Lyman-alpha radiative transfer"
}
| null | null | null | null | true | null |
12025
| null |
Default
| null | null |
null |
{
"abstract": " We present a theoretical study of the finite-temperature Kosterlitz-Thouless\n(KT) and vortex-antivortex lattice (VAL) melting transitions in two-dimensional\nFermi gases with $p$- or $d$-wave pairing. For both pairings, when the\ninteraction is tuned from weak to strong attractions, we observe a quantum\nphase transition from the Bardeen-Cooper-Schrieffer (BCS) superfluidity to the\nBose-Einstein condensation (BEC) of difermions. The KT and VAL transition\ntemperatures increase during this BCS-BEC transition and approach constant\nvalues in the deep BEC region. The BCS-BEC transition is characterized by the\nnon-analyticities of the chemical potential, the superfluid order parameter,\nand the sound velocities as functions of the interaction strength at both zero\nand finite temperatures; however, the temperature effect tends to weaken the\nnon-analyticities comparing to the zero temperature case. The effect of\nmismatched Fermi surfaces on the $d$-wave pairing is also studied.\n",
"title": "Kosterlitz-Thouless transition and vortex-antivortex lattice melting in two-dimensional Fermi gases with $p$- or $d$-wave pairing"
}
| null | null | null | null | true | null |
12026
| null |
Default
| null | null |
null |
{
"abstract": " The chiral optical Tamm state (COTS) is a special localized state at the\ninterface of a handedness-preserving mirror and a structurally chiral medium\nsuch as a cholesteric liquid crystal or a chiral sculptured thin film. The\nspectral behavior of COTS, observed as reflection resonances, is described by\nthe temporal coupled-mode theory. Mode coupling is different for two circular\nlight polarizations because COTS has a helix structure replicating that of the\ncholesteric. The mode coupling for co-handed circularly polarized light\nexponentially attenuates with the cholesteric layer thickness since the COTS\nfrequency falls into the stop band. Cross-handed circularly polarized light\nfreely goes through the cholesteric layer and can excite COTS when reflected\nfrom the handedness-preserving mirror. The coupling in this case is\nproportional to anisotropy of the cholesteric and theoretically it is only\nanisotropy of magnetic permittivity that can ultimately cancel this coupling.\nThese two couplings being equal results in a polarization crossover (the\nKopp--Genack effect) for which a linear polarization is optimal to excite COTS.\nThe corresponding cholesteric thickness and scattering matrix for COTS are\ngenerally described by simple expressions.\n",
"title": "Chiral Optical Tamm States: Temporal Coupled-Mode Theory"
}
| null | null | null | null | true | null |
12027
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, our aim is to show some mean value inequalities for the\nFox-Wright functions, such as Turán--type inequalities, Lazarević and\nWilker--type inequalities. As applications we derive some new type inequalities\nfor hypergeometric functions and the four--parametric Mittag--Leffler\nfunctions. Furthermore, we prove monotonicity of ratios for sections of series\nof Fox-Wright functions, the results is also closely connected with\nTurán--type inequalities. Moreover, some other type inequalities are also\npresented. At the end of the paper, some problems stated, which may be of\ninterest for further research.\n",
"title": "Functional inequalities for Fox-Wright functions"
}
| null | null | null | null | true | null |
12028
| null |
Default
| null | null |
null |
{
"abstract": " Research Objects (ROs) are semantically enhanced aggregations of resources\nassociated to scientific experiments, such as data, provenance of these data,\nthe scientific workflow used to run the experiment, intermediate results, logs\nand the interpretation of the results. As the number of ROs increases, it is\nbecoming difficult to find ROs to be used, reused or re-purposed. New search\nand retrieval techniques are required to find the most appropriate ROs for a\ngiven researcher, paying attention to provide an intuitive user interface. In\nthis paper we show CollabSpheres, a user interface that provides a new visual\nmetaphor to find ROs by means of a recommendation system that takes advantage\nof the social aspects of ROs. The experimental evaluation of this tool shows\nthat users perceive high values of usability, user satisfaction, usefulness and\nease of use. From the analysis of these results we argue that users perceive\nthe simplicity, intuitiveness and cleanness of this tool, as well as this tool\nincreases collaboration and reuse of research objects.\n",
"title": "Collaboration Spheres: a Visual Metaphor to Share and Reuse Research Objects"
}
| null | null | null | null | true | null |
12029
| null |
Default
| null | null |
null |
{
"abstract": " This work introduces our approach to the flat and textureless object grasping\nproblem. In particular, we address the tableware and cutlery manipulation\nproblem where a service robot has to clean up a table. Our solution integrates\ncolour and 2D and 3D geometry information to describe objects, and this\ninformation is given to the robot action planner to find the best grasping\ntrajectory depending on the object class. Furthermore, we use visual feedback\nas a verification step to determine if the grasping process has successfully\noccurred. We evaluate our approach in both an open and a standard service robot\nplatform following the RoboCup@Home international tournament regulations.\n",
"title": "Intelligent flat-and-textureless object manipulation in Service Robots"
}
| null | null |
[
"Computer Science"
] | null | true | null |
12030
| null |
Validated
| null | null |
null |
{
"abstract": " Pair creation on the cosmic infrared background and subsequent\ninverse-Compton scattering on the CMB potentially reprocesses the TeV emission\nof blazars into faint GeV halos with structures sensitive to intergalactic\nmagnetic fields (IGMF). We attempt to detect such halos exploiting their highly\nanisotropic shape. Their persistent nondetection excludes at greater than\n$3.9\\sigma$ an IGMF with correlation lengths >100 Mpc and current-day strengths\nin the range $10^{-16}$ to $10^{-15}$ G, and at 2 sigma from $10^{-17}$ to\n$10^{-14}$ G, covering the range implied by gamma-ray spectra of nearby TeV\nemitters. Alternatively, plasma processes could pre-empt the inverse-Compton\ncascade.\n",
"title": "Constraints on the Intergalactic Magnetic Field from Bow Ties in the Gamma-ray Sky"
}
| null | null | null | null | true | null |
12031
| null |
Default
| null | null |
null |
{
"abstract": " Knowledge Transfer (KT) techniques tackle the problem of transferring the\nknowledge from a large and complex neural network into a smaller and faster\none. However, existing KT methods are tailored towards classification tasks and\nthey cannot be used efficiently for other representation learning tasks. In\nthis paper a novel knowledge transfer technique, that is capable of training a\nstudent model that maintains the same amount of mutual information between the\nlearned representation and a set of (possible unknown) labels as the teacher\nmodel, is proposed. Apart from outperforming existing KT techniques, the\nproposed method allows for overcoming several limitations of existing methods\nproviding new insight into KT as well as novel KT applications, ranging from\nknowledge transfer from handcrafted feature extractors to {cross-modal} KT from\nthe textual modality into the representation extracted from the visual modality\nof the data.\n",
"title": "Learning Deep Representations with Probabilistic Knowledge Transfer"
}
| null | null | null | null | true | null |
12032
| null |
Default
| null | null |
null |
{
"abstract": " This article is an attempt to generalize Riemann's bilinear relations on\ncompact Riemann surface of genus at least 2, which may lead to new structures\nin the theory of hyperbolic Riemann surfaces. No significant result is\nobtained, the article serves to bring the readers' attention to the observation\nmade by [Bol-1949], and some easy consequences.\n",
"title": "Pluricanonical Periods over Compact Riemann Surfaces of Genus at least 2"
}
| null | null | null | null | true | null |
12033
| null |
Default
| null | null |
null |
{
"abstract": " When applied to training deep neural networks, stochastic gradient descent\n(SGD) often incurs steady progression phases, interrupted by catastrophic\nepisodes in which loss and gradient norm explode. A possible mitigation of such\nevents is to slow down the learning process. This paper presents a novel\napproach to control the SGD learning rate, that uses two statistical tests. The\nfirst one, aimed at fast learning, compares the momentum of the normalized\ngradient vectors to that of random unit vectors and accordingly gracefully\nincreases or decreases the learning rate. The second one is a change point\ndetection test, aimed at the detection of catastrophic learning episodes; upon\nits triggering the learning rate is instantly halved. Both abilities of\nspeeding up and slowing down the learning rate allows the proposed approach,\ncalled SALeRA, to learn as fast as possible but not faster. Experiments on\nstandard benchmarks show that SALeRA performs well in practice, and compares\nfavorably to the state of the art.\n",
"title": "Stochastic Gradient Descent: Going As Fast As Possible But Not Faster"
}
| null | null | null | null | true | null |
12034
| null |
Default
| null | null |
null |
{
"abstract": " This paper studies the detection of bird calls in audio segments using\nstacked convolutional and recurrent neural networks. Data augmentation by\nblocks mixing and domain adaptation using a novel method of test mixing are\nproposed and evaluated in regard to making the method robust to unseen data.\nThe contributions of two kinds of acoustic features (dominant frequency and log\nmel-band energy) and their combinations are studied in the context of bird\naudio detection. Our best achieved AUC measure on five cross-validations of the\ndevelopment data is 95.5% and 88.1% on the unseen evaluation data.\n",
"title": "Stacked Convolutional and Recurrent Neural Networks for Bird Audio Detection"
}
| null | null | null | null | true | null |
12035
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we consider a Bayesian framework for making inferences about\ndynamical systems from ergodic observations. The proposed Bayesian procedure is\nbased on the Gibbs posterior, a decision theoretic generalization of standard\nBayesian inference. We place a prior over a model class consisting of a\nparametrized family of Gibbs measures on a mixing shift of finite type. This\nmodel class generalizes (hidden) Markov chain models by allowing for long range\ndependencies, including Markov chains of arbitrarily large orders. We\ncharacterize the asymptotic behavior of the Gibbs posterior distribution on the\nparameter space as the number of observations tends to infinity. In particular,\nwe define a limiting variational problem over the space of joinings of the\nmodel system with the observed system, and we show that the Gibbs posterior\ndistributions concentrate around the solution set of this variational problem.\nIn the case of properly specified models our convergence results may be used to\nestablish posterior consistency. This work establishes tight connections\nbetween Gibbs posterior inference and the thermodynamic formalism, which may\ninspire new proof techniques in the study of Bayesian posterior consistency for\ndependent processes.\n",
"title": "Gibbs posterior convergence and the thermodynamic formalism"
}
| null | null | null | null | true | null |
12036
| null |
Default
| null | null |
null |
{
"abstract": " We analyze the emission spectrum of the hot Jupiter WASP-12b using our\nHELIOS-R retrieval code and HELIOS-K opacity calculator. When interpreting\nHubble and Spitzer data, the retrieval outcomes are found to be\nprior-dominated. When the prior distributions of the molecular abundances are\nassumed to be log-uniform, the volume mixing ratio of HCN is found to be\nimplausibly high. A VULCAN chemical kinetics model of WASP-12b suggests that\nchemical equilibrium is a reasonable assumption even when atmospheric mixing is\nimplausibly rigorous. Guided by (exo)planet formation theory, we set Gaussian\npriors on the elemental abundances of carbon, oxygen and nitrogen with the\nGaussian peaks being centered on the measured C/H, O/H and N/H values of the\nstar. By enforcing chemical equilibrium, we find substellar O/H and stellar to\nslightly superstellar C/H for the dayside atmosphere of WASP-12b. The\nsuperstellar carbon-to-oxygen ratio is just above unity, regardless of whether\nclouds are included in the retrieval analysis, consistent with Madhusudhan et\nal. (2011). Furthermore, whether a temperature inversion exists in the\natmosphere depends on one's assumption for the Gaussian width of the priors.\nOur retrieved posterior distributions are consistent with the formation of\nWASP-12b in a solar-composition protoplanetary disk, beyond the water iceline,\nvia gravitational instability or pebble accretion (without core erosion) and\nmigration inwards to its present orbital location via a disk-free mechanism,\nand are inconsistent with both in-situ formation and core accretion with disk\nmigration, as predicted by Madhusudhan et al. (2017). We predict that the\ninterpretation of James Webb Space Telescope WASP-12b data will not be\nprior-dominated.\n",
"title": "Retrieval Analysis of the Emission Spectrum of WASP-12b: Sensitivity of Outcomes to Prior Assumptions and Implications for Formation History"
}
| null | null | null | null | true | null |
12037
| null |
Default
| null | null |
null |
{
"abstract": " We propose an exploration method that incorporates look-ahead search over\nbasic learnt skills and their dynamics, and use it for reinforcement learning\n(RL) of manipulation policies . Our skills are multi-goal policies learned in\nisolation in simpler environments using existing multigoal RL formulations,\nanalogous to options or macroactions. Coarse skill dynamics, i.e., the state\ntransition caused by a (complete) skill execution, are learnt and are unrolled\nforward during lookahead search. Policy search benefits from temporal\nabstraction during exploration, though itself operates over low-level primitive\nactions, and thus the resulting policies does not suffer from suboptimality and\ninflexibility caused by coarse skill chaining. We show that the proposed\nexploration strategy results in effective learning of complex manipulation\npolicies faster than current state-of-the-art RL methods, and converges to\nbetter policies than methods that use options or parametrized skills as\nbuilding blocks of the policy itself, as opposed to guiding exploration. We\nshow that the proposed exploration strategy results in effective learning of\ncomplex manipulation policies faster than current state-of-the-art RL methods,\nand converges to better policies than methods that use options or parameterized\nskills as building blocks of the policy itself, as opposed to guiding\nexploration.\n",
"title": "Model Learning for Look-ahead Exploration in Continuous Control"
}
| null | null | null | null | true | null |
12038
| null |
Default
| null | null |
null |
{
"abstract": " Single-image-based view generation (SIVG) is important for producing 3D\nstereoscopic content. Here, handling different spatial resolutions as input and\noptimizing both reconstruction accuracy and processing speed is desirable.\nLatest approaches are based on convolutional neural network (CNN), and they\ngenerate promising results. However, their use of fully connected layers as\nwell as pre-trained VGG forces a compromise between reconstruction accuracy and\nprocessing speed. In addition, this approach is limited to the use of a\nspecific spatial resolution. To remedy these problems, we propose exploiting\nfully convolutional networks (FCN) for SIVG. We present two FCN architectures\nfor SIVG. The first one is based on combination of an FCN and a view-rendering\nnetwork called DeepView$_{ren}$. The second one consists of decoupled networks\nfor luminance and chrominance signals, denoted by DeepView$_{dec}$. To train\nour solutions we present a large dataset of 2M stereoscopic images. Results\nshow that both of our architectures improve accuracy and speed over the state\nof the art. DeepView$_{ren}$ generates competitive accuracy to the state of the\nart, however, with the fastest processing speed of all. That is x5 times faster\nspeed and x24 times lower memory consumption compared to the state of the art.\nDeepView$_{dec}$ has much higher accuracy, but with x2.5 times faster speed and\nx12 times lower memory consumption. We evaluated our approach with both\nobjective and subjective studies.\n",
"title": "Efficient and Scalable View Generation from a Single Image using Fully Convolutional Networks"
}
| null | null | null | null | true | null |
12039
| null |
Default
| null | null |
null |
{
"abstract": " With applications to many disciplines, the traveling salesman problem (TSP)\nis a classical computer science optimization problem with applications to\nindustrial engineering, theoretical computer science, bioinformatics, and\nseveral other disciplines. In recent years, there have been a plethora of novel\napproaches for approximate solutions ranging from simplistic greedy to\ncooperative distributed algorithms derived from artificial intelligence. In\nthis paper, we perform an evaluation and analysis of cornerstone algorithms for\nthe Euclidean TSP. We evaluate greedy, 2-opt, and genetic algorithms. We use\nseveral datasets as input for the algorithms including a small dataset, a\nmediumsized dataset representing cities in the United States, and a synthetic\ndataset consisting of 200 cities to test algorithm scalability. We discover\nthat the greedy and 2-opt algorithms efficiently calculate solutions for\nsmaller datasets. Genetic algorithm has the best performance for optimality for\nmedium to large datasets, but generally have longer runtime. Our\nimplementations is public available.\n",
"title": "An Empirical Analysis of Approximation Algorithms for the Euclidean Traveling Salesman Problem"
}
| null | null | null | null | true | null |
12040
| null |
Default
| null | null |
null |
{
"abstract": " The concept of dynamical compensation has been recently introduced to\ndescribe the ability of a biological system to keep its output dynamics\nunchanged in the face of varying parameters. Here we show that, according to\nits original definition, dynamical compensation is equivalent to lack of\nstructural identifiability. This is relevant if model parameters need to be\nestimated, which is often the case in biological modelling. This realization\nprompts us to warn that care should we taken when using an unidentifiable model\nto extract biological insight: the estimated values of structurally\nunidentifiable parameters are meaningless, and model predictions about\nunmeasured state variables can be wrong. Taking this into account, we explore\nalternative definitions of dynamical compensation that do not necessarily imply\nstructural unidentifiability. Accordingly, we show different ways in which a\nmodel can be made identifiable while exhibiting dynamical compensation. Our\nanalyses enable the use of the new concept of dynamical compensation in the\ncontext of parameter identification, and reconcile it with the desirable\nproperty of structural identifiability.\n",
"title": "Dynamical compensation and structural identifiability: analysis, implications, and reconciliation"
}
| null | null | null | null | true | null |
12041
| null |
Default
| null | null |
null |
{
"abstract": " A 2.1 MeV, 10 mA CW RFQ has been installed and commissioned at Fermilab's\ntest accelerator known as PIP-II Injector Test. This report describes the\nmeasurements of the beam properties after acceleration in the RFQ, including\nthe energy and emittance.\n",
"title": "Characterization of the beam from the RFQ of the PIP-II Injector Test"
}
| null | null | null | null | true | null |
12042
| null |
Default
| null | null |
null |
{
"abstract": " Softmax is a standard final layer used in Neural Nets (NNs) to summarize\ninformation encoded in the trained NN and return a prediction. However, Softmax\nleverages only a subset of the class-specific structure encoded in the trained\nmodel and ignores potentially valuable information: During training, models\nencode an array $D$ of class response distributions, where $D_{ij}$ is the\ndistribution of the $j^{th}$ pre-Softmax readout neuron's responses to the\n$i^{th}$ class. Given a test sample, Softmax implicitly uses only the row of\nthis array $D$ that corresponds to the readout neurons' responses to the\nsample's true class. Leveraging more of this array $D$ can improve classifier\naccuracy, because the likelihoods of two competing classes can be encoded in\nother rows of $D$.\nTo explore this potential resource, we develop a hybrid classifier\n(Softmax-Pooling Hybrid, $SPH$) that uses Softmax on high-scoring samples, but\non low-scoring samples uses a log-likelihood method that pools the information\nfrom the full array $D$. We apply $SPH$ to models trained on a vectorized MNIST\ndataset to varying levels of accuracy. $SPH$ replaces only the final Softmax\nlayer in the trained NN, at test time only. All training is the same as for\nSoftmax. Because the pooling classifier performs better than Softmax on\nlow-scoring samples, $SPH$ reduces test set error by 6% to 23%, using the exact\nsame trained model, whatever the baseline Softmax accuracy. This reduction in\nerror reflects hidden capacity of the trained NN that is left unused by\nSoftmax.\n",
"title": "Money on the Table: Statistical information ignored by Softmax can improve classifier accuracy"
}
| null | null | null | null | true | null |
12043
| null |
Default
| null | null |
null |
{
"abstract": " In this note, we recall Kummer's Fourier series expansion of the 1-periodic\nfunction that coincides with the logarithm of the Gamma function on the unit\ninterval $(0,1)$, and we use it to find closed forms for some numerical series\nrelated to the generalized Stieltjes constants, and some integrals involving\nthe function $x\\mapsto \\ln \\ln(1/x)$.\n",
"title": "Generalized Stieltjes constants and integrals involving the log-log function: Kummer's Theorem in action"
}
| null | null | null | null | true | null |
12044
| null |
Default
| null | null |
null |
{
"abstract": " In the study of extensions of polytopes of combinatorial optimization\nproblems, a notorious open question is that for the size of the smallest\nextended formulation of the Minimum Spanning Tree problem on a complete graph\nwith $n$ nodes. The best known lower bound is the trival (dimension) bound,\n$\\Omega(n^2)$, the best known upper bound is the extended formulation by Wong\n(1980) of size $O(n^3)$ (also Martin, 1991).\nIn this note we give a nondeterministic communication protocol with cost\n$\\log_2(n^2\\log n)+O(1)$ for the support of the spanning tree slack matrix.\nThis means that the combinatorial lower bounds can improve the trivial lower\nbound only by a factor of (at most) $O(\\log n)$.\n",
"title": "On the Combinatorial Lower Bound for the Extension Complexity of the Spanning Tree Polytope"
}
| null | null | null | null | true | null |
12045
| null |
Default
| null | null |
null |
{
"abstract": " DEVS is a popular formalism for modelling complex dynamic systems using a\ndiscrete-event abstraction. At this abstraction level, a timed sequence\nofpertinent \"events\" input to a system (or internal, in the case of timeouts)\ncause instantaneous changes to the state of the system. Between events, the\nstate does not change, resulting in a a piecewise constant state trajectory.\nMain advantages of DEVS are its rigorous formal definition, and its support for\nmodular composition.\nThis chapter introduces the Classic DEVS formalism in a bottom-up fashion,\nusing a simple traffic light example. The syntax and operational semantics of\nAtomic (i.e., non-hierarchical) models are intruced first. The semantics of\nCoupled (hierarchical) models is then given by translation into Atomic DEVS\nmodels. As this formal \"flattening\" is not efficient, a modular abstract\nsimulator which operates directly on the coupled model is also presented. This\nis the common basis for subsequent efficient implementations. We continue to\nactual applications of DEVS modelling and simulation, as seen in performance\nanalysis for queueing systems. Finally, we present some of the shortcomings in\nthe Classic DEVS formalism, and show solutions to them in the form of variants\nof the original formalism.\n",
"title": "An Introduction to Classic DEVS"
}
| null | null | null | null | true | null |
12046
| null |
Default
| null | null |
null |
{
"abstract": " Several active areas of research in novel energy storage technologies,\nincluding three-dimensional solid state batteries and passivation coatings for\nreactive battery electrode components, require conformal solid state\nelectrolytes. We describe an atomic layer deposition (ALD) process for a member\nof the lithium phosphorus oxynitride (LiPON) family, which is employed as a\nthin film lithium-conducting solid electrolyte. The reaction between lithium\ntert-butoxide (LiO$^t$Bu) and diethyl phosphoramidate (DEPA) produces\nconformal, ionically conductive thin films with a stoichiometry close to\nLi$_2$PO$_2$N between 250 and 300$^\\circ$C. The P/N ratio of the films is\nalways 1, indicative of a particular polymorph of LiPON which closely resembles\na polyphosphazene. Films grown at 300$^\\circ$C have an ionic conductivity of\n$6.51\\:(\\pm0.36)\\times10^{-7}$ S/cm at 35$^\\circ$C, and are functionally\nelectrochemically stable in the window from 0 to 5.3V vs. Li/Li$^+$. We\ndemonstrate the viability of the ALD-grown electrolyte by integrating it into\nfull solid state batteries, including thin film devices using LiCoO$_2$ as the\ncathode and Si as the anode operating at up to 1 mA/cm$^2$. The high quality of\nthe ALD growth process allows pinhole-free deposition even on rough crystalline\nsurfaces, and we demonstrate the fabrication and operation of thin film\nbatteries with the thinnest (<100nm) solid state electrolytes yet reported.\nFinally, we show an additional application of the moderate-temperature ALD\nprocess by demonstrating a flexible solid state battery fabricated on a polymer\nsubstrate.\n",
"title": "Nanoscale Solid State Batteries Enabled By Thermal Atomic Layer Deposition of a Lithium Polyphosphazene Solid State Electrolyte"
}
| null | null | null | null | true | null |
12047
| null |
Default
| null | null |
null |
{
"abstract": " The Machine Recognition of Crystallization Outcomes (MARCO) initiative has\nassembled roughly half a million annotated images of macromolecular\ncrystallization experiments from various sources and setups. Here,\nstate-of-the-art machine learning algorithms are trained and tested on\ndifferent parts of this data set. We find that more than 94% of the test images\ncan be correctly labeled, irrespective of their experimental origin. Because\ncrystal recognition is key to high-density screening and the systematic\nanalysis of crystallization experiments, this approach opens the door to both\nindustrial and fundamental research applications.\n",
"title": "Classification of crystallization outcomes using deep convolutional neural networks"
}
| null | null | null | null | true | null |
12048
| null |
Default
| null | null |
null |
{
"abstract": " This research investigated the potential for improving Peer-to-Peer (P2P)\ncredit scoring by using \"private information\" about communications and travels\nof borrowers. We found that P2P borrowers' ego networks exhibit scale-free\nbehavior driven by underlying preferential attachment mechanisms that connect\nborrowers in a fashion that can be used to predict loan profitability. The\nprojection of these private networks onto networks of mobile phone\ncommunication and geographical locations from mobile phone GPS potentially give\nloan providers access to private information through graph and location metrics\nwhich we used to predict loan profitability. Graph topology was found to be an\nimportant predictor of loan profitability, explaining over 5.5% of variability.\nNetworks of borrower location information explain an additional 19% of the\nprofitability. Machine learning algorithms were applied to the data set\npreviously analyzed to develop the predictive model and resulted in a 4%\nreduction in mean squared error.\n",
"title": "Private Information, Credit Risk and Graph Structure in P2P Lending Networks"
}
| null | null | null | null | true | null |
12049
| null |
Default
| null | null |
null |
{
"abstract": " Magnetotransport measurements in combination with molecular dynamics (MD)\nsimulations on two-dimensional disordered Lorentz gases in the classical regime\nare reported. In quantitative agreement between experiment and simulation, the\nmagnetoconductivity displays a pronounced peak as a function of perpendicular\nmagnetic field $B$ which cannot be explained in the framework of existing\nkinetic theories. We show that this peak is linked to the onset of a directed\nmotion of the electrons along the contour of the disordered obstacle matrix\nwhen the cyclotron radius becomes smaller than the size of the obstacles. This\ndirected motion leads to transient superdiffusive motion and strong scaling\ncorrections in the vicinity of the insulator-to-conductor transitions of the\nLorentz gas.\n",
"title": "Nonmonotonous classical magneto-conductivity of a two-dimensional electron gas in a disordered array of obstacles"
}
| null | null | null | null | true | null |
12050
| null |
Default
| null | null |
null |
{
"abstract": " Science education is a crucial issue with long-term impacts for Europe as the\nlow enrolment rates in the STEM-fields, including (natural) science,\ntechnology, engineering and mathematics, will lead to a workforce problem in\nresearch and development. In order to address this challenge, the EU-funded\nresearch project SciChallenge (project.scichallenge.eu) aims to find a new way\nfor getting young people more interested in STEM. For this purpose, the project\ndeveloped and implemented a social-media-based STEM-contest for young people,\nwhich aims at increasing the attractiveness of science education and careers\namong young people. In the first two parts, the paper reflects on the problem,\nintroduces the project and highlights the main steps of the preparation of the\ncontest. The third section of the paper presents the idea, design and\nimplementation of the digital contest platform (www.scichallenge.eu), which\nserves as the core of the challenge. The fourth part of the paper will provide\na status update on the contest pilot. It will provide a summary of the\nexperiences that the consortium made with this novel approach as well as the\nmain obstacles that the consortium was facing. The paper will conclude with a\npreliminary reflection on the question if such an approach can help to increase\nthe interest of young people in STEM-education and careers.\n",
"title": "Pushing STEM-education through a social-media-based contest format - experiences and lessons-learned from the H2020-project SciChallenge"
}
| null | null | null | null | true | null |
12051
| null |
Default
| null | null |
null |
{
"abstract": " We investigated the effect of out-of-plane crumpling on the mechanical\nresponse of graphene membranes. In our experiments, stress was applied to\ngraphene membranes using pressurized gas while the strain state was monitored\nthrough two complementary techniques: interferometric profilometry and Raman\nspectroscopy. By comparing the data obtained through these two techniques, we\ndetermined the geometric hidden area which quantifies the crumpling strength.\nWhile the devices with hidden area $\\sim0~\\%$ obeyed linear mechanics with\nbiaxial stiffness $428\\pm10$ N/m, specimens with hidden area in the range\n$0.5-1.0~\\%$ were found to obey an anomalous Hooke's law with an exponent\n$\\sim0.1$.\n",
"title": "Hidden area and mechanical nonlinearities in freestanding graphene"
}
| null | null | null | null | true | null |
12052
| null |
Default
| null | null |
null |
{
"abstract": " Solutions of partial differential equations (PDEs) on manifolds have provided\nimportant applications in different fields in science and engineering. Existing\nmethods are majorly based on discretization of manifolds as implicit functions,\ntriangle meshes, or point clouds, where the manifold structure is approximated\nby either zero level set of an implicit function or a set of points. In many\napplications, manifolds might be only provided as an inter-point distance\nmatrix with possible missing values. This paper discusses a framework to\ndiscretize PDEs on manifolds represented as incomplete inter-point distance\ninformation. Without conducting a time-consuming global coordinates\nreconstruction, we propose a more efficient strategy by discretizing\ndifferential operators only based on point-wisely local reconstruction. Our\nlocal reconstruction model is based on the recent advances of low-rank matrix\ncompletion theory, where only a very small random portion of distance\ninformation is required. This method enables us to conduct analyses of\nincomplete distance data using solutions of special designed PDEs such as the\nLaplace-Beltrami (LB) eigen-system. As an application, we demonstrate a new way\nof manifold reconstruction from an incomplete distance by stitching patches\nusing the spectrum of the LB operator. Intensive numerical experiments\ndemonstrate the effectiveness of the proposed methods.\n",
"title": "Solving Partial Differential Equations on Manifolds From Incomplete Inter-Point Distance"
}
| null | null | null | null | true | null |
12053
| null |
Default
| null | null |
null |
{
"abstract": " All four giant planets in the Solar System feature zonal flows on the order\nof 100 m/s in the cloud deck, and large-scale intrinsic magnetic fields on the\norder of 1 Gauss near the surface. The vertical structure of the zonal flows\nremains obscure. The end-member scenarios are shallow flows confined in the\nradiative atmosphere and deep flows throughout the entire planet. The\nelectrical conductivity increases rapidly yet smoothly as a function of depth\ninside Jupiter and Saturn. Deep zonal flows will inevitably interact with the\nmagnetic field, at depth with even modest electrical conductivity. Here we\ninvestigate the interaction between zonal flows and magnetic fields in the\nsemi-conducting region of giant planets. Employing mean-field electrodynamics,\nwe show that the interaction will generate detectable poloidal magnetic field\nperturbations spatially correlated with the deep zonal flows. Assuming the peak\namplitude of the dynamo alpha-effect to be 0.1 mm/s, deep zonal flows on the\norder of 0.1 - 1 m/s in the semi-conducting region of Jupiter and Saturn would\ngenerate poloidal magnetic perturbations on the order of 0.01% - 1% of the\nbackground dipole field. These poloidal perturbations should be detectable with\nthe in-situ magnetic field measurements from the Juno mission and the Cassini\nGrand Finale. This implies that magnetic field measurements can be employed to\nconstrain the properties of deep zonal flows in the semi-conducting region of\ngiant planets.\n",
"title": "Zonal Flow Magnetic Field Interaction in the Semi-Conducting Region of Giant Planets"
}
| null | null | null | null | true | null |
12054
| null |
Default
| null | null |
null |
{
"abstract": " Semantic segmentation, like other fields of computer vision, has seen a\nremarkable performance advance by the use of deep convolution neural networks.\nHowever, considering that neighboring pixels are heavily dependent on each\nother, both learning and testing of these methods have a lot of redundant\noperations. To resolve this problem, the proposed network is trained and tested\nwith only 0.37% of total pixels by superpixel-based sampling and largely\nreduced the complexity of upsampling calculation. The hypercolumn feature maps\nare constructed by pyramid module in combination with the convolution layers of\nthe base network. Since the proposed method uses a very small number of sampled\npixels, the end-to-end learning of the entire network is difficult with a\ncommon learning rate for all the layers. In order to resolve this problem, the\nlearning rate after sampling is controlled by statistical process control (SPC)\nof gradients in each layer. The proposed method performs better than or equal\nto the conventional methods that use much more samples on Pascal Context,\nSUN-RGBD dataset.\n",
"title": "Superpixel-based Semantic Segmentation Trained by Statistical Process Control"
}
| null | null | null | null | true | null |
12055
| null |
Default
| null | null |
null |
{
"abstract": " Photometric Stereo methods seek to reconstruct the 3d shape of an object from\nmotionless images obtained with varying illumination. Most existing methods\nsolve a restricted problem where the physical reflectance model, such as\nLambertian reflectance, is known in advance. In contrast, we do not restrict\nourselves to a specific reflectance model. Instead, we offer a method that\nworks on a wide variety of reflectances. Our approach uses a simple yet\nuncommonly used property of the problem - the sought after normals are points\non a unit hemisphere. We present a novel embedding method that maps pixels to\nnormals on the unit hemisphere. Our experiments demonstrate that this approach\noutperforms existing manifold learning methods for the task of hemisphere\nembedding. We further show successful reconstructions of objects from a wide\nvariety of reflectances including smooth, rough, diffuse and specular surfaces,\neven in the presence of significant attached shadows. Finally, we empirically\nprove that under these challenging settings we obtain more accurate shape\nreconstructions than existing methods.\n",
"title": "Photometric Stereo by Hemispherical Metric Embedding"
}
| null | null |
[
"Computer Science"
] | null | true | null |
12056
| null |
Validated
| null | null |
null |
{
"abstract": " We introduce a web of strongly correlated interacting 3+1D topological\nsuperconductors/insulators of 10 particular global symmetry groups of Cartan\nclasses, realizable in electronic condensed matter systems, and their new SU(N)\ngeneralizations. The symmetries include SU(N), SU(2), U(1), fermion parity,\ntime reversal and relate to each other through symmetry embeddings. We overview\nthe lattice Hamiltonian formalism. We complete the list of field theories of\nbulk symmetry-protected topological invariants (SPT invariants/partition\nfunctions that exhibit boundary 't Hooft anomalies) via cobordism calculations,\nmatching their full classification. We also present explicit 4-manifolds that\ndetect these SPTs. On the other hand, once we dynamically gauge part of their\nglobal symmetries, we arrive in various new phases of SU(N) Yang-Mills (YM)\ngauge theories, analogous to quantum spin liquids with emergent gauge fields.\nWe discuss how coupling YM theories to time reversal-SPTs affects the strongly\ncoupled theories at low energy. For example, we point out a possibility of\nhaving two deconfined gapless time-reversal symmetric SU(2) YM theories at\n$\\theta=\\pi$ as two distinct conformal field theories, which although are\nsecretly indistinguishable by correlators of local operators on orientable\nspacetimes nor by gapped SPT states, can be distinguished on non-orientable\nspacetimes or potentially by correlators of extended operators.\n",
"title": "Time Reversal, SU(N) Yang-Mills and Cobordisms: Interacting Topological Superconductors/Insulators and Quantum Spin Liquids in 3+1D"
}
| null | null | null | null | true | null |
12057
| null |
Default
| null | null |
null |
{
"abstract": " Rainfall ensemble forecasts have to be skillful for both low precipitation\nand extreme events. We present statistical post-processing methods based on\nQuantile Regression Forests (QRF) and Gradient Forests (GF) with a parametric\nextension for heavy-tailed distributions. Our goal is to improve ensemble\nquality for all types of precipitation events, heavy-tailed included, subject\nto a good overall performance. Our hybrid proposed methods are applied to daily\n51-h forecasts of 6-h accumulated precipitation from 2012 to 2015 over France\nusing the M{é}t{é}o-France ensemble prediction system called PEARP. They\nprovide calibrated pre-dictive distributions and compete favourably with\nstate-of-the-art methods like Analogs method or Ensemble Model Output\nStatistics. In particular, hybrid forest-based procedures appear to bring an\nadded value to the forecast of heavy rainfall.\n",
"title": "Forest-based methods and ensemble model output statistics for rainfall ensemble forecasting"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
12058
| null |
Validated
| null | null |
null |
{
"abstract": " A basic problem in information theory is the following: Let $\\mathbf{P} =\n(\\mathbf{X}, \\mathbf{Y})$ be an arbitrary distribution where the marginals\n$\\mathbf{X}$ and $\\mathbf{Y}$ are (potentially) correlated. Let Alice and Bob\nbe two players where Alice gets samples $\\{x_i\\}_{i \\ge 1}$ and Bob gets\nsamples $\\{y_i\\}_{i \\ge 1}$ and for all $i$, $(x_i, y_i) \\sim \\mathbf{P}$. What\njoint distributions $\\mathbf{Q}$ can be simulated by Alice and Bob without any\ninteraction?\nClassical works in information theory by G{á}cs-K{ö}rner and Wyner answer\nthis question when at least one of $\\mathbf{P}$ or $\\mathbf{Q}$ is the\ndistribution on $\\{0,1\\} \\times \\{0,1\\}$ where each marginal is unbiased and\nidentical. However, other than this special case, the answer to this question\nis understood in very few cases. Recently, Ghazi, Kamath and Sudan showed that\nthis problem is decidable for $\\mathbf{Q}$ supported on $\\{0,1\\} \\times\n\\{0,1\\}$. We extend their result to $\\mathbf{Q}$ supported on any finite\nalphabet.\nWe rely on recent results in Gaussian geometry (by the authors) as well as a\nnew \\emph{smoothing argument} inspired by the method of \\emph{boosting} from\nlearning theory and potential function arguments from complexity theory and\nadditive combinatorics.\n",
"title": "Non interactive simulation of correlated distributions is decidable"
}
| null | null | null | null | true | null |
12059
| null |
Default
| null | null |
null |
{
"abstract": " The computable model theory of modal logic was initiated by Suman Ganguli and\nAnil Nerode in [4]. They use an effective Henkin-type construction to\neffectivize various completeness theorems from classical modal logic. This\nconstruction has the feature of only producing models whose frames can be\nobtained by adding edges to a tree digraph. Consequently, this construction\ncannot prove an effective version of a well-known completeness theorem which\nstates that every $\\mathsf{S4.3.1}$-theory has a model whose accessibility\nrelation is a linear order of order type $\\omega$. We prove an effectivization\nof that theorem by means of a new construction adapted from that of Ganguli and\nNerode.\n",
"title": "Effective Completeness for S4.3.1-Theories with Respect to Discrete Linear Models"
}
| null | null |
[
"Mathematics"
] | null | true | null |
12060
| null |
Validated
| null | null |
null |
{
"abstract": " DeepTingle is a text prediction and classification system trained on the\ncollected works of the renowned fantastic gay erotica author Chuck Tingle.\nWhereas the writing assistance tools you use everyday (in the form of\npredictive text, translation, grammar checking and so on) are trained on\ngeneric, purportedly \"neutral\" datasets, DeepTingle is trained on a very\nspecific, internally consistent but externally arguably eccentric dataset. This\nallows us to foreground and confront the norms embedded in data-driven\ncreativity and productivity assistance tools. As such tools effectively\nfunction as extensions of our cognition into technology, it is important to\nidentify the norms they embed within themselves and, by extension, us.\nDeepTingle is realized as a web application based on LSTM networks and the\nGloVe word embedding, implemented in JavaScript with Keras-JS.\n",
"title": "DeepTingle"
}
| null | null | null | null | true | null |
12061
| null |
Default
| null | null |
null |
{
"abstract": " The ability of physical layer relay caching to increase the degrees of\nfreedom (DoF) of a single cell was recently illustrated. In this paper, we\nextend this result to the case of multiple cells in which a caching relay is\nshared among multiple non-cooperative base stations (BSs). In particular, we\nshow that a large DoF gain can be achieved by exploiting the benefits of having\na shared relay that cooperates with the BSs. We first propose a cache-assisted\nrelaying protocol that improves the cooperation opportunity between the BSs and\nthe relay. Next, we consider the cache content placement problem that aims to\ndesign the cache content at the relay such that the DoF gain is maximized. We\npropose an optimal algorithm and a near-optimal low-complexity algorithm for\nthe cache content placement problem. Simulation results show significant\nimprovement in the DoF gain using the proposed relay-caching protocol.\n",
"title": "Degrees of Freedom in Cached MIMO Relay Networks With Multiple Base Stations"
}
| null | null | null | null | true | null |
12062
| null |
Default
| null | null |
null |
{
"abstract": " The theoretical description of the thermodynamics of water is challenged by\nthe structural transition towards tetrahedral symmetry at ambient conditions.\nAs perturbation theories typically assume a spherically symmetric reference\nfluid, they are incapable of accurately describing the liquid properties of\nwater at ambient conditions. In this paper we solve this problem, by\nintroducing the concept of an associated reference perturbation theory (APT).\nIn APT we treat the reference fluid as an associating hard sphere fluid which\ntransitions to tetrahedral symmetry in the fully hydrogen bonded limit. We\ncalculate this transition in a theoretically self-consistent manner without\nappealing to molecular simulations. This associated reference provides the\nreference fluid for a second order Barker-Hendersen perturbative treatment of\nthe long-range attractions. We demonstrate that this new approach gives a\nsignificantly improved description of water as compared to standard\nperturbation theories.\n",
"title": "A perturbation theory for water with an associating reference fluid"
}
| null | null | null | null | true | null |
12063
| null |
Default
| null | null |
null |
{
"abstract": " We present bounds for the finite sample error of sequential Monte Carlo\nsamplers on static spaces. Our approach explicitly relates the performance of\nthe algorithm to properties of the chosen sequence of distributions and mixing\nproperties of the associated Markov kernels. This allows us to give the first\nfinite sample comparison to other Monte Carlo schemes. We obtain bounds for the\ncomplexity of sequential Monte Carlo approximations for a variety of target\ndistributions including finite spaces, product measures, and log-concave\ndistributions including Bayesian logistic regression. The bounds obtained are\nwithin a logarithmic factor of similar bounds obtainable for Markov chain Monte\nCarlo.\n",
"title": "Finite Sample Complexity of Sequential Monte Carlo Estimators"
}
| null | null |
[
"Statistics"
] | null | true | null |
12064
| null |
Validated
| null | null |
null |
{
"abstract": " Butanol has received significant research attention as a second-generation\nbiofuel in the past few years. In the present study, skeletal mechanisms for\nfour butanol isomers were generated from two widely accepted, well-validated\ndetailed chemical kinetic models for the butanol isomers. The detailed models\nwere reduced using a two-stage approach consisting of the directed relation\ngraph with error propagation and sensitivity analysis. During the reduction\nprocess, issues were encountered with pressure-dependent reactions formulated\nusing the logarithmic pressure interpolation approach; these issues are\ndiscussed and recommendations made to avoid ambiguity in its future\nimplementation in mechanism development. The performance of the skeletal\nmechanisms generated here was compared with that of detailed mechanisms in\nsimulations of autoignition delay times, laminar flame speeds, and perfectly\nstirred reactor temperature response curves and extinction residence times,\nover a wide range of pressures, temperatures, and equivalence ratios. The\ndetailed and skeletal mechanisms agreed well, demonstrating the adequacy of the\nresulting reduced chemistry for all the butanol isomers in predicting global\ncombustion phenomena. In addition, the skeletal mechanisms closely predicted\nthe time-histories of fuel mass fractions in homogeneous compression-ignition\nengine simulations. The performance of each butanol isomer was additionally\ncompared with that of a gasoline surrogate with an antiknock index of 87 in a\nhomogeneous compression-ignition engine simulation. The gasoline surrogate was\nconsumed faster than any of the butanol isomers, with tert-butanol exhibiting\nthe slowest fuel consumption rate. While n-butanol and isobutanol displayed the\nmost similar consumption profiles relative to the gasoline surrogate, the two\nliterature chemical kinetic models predicted different orderings.\n",
"title": "Reduced chemistry for butanol isomers at engine-relevant conditions"
}
| null | null | null | null | true | null |
12065
| null |
Default
| null | null |
null |
{
"abstract": " Mutual Information (MI) is an useful tool for the recognition of mutual\ndependence berween data sets. Differen methods for the estimation of MI have\nbeen developed when both data sets are discrete or when both data sets are\ncontinuous. The MI estimation between a discrete data set and a continuous data\nset has not received so much attention. We present here a method for the\nestimation of MI for this last case based on the kernel density approximation.\nThe calculation may be of interest in diverse contexts. Since MI is closely\nrelated to Jensen Shannon divergence, the method here developed is of\nparticular interest in the problem of sequence segmentation.\n",
"title": "A Measure of Dependence Between Discrete and Continuous Variables"
}
| null | null | null | null | true | null |
12066
| null |
Default
| null | null |
null |
{
"abstract": " Bayesian online changepoint detection (BOCPD) (Adams & MacKay, 2007) offers a\nrigorous and viable way to identity changepoints in complex systems. In this\nwork, we introduce a Stein variational online changepoint detection (SVOCD)\nmethod to provide a computationally tractable generalization of BOCPD beyond\nthe exponential family of probability distributions. We integrate the recently\ndeveloped Stein variational Newton (SVN) method (Detommaso et al., 2018) and\nBOCPD to offer a full online Bayesian treatment for a large number of\nsituations with significant importance in practice. We apply the resulting\nmethod to two challenging and novel applications: Hawkes processes and long\nshort-term memory (LSTM) neural networks. In both cases, we successfully\ndemonstrate the efficacy of our method on real data.\n",
"title": "Stein Variational Online Changepoint Detection with Applications to Hawkes Processes and Neural Networks"
}
| null | null | null | null | true | null |
12067
| null |
Default
| null | null |
null |
{
"abstract": " Let $\\mathcal{B}_d$ be the unital $C^*$-algebra generated by the elements\n$u_{jk}, \\, 0 \\le i, j \\le d-1$, satisfying the relations that $[u_{j,k}]$ is a\nunitary operator, and let $C^*(\\mathbb{F}_{d^2})$ be the full group\n$C^*$-algebra of free group of $d^2$ generators. Based on the idea of\nteleportation and super-dense coding in quantum information theory, we exhibit\nthe two $*$-isomorphisms $M_d(C^*(\\mathbb{F}_{d^2}))\\cong \\mathcal{B}_d\\rtimes\n\\mathbb{Z}_d\\rtimes \\mathbb{Z}_d$ and $M_d(\\mathcal{B}_d)\\cong\nC^*(\\mathbb{F}_{d^2})\\rtimes \\mathbb{Z}_d\\rtimes \\mathbb{Z}_d$, for certain\nactions of $\\mathbb{Z}_d$. As an application, we show that for any $d,m\\ge 2$\nwith $(d,m)\\neq (2,2)$, the matrix-valued generalization of the (tensor\nproduct) quantum correlation set of $d$ inputs and $m$ outputs is not closed.\n",
"title": "Quantum Teleportation and Super-dense Coding in Operator Algebras"
}
| null | null |
[
"Mathematics"
] | null | true | null |
12068
| null |
Validated
| null | null |
null |
{
"abstract": " The CANDECOMP/PARAFAC (CP) decomposition is a leading method for the analysis\nof multiway data. The standard alternating least squares algorithm for the CP\ndecomposition (CP-ALS) involves a series of highly overdetermined linear least\nsquares problems. We extend randomized least squares methods to tensors and\nshow the workload of CP-ALS can be drastically reduced without a sacrifice in\nquality. We introduce techniques for efficiently preprocessing, sampling, and\ncomputing randomized least squares on a dense tensor of arbitrary order, as\nwell as an efficient sampling-based technique for checking the stopping\ncondition. We also show more generally that the Khatri-Rao product (used within\nthe CP-ALS iteration) produces conditions favorable for direct sampling. In\nnumerical results, we see improvements in speed, reductions in memory\nrequirements, and robustness with respect to initialization.\n",
"title": "A Practical Randomized CP Tensor Decomposition"
}
| null | null | null | null | true | null |
12069
| null |
Default
| null | null |
null |
{
"abstract": " A potential flow around a circular cylinder is a commonly examined problem in\nan introductory physics class. We pose a similar problem but with different\nboundary conditions where a rectangular pole replaces a circular cylinder. We\ndemonstrate to solve the problem by deriving a general solution for the flow in\nthe form of an infinite series and determining the coefficients in the series\nusing a multiple linear regression. When the size of a pole is specified, our\nsolution provides a quantitative estimate of the characteristic length scale of\nthe potential flow. Our analysis implies that the potential flow around a\nrectangular pole of the diagonal 1 is equivalent to the potential flow around a\ncircle of diameter 0.78 to a distant observer.\n",
"title": "Two dimensional potential flow around a rectangular pole solved by a multiple linear regression"
}
| null | null | null | null | true | null |
12070
| null |
Default
| null | null |
null |
{
"abstract": " Bilateral trade is a fundamental economic scenario comprising a strategically\nacting buyer and seller, each holding valuations for the item, drawn from\npublicly known distributions. A mechanism is supposed to facilitate trade\nbetween these agents, if such trade is beneficial. It was recently shown that\nthe only mechanisms that are simultaneously DSIC, SBB, and ex-post IR, are\nfixed price mechanisms, i.e., mechanisms that are parametrised by a price p,\nand trade occurs if and only if the valuation of the buyer is at least p and\nthe valuation of the seller is at most p. The gain from trade is the increase\nin welfare that results from applying a mechanism; here we study the gain from\ntrade achievable by fixed price mechanisms. We explore this question for both\nthe bilateral trade setting, and a double auction setting where there are\nmultiple buyers and sellers. We first identify a fixed price mechanism that\nachieves a gain from trade of at least 2/r times the optimum, where r is the\nprobability that the seller's valuation does not exceed the buyer's valuation.\nThis extends a previous result by McAfee. Subsequently, we improve this\napproximation factor in an asymptotic sense, by showing that a more\nsophisticated rule for setting the fixed price results in an expected gain from\ntrade within a factor O(log(1/r)) of the optimal gain from trade. This is\nasymptotically the best approximation factor possible. Lastly, we extend our\nstudy of fixed price mechanisms to the double auction setting defined by a set\nof multiple i.i.d. unit demand buyers, and i.i.d. unit supply sellers. We\npresent a fixed price mechanism that achieves a gain from trade that achieves\nfor all epsilon > 0 a gain from trade of at least (1-epsilon) times the\nexpected optimal gain from trade with probability 1 - 2/e^{#T epsilon^2 /2},\nwhere #T is the expected number of trades resulting from the double auction.\n",
"title": "Fixed Price Approximability of the Optimal Gain From Trade"
}
| null | null | null | null | true | null |
12071
| null |
Default
| null | null |
null |
{
"abstract": " We report $^{139}$La and $^{63}$Cu NMR investigation of the successive charge\norder, spin order, and superconducting transitions in super-oxygenated\nLa$_2$CuO$_{4+y}$ single crystal with stage-4 excess oxygen order at\n$T_{stage}\\simeq 290$ K. We show that the stage-4 order induces tilting of\nCuO$_6$ octahedra below $T_{stage}$, which in turn causes $^{139}$La NMR line\nbroadening. The structural distortion continues to develop far below\n$T_{stage}$, and completes at $T_{charge}\\simeq 60$ K, where charge order sets\nin. This sequence is reminiscent of the the charge order transition in Nd\nco-doped La$_{1.88}$Sr$_{0.12}$CuO$_4$ that sets in once the low temperature\ntetragonal (LTT) phase is established. We also show that the paramagnetic\n$^{63}$Cu NMR signals are progressively wiped out below $T_{charge}$ due to\nenhanced low frequency spin fluctuations, but the residual $^{63}$Cu NMR\nsignals continue to exhibit the characteristics expected for optimally doped\nsuperconducting CuO$_2$ planes. This indicates that charge order in\nLa$_2$CuO$_{4+y}$ does not take place uniformly in space. Low frequency Cu spin\nfluctuations as probed by $^{139}$La nuclear spin-lattice relaxation rate are\nmildly glassy, and do not exhibit critical divergence at $T_{spin}$($\\simeq\nT_{c}$)=42 K. These findings, including the spatially inhomogeneous nature of\nthe charge ordered state, are qualitatively similar to the case of\nLa$_{1.885}$Sr$_{0.115}$CuO$_4$ [T. Imai et al., Phys. Rev. B 96 (2017) 224508,\nand A. Arsenault et al., Phys. Rev. B 97 (2018) 064511], but both charge and\nspin order take place more sharply in the present case.\n",
"title": "$^{139}$La and $^{63}$Cu NMR investigation of charge order in La$_{2}$CuO$_{4+y}$ ($T_{c}=42$K)"
}
| null | null | null | null | true | null |
12072
| null |
Default
| null | null |
null |
{
"abstract": " Field failures, that is, failures caused by faults that escape the testing\nphase leading to failures in the field, are unavoidable. Improving verification\nand validation activities before deployment can identify and timely remove many\nbut not all faults, and users may still experience a number of annoying\nproblems while using their software systems. This paper investigates the nature\nof field failures, to understand to what extent further improving in-house\nverification and validation activities can reduce the number of failures in the\nfield, and frames the need of new approaches that operate in the field. We\nreport the results of the analysis of the bug reports of five applications\nbelonging to three different ecosystems, propose a taxonomy of field failures,\nand discuss the reasons why failures belonging to the identified classes cannot\nbe detected at design time but shall be addressed at runtime. We observe that\nmany faults (70%) are intrinsically hard to detect at design-time.\n",
"title": "An Exploratory Study of Field Failures"
}
| null | null | null | null | true | null |
12073
| null |
Default
| null | null |
null |
{
"abstract": " Crowdsourcing has been successfully applied in many domains including\nastronomy, cryptography and biology. In order to test its potential for useful\napplication in a Smart Grid context, this paper investigates the extent to\nwhich a crowd can contribute predictive hypotheses to a model of residential\nelectric energy consumption. In this experiment, the crowd generated hypotheses\nabout factors that make one home different from another in terms of monthly\nenergy usage. To implement this concept, we deployed a web-based system within\nwhich 627 residential electricity customers posed 632 questions that they\nthought predictive of energy usage. While this occurred, the same group\nprovided 110,573 answers to these questions as they accumulated. Thus users\nboth suggested the hypotheses that drive a predictive model and provided the\ndata upon which the model is built. We used the resulting question and answer\ndata to build a predictive model of monthly electric energy consumption, using\nrandom forest regression. Because of the sparse nature of the answer data,\ncareful statistical work was needed to ensure that these models are valid. The\nresults indicate that the crowd can generate useful hypotheses, despite the\nsparse nature of the dataset.\n",
"title": "Crowdsourcing Predictors of Residential Electric Energy Usage"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
12074
| null |
Validated
| null | null |
null |
{
"abstract": " We study a class of one-dimensional classical fluids with penetrable\nparticles interacting through positive, purely repulsive, pair-potentials.\nStarting from some lower bounds to the total potential energy, we draw results\non the thermodynamic limit of the given model.\n",
"title": "One-dimensional fluids with positive potentials"
}
| null | null | null | null | true | null |
12075
| null |
Default
| null | null |
null |
{
"abstract": " In this letter, we propose an algorithm for recovery of sparse and low rank\ncomponents of matrices using an iterative method with adaptive thresholding. In\neach iteration, the low rank and sparse components are obtained using a\nthresholding operator. This algorithm is fast and can be implemented easily. We\ncompare it with one of the most common fast methods in which the rank and\nsparsity are approximated by $\\ell_1$ norm. We also apply it to some real\napplications where the noise is not so sparse. The simulation results show that\nit has a suitable performance with low run-time.\n",
"title": "Recovery of Sparse and Low Rank Components of Matrices Using Iterative Method with Adaptive Thresholding"
}
| null | null | null | null | true | null |
12076
| null |
Default
| null | null |
null |
{
"abstract": " This contributions discusses the simulation of magnetothermal effects in\nsuperconducting magnets as used in particle accelerators. An iterative coupling\nscheme using reduced order models between a magnetothermal partial differential\nmodel and an electrical lumped-element circuit is demonstrated. The\nmultiphysics, multirate and multiscale problem requires a consistent\nformulation and framework to tackle the challenging transient effects occurring\nat both system and device level.\n",
"title": "Reduced Order Modelling for the Simulation of Quenches in Superconducting Magnets"
}
| null | null | null | null | true | null |
12077
| null |
Default
| null | null |
null |
{
"abstract": " With the wide adoption of the multi-community setting in many popular social\nmedia platforms, the increasing user engagements across multiple online\ncommunities warrant research attention. In this paper, we introduce a novel\nanalogy between the movements in the cyber space and the physical space. This\nanalogy implies a new way of studying human online activities by modelling the\nactivities across online communities in a similar fashion as the movements\namong locations. First, we quantitatively validate the analogy by comparing\nseveral important properties of human online activities and physical movements.\nOur experiments reveal striking similarities between the cyber space and the\nphysical space. Next, inspired by the established methodology on human mobility\nin the physical space, we propose a framework to study human \"mobility\" across\nonline platforms. We discover three interesting patterns of user engagements in\nonline communities. Furthermore, our experiments indicate that people with\ndifferent mobility patterns also exhibit divergent preferences to online\ncommunities. This work not only attempts to achieve a better understanding of\nhuman online activities, but also intends to open a promising research\ndirection with rich implications and applications.\n",
"title": "Life in the \"Matrix\": Human Mobility Patterns in the Cyber Space"
}
| null | null | null | null | true | null |
12078
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we introduce and investigate a novel class of analytic and\nunivalent functions of negative coefficients in the open unit disk. For this\nfunction class, we obtain characterization and distortion theorems as well as\nthe radii of close-to-convexity, starlikeness and convexity by using fractional\ncalculus techniques.\n",
"title": "A Novel Subclass of Univalent Functions Involving Operators of Fractional Calculus"
}
| null | null | null | null | true | null |
12079
| null |
Default
| null | null |
null |
{
"abstract": " We propose an adaptive estimator for the stationary distribution of a\nbifurcating Markov Chain on $\\mathbb R^d$. Bifurcating Markov chains (BMC for\nshort) are a class of stochastic processes indexed by regular binary trees. A\nkernel estimator is proposed whose bandwidth is selected by a method inspired\nby the works of Goldenshluger and Lepski [18]. Drawing inspiration from\ndimension jump methods for model selection, we also provide an algorithm to\nselect the best constant in the penalty.\n",
"title": "Local bandwidth selection for kernel density estimation in bifurcating Markov chain model"
}
| null | null | null | null | true | null |
12080
| null |
Default
| null | null |
null |
{
"abstract": " We describe a high-performance implementation of the lattice-Boltzmann method\n(LBM) for sparse geometries on graphic processors. In our implementation we\ncover the whole geometry with a uniform mesh of small tiles and carry out\ncalculations for each tile independently with a proper data synchronization at\ntile edges. For this method we provide both the theoretical analysis of\ncomplexity and the results for real implementations for 2D and 3D geometries.\nBased on the theoretical model, we show that tiles offer significantly smaller\nbandwidth overhead than solutions based on indirect addressing. For\n2-dimensional lattice arrangements a reduction of memory usage is also\npossible, though at the cost of diminished performance. We reached the\nperformance of 682 MLUPS on GTX Titan (72\\% of peak theoretical memory\nbandwidth) for D3Q19 lattice arrangement and double precision data.\n",
"title": "Sparse geometries handling in lattice-Boltzmann method implementation for graphic processors"
}
| null | null | null | null | true | null |
12081
| null |
Default
| null | null |
null |
{
"abstract": " The multimodal web elements such as text and images are associated with\ninherent memory costs to store and transfer over the Internet. With the limited\nnetwork connectivity in developing countries, webpage rendering gets delayed in\nthe presence of high-memory demanding elements such as images (relative to\ntext). To overcome this limitation, we propose a Canonical Correlation Analysis\n(CCA) based computational approach to replace high-cost modality with an\nequivalent low-cost modality. Our model learns a common subspace for low-cost\nand high-cost modalities that maximizes the correlation between their visual\nfeatures. The obtained common subspace is used for determining the low-cost\n(text) element of a given high-cost (image) element for the replacement. We\nanalyze the cost-saving performance of the proposed approach through an\neye-tracking experiment conducted on real-world webpages. Our approach reduces\nthe memory-cost by at least 83.35% by replacing images with text.\n",
"title": "From Multimodal to Unimodal Webpages for Developing Countries"
}
| null | null | null | null | true | null |
12082
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the light-curve properties of a sample of 26 spectroscopically\nconfirmed hydrogen-poor superluminous supernovae (SLSNe-I) in the Palomar\nTransient Factory (PTF) survey. These events are brighter than SNe Ib/c and SNe\nIc-BL, on average, by about 4 and 2~mag, respectively. The peak absolute\nmagnitudes of SLSNe-I in rest-frame $g$ band span $-22\\lesssim M_g\n\\lesssim-20$~mag, and these peaks are not powered by radioactive $^{56}$Ni,\nunless strong asymmetries are at play. The rise timescales are longer for SLSNe\nthan for normal SNe Ib/c, by roughly 10 days, for events with similar decay\ntimes. Thus, SLSNe-I can be considered as a separate population based on\nphotometric properties. After peak, SLSNe-I decay with a wide range of slopes,\nwith no obvious gap between rapidly declining and slowly declining events. The\nlatter events show more irregularities (bumps) in the light curves at all\ntimes. At late times, the SLSN-I light curves slow down and cluster around the\n$^{56}$Co radioactive decay rate. Powering the late-time light curves with\nradioactive decay would require between 1 and 10${\\rm M}_\\odot$ of Ni masses.\nAlternatively, a simple magnetar model can reasonably fit the majority of\nSLSNe-I light curves, with four exceptions, and can mimic the radioactive decay\nof $^{56}$Co, up to $\\sim400$ days from explosion. The resulting spin values do\nnot correlate with the host-galaxy metallicities. Finally, the analysis of our\nsample cannot strengthen the case for using SLSNe-I for cosmology.\n",
"title": "Light curves of hydrogen-poor Superluminous Supernovae from the Palomar Transient Factory"
}
| null | null | null | null | true | null |
12083
| null |
Default
| null | null |
null |
{
"abstract": " Airbnb, an online marketplace for accommodations, has experienced a\nstaggering growth accompanied by intense debates and scattered regulations\naround the world. Current discourses, however, are largely focused on opinions\nrather than empirical evidences. Here, we aim to bridge this gap by presenting\nthe first large-scale measurement study on Airbnb, using a crawled data set\ncontaining 2.3 million listings, 1.3 million hosts, and 19.3 million reviews.\nWe measure several key characteristics at the heart of the ongoing debate and\nthe sharing economy. Among others, we find that Airbnb has reached a global yet\nheterogeneous coverage. The majority of its listings across many countries are\nentire homes, suggesting that Airbnb is actually more like a rental marketplace\nrather than a spare-room sharing platform. Analysis on star-ratings reveals\nthat there is a bias toward positive ratings, amplified by a bias toward using\npositive words in reviews. The extent of such bias is greater than Yelp\nreviews, which were already shown to exhibit a positive bias. We investigate a\nkey issue---commercial hosts who own multiple listings on Airbnb---repeatedly\ndiscussed in the current debate. We find that their existence is prevalent,\nthey are early-movers towards joining Airbnb, and their listings are\ndisproportionately entire homes and located in the US. Our work advances the\ncurrent understanding of how Airbnb is being used and may serve as an\nindependent and empirical reference to inform the debate.\n",
"title": "Sharing Means Renting?: An Entire-marketplace Analysis of Airbnb"
}
| null | null |
[
"Computer Science",
"Physics"
] | null | true | null |
12084
| null |
Validated
| null | null |
null |
{
"abstract": " Dynamic race detection is the problem of determining if an observed program\nexecution reveals the presence of a data race in a program. The classical\napproach to solving this problem is to detect if there is a pair of conflicting\nmemory accesses that are unordered by Lamport's happens-before (HB) relation.\nHB based race detection is known to not report false positives, i.e., it is\nsound. However, the soundness guarantee of HB only promises that the first pair\nof unordered, conflicting events is a schedulable data race. That is, there can\nbe pairs of HB-unordered conflicting data accesses that are not schedulable\nraces because there is no reordering of the events of the execution, where the\nevents in race can be executed immediately after each other. We introduce a new\npartial order, called schedulable happens-before (SHB) that exactly\ncharacterizes the pairs of schedulable data races --- every pair of conflicting\ndata accesses that are identified by SHB can be scheduled, and every HB-race\nthat can be scheduled is identified by SHB. Thus, the SHB partial order is\ntruly sound. We present a linear time, vector clock algorithm to detect\nschedulable races using SHB. Our experiments demonstrate the value of our\nalgorithm for dynamic race detection --- SHB incurs only little performance\noverhead and can scale to executions from real-world software applications\nwithout compromising soundness.\n",
"title": "What Happens - After the First Race? Enhancing the Predictive Power of Happens - Before Based Dynamic Race Detection"
}
| null | null | null | null | true | null |
12085
| null |
Default
| null | null |
null |
{
"abstract": " The process of exploring and exploiting Oil and Gas (O&G) generates a lot of\ndata that can bring more efficiency to the industry. The opportunities for\nusing data mining techniques in the \"digital oil-field\" remain largely\nunexplored or uncharted. With the high rate of data expansion, companies are\nscrambling to develop ways to develop near-real-time predictive analytics, data\nmining and machine learning capabilities, and are expanding their data storage\ninfrastructure and resources. With these new goals, come the challenges of\nmanaging data growth, integrating intelligence tools, and analyzing the data to\nglean useful insights. Oil and Gas companies need data solutions to\neconomically extract value from very large volumes of a wide variety of data\ngenerated from exploration, well drilling and production devices and sensors.\nData mining for oil and gas industry throughout the lifecycle of the\nreservoir includes the following roles: locating hydrocarbons, managing\ngeological data, drilling and formation evaluation, well construction, well\ncompletion, and optimizing production through the life of the oil field. For\neach of these phases during the lifecycle of oil field, data mining play a\nsignificant role. Based on which phase were talking about, knowledge creation\nthrough scientific models, data analytics and machine learning, a effective,\nproductive, and on demand data insight is critical for decision making within\nthe organization.\nThe significant challenges posed by this complex and economically vital field\njustify a meeting of data scientists that are willing to share their experience\nand knowledge. Thus, the Worskhop on Data Mining for Oil and Gas (DM4OG) aims\nto provide a quality forum for researchers that work on the significant\nchallenges arising from the synergy between data science, machine learning, and\nthe modeling and optimization problems in the O&G industry.\n",
"title": "Proceedings of the Workshop on Data Mining for Oil and Gas"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
12086
| null |
Validated
| null | null |
null |
{
"abstract": " We study a photonic analog of the chiral magnetic (vortical) effect. We\ndiscuss that the vector component of magnetoelectric tensors plays a role of\n\"vector potential,\" and its rotation is understood as \"magnetic field\" of a\nlight. Using the geometrical optics approximation, we show that \"magnetic\nfields\" cause an anomalous shift of a wave packet of a light through an\ninterplay with the Berry curvature of photons. The mechanism is the same as\nthat of the chiral magnetic (vortical) effect of a chiral fermion, so that we\nterm the anomalous shift \"chiral magnetic effect of a light.\" We further study\nthe chiral magnetic effect of a light beyond geometric optics by directly\nsolving the transmission problem of a wave packet at a surface of a\nmagnetoelectric material. We show that the experimental signal of the chiral\nmagnetic effect of a light is the nonvanishing of transverse displacements for\nthe beam normally incident to a magnetoelectric material.\n",
"title": "Chiral magnetic effect of light"
}
| null | null | null | null | true | null |
12087
| null |
Default
| null | null |
null |
{
"abstract": " The effects of pressure on the crystal structure of the three known\npolymorphs of magnesium sulfate have been theoretically study by means of DFT\ncalculations up to 45 GPa. We determined that at ambient conditions gamma MgSO4\nis an unstable polymorph, which decompose into MgO and SO3, and that the\nresponse of the other two polymorphs to hydrostatic pressure is non isotropic.\nAdditionally we found that at all pressures beta MgSO4 has a largest enthalpy\nthan alpha MgSO4. This indicates that beta MgSO4 is thermodynamically unstable\nversus alpha MgSO4 and predicts the occurrence of a beta alpha phase transition\nunder moderate compression. Our calculations also predict the existence under\npressure of additional phase transitions to two new polymorphs of MgSO4, which\nwe named as delta MgSO4 and epsilon MgSO4. The alpha delta transition is\npredicted to occur at 17.5 GPa, and the delta epsilon transition at 35 GPa,\npressures that nowadays can be experimentally easily achieved. All the\npredicted structural transforma ions are characterized as first order\ntransitions. This suggests that they can be non reversible, and therefore the\nnew polymorphs could be recovered as metastable polymorphs at ambient\nconditions. The crystal structure of the two new polymorphs is reported. In\nthem, the coordination number of sulfur is four as in the previously known\npolymorphs, but the coordination number of magnesium is eight instead of six.\nIn the article we will report the axial and bond compressibility for the four\npolymorphs of MgSO4. The pressure volume equation of state of each phase is\nalso given. The values obtained for the bulk modulus are 62 GPa, 57 GPa, 102\nGPa, and 119 GPa for alpha MgSO4, beta MgSO4, delta MgSO4, and epsilon MgSO4,\nrespectively. Finally, the electronic band structure of these four polymorphs\nof MgSO4 has been calculated by the first time.\n",
"title": "New Pressure-Induced Polymorphic Transitions of Anhydrous Magnesium Sulfate"
}
| null | null |
[
"Physics"
] | null | true | null |
12088
| null |
Validated
| null | null |
null |
{
"abstract": " We link the theory of optimal transportation to the theory of integer\npartitions. Let $\\mathscr P(n)$ denote the set of integer partitions of $n \\in\n\\mathbb N$ and write partitions $\\pi \\in \\mathscr P(n)$ as $(n_1, \\dots,\nn_{k(\\pi)})$. Using terminology from optimal transport, we characterize certain\nclasses of partitions like symmetric partitions and those in Euler's identity\n$|\\{ \\pi \\in \\mathscr P(n) |$ all $ n_i $ distinct $ \\} | = | \\{ \\pi \\in\n\\mathscr P(n) | $ all $ n_i $ odd $ \\}|$.\nThen we sketch how optimal transport might help to understand higher\ndimensional partitions.\n",
"title": "Optimal transport and integer partitions"
}
| null | null | null | null | true | null |
12089
| null |
Default
| null | null |
null |
{
"abstract": " This paper studies an intelligent ultimate technique for health-monitoring\nand prognostic of common rotary machine components, particularly bearings.\nDuring a run-to-failure experiment, rich unsupervised features from vibration\nsensory data are extracted by a trained sparse auto-encoder. Then, the\ncorrelation of the extracted attributes of the initial samples (presumably\nhealthy at the beginning of the test) with the succeeding samples is calculated\nand passed through a moving-average filter. The normalized output is named\nauto-encoder correlation-based (AEC) rate which stands for an informative\nattribute of the system depicting its health status and precisely identifying\nthe degradation starting point. We show that AEC technique well-generalizes in\nseveral run-to-failure tests. AEC collects rich unsupervised features form the\nvibration data fully autonomous. We demonstrate the superiority of the AEC over\nmany other state-of-the-art approaches for the health monitoring and prognostic\nof machine bearings.\n",
"title": "An Automated Auto-encoder Correlation-based Health-Monitoring and Prognostic Method for Machine Bearings"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
12090
| null |
Validated
| null | null |
null |
{
"abstract": " This book introduces a temporal type theory, the first of its kind as far as\nwe know. It is based on a standard core, and as such it can be formalized in a\nproof assistant such as Coq or Lean by adding a number of axioms. Well-known\ntemporal logics---such as Linear and Metric Temporal Logic (LTL and\nMTL)---embed within the logic of temporal type theory.\nThe types in this theory represent \"behavior types\". The language is rich\nenough to allow one to define arbitrary hybrid dynamical systems, which are\nmixtures of continuous dynamics---e.g. as described by a differential\nequation---and discrete jumps. In particular, the derivative of a continuous\nreal-valued function is internally defined.\nWe construct a semantics for the temporal type theory in the topos of sheaves\non a translation-invariant quotient of the standard interval domain. In fact,\ndomain theory plays a recurring role in both the semantics and the type theory.\n",
"title": "Temporal Type Theory: A topos-theoretic approach to systems and behavior"
}
| null | null | null | null | true | null |
12091
| null |
Default
| null | null |
null |
{
"abstract": " We provide a direct construction of Poletsky discs via local arc\napproximation and a Runge-type theorem by A. Gournay.\n",
"title": "On Poletsky theory of discs in compact manifolds"
}
| null | null | null | null | true | null |
12092
| null |
Default
| null | null |
null |
{
"abstract": " DR-submodular continuous functions are important objectives with wide\nreal-world applications spanning MAP inference in determinantal point processes\n(DPPs), and mean-field inference for probabilistic submodular models, amongst\nothers. DR-submodularity captures a subclass of non-convex functions that\nenables both exact minimization and approximate maximization in polynomial\ntime.\nIn this work we study the problem of maximizing non-monotone DR-submodular\ncontinuous functions under general down-closed convex constraints. We start by\ninvestigating geometric properties that underlie such objectives, e.g., a\nstrong relation between (approximately) stationary points and global optimum is\nproved. These properties are then used to devise two optimization algorithms\nwith provable guarantees. Concretely, we first devise a \"two-phase\" algorithm\nwith $1/4$ approximation guarantee. This algorithm allows the use of existing\nmethods for finding (approximately) stationary points as a subroutine, thus,\nharnessing recent progress in non-convex optimization. Then we present a\nnon-monotone Frank-Wolfe variant with $1/e$ approximation guarantee and\nsublinear convergence rate. Finally, we extend our approach to a broader class\nof generalized DR-submodular continuous functions, which captures a wider\nspectrum of applications. Our theoretical findings are validated on synthetic\nand real-world problem instances.\n",
"title": "Continuous DR-submodular Maximization: Structure and Algorithms"
}
| null | null | null | null | true | null |
12093
| null |
Default
| null | null |
null |
{
"abstract": " A symmetric matrix is Robinsonian if its rows and columns can be\nsimultaneously reordered in such a way that entries are monotone nondecreasing\nin rows and columns when moving toward the diagonal. The adjacency matrix of a\ngraph is Robinsonian precisely when the graph is a unit interval graph, so that\nRobinsonian matrices form a matrix analogue of the class of unit interval\ngraphs. Here we provide a structural characterization for Robinsonian matrices\nin terms of forbidden substructures, extending the notion of asteroidal triples\nto weighted graphs. This implies the known characterization of unit interval\ngraphs and leads to an efficient algorithm for certifying that a matrix is not\nRobinsonian.\n",
"title": "A Structural Characterization for Certifying Robinsonian Matrices"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
12094
| null |
Validated
| null | null |
null |
{
"abstract": " The detection of gravity plays a fundamental role during the growth and\nevolution of plants. Although progress has been made in our understanding of\nthe molecular, cellular and physical mechanisms involved in the gravity\ndetection, a coherent scenario consistent with all the observations is still\nlacking. In this perspective paper we discuss recent experiments showing that\nthe response to inclination of shoots is independent of the gravity intensity,\nmeaning that the gravity sensor detects an inclination and not a force. This\nresult questions some of the commonly accepted hypotheses and leads to propose\na new \"position sensor hypothesis\". The implications of this new scenario are\ndiscussed in the light of different observations available in the literature.\n",
"title": "A new scenario for gravity detection in plants: the position sensor hypothesis"
}
| null | null | null | null | true | null |
12095
| null |
Default
| null | null |
null |
{
"abstract": " Markov Chain Monte Carlo (MCMC) methods such as Gibbs sampling are finding\nwidespread use in applied statistics and machine learning. These often lead to\ndifficult computational problems, which are increasingly being solved on\nparallel and distributed systems such as compute clusters. Recent work has\nproposed running iterative algorithms such as gradient descent and MCMC in\nparallel asynchronously for increased performance, with good empirical results\nin certain problems. Unfortunately, for MCMC this parallelization technique\nrequires new convergence theory, as it has been explicitly demonstrated to lead\nto divergence on some examples. Recent theory on Asynchronous Gibbs sampling\ndescribes why these algorithms can fail, and provides a way to alter them to\nmake them converge. In this article, we describe how to apply this theory in a\ngeneric setting, to understand the asynchronous behavior of any MCMC algorithm,\nincluding those implemented using parameter servers, and those not based on\nGibbs sampling.\n",
"title": "Techniques for proving Asynchronous Convergence results for Markov Chain Monte Carlo methods"
}
| null | null | null | null | true | null |
12096
| null |
Default
| null | null |
null |
{
"abstract": " Hurricanes are cyclones circulating about a defined center whose closed wind\nspeeds exceed 75 mph originating over tropical and subtropical waters. At\nlandfall, hurricanes can result in severe disasters. The accuracy of predicting\ntheir trajectory paths is critical to reduce economic loss and save human\nlives. Given the complexity and nonlinearity of weather data, a recurrent\nneural network (RNN) could be beneficial in modeling hurricane behavior. We\npropose the application of a fully connected RNN to predict the trajectory of\nhurricanes. We employed the RNN over a fine grid to reduce typical truncation\nerrors. We utilized their latitude, longitude, wind speed, and pressure\npublicly provided by the National Hurricane Center (NHC) to predict the\ntrajectory of a hurricane at 6-hour intervals. Results show that this proposed\ntechnique is competitive to methods currently employed by the NHC and can\npredict up to approximately 120 hours of hurricane path.\n",
"title": "Predicting Hurricane Trajectories using a Recurrent Neural Network"
}
| null | null | null | null | true | null |
12097
| null |
Default
| null | null |
null |
{
"abstract": " A stochastic optimal control problem driven by an abstract evolution equation\nin a separable Hilbert space is considered. Thanks to the identification of the\nmild solution of the state equation as $\\nu$-weak Dirichlet process, the value\nprocesses is proved to be a real weak Dirichlet process. The uniqueness of the\ncorresponding decomposition is used to prove a verification theorem. Through\nthat technique several of the required assumptions are milder than those\nemployed in previous contributions about non-regular solutions of\nHamilton-Jacobi-Bellman equations.\n",
"title": "HJB equations in infinite dimension and optimal control of stochastic evolution equations via generalized Fukushima decomposition"
}
| null | null | null | null | true | null |
12098
| null |
Default
| null | null |
null |
{
"abstract": " Magnetic induction was first proposed as a planetary heating mechanism by\nSonett and Colburn in 1968, in recent years this theory has lost favor as a\nplausible source of heating in the early solar system. However, new models of\nproto-planetary disk evolution suggest that magnetic fields play an important\nrole in solar system formation. In particular, the magneto-hydrodynamic\nbehavior of proto-planetary disks is believed to be responsible for the net\noutward flow of angular momentum in the solar system. It is important to\nre-evaluate the plausibility of magnetic induction based on the intense\nmagnetic field environments described by the most recent models of\nproto-planetary disk evolution.\nIn order to re-evaluate electromagnetic induction theory the electrical\nconductivity of meteorites must be determined. To develop a technique capable\nof making these measurements, a time-varying magnetic field was generated to\ninductively heat metallic control samples. The thermal response of each sample,\nwhich depends on electrical conductivity, was monitored until a thermal steady\nstate was achieved. The relationship between conductivity and thermal response\ncan be exploited to estimate the electrical conductivity of unknown samples.\nAfter applying the technique to various metals it was recognized that this\nmethod is not capable of making precise electrical conductivity measurements.\nHowever, this method can constrain the product of the electrical conductivity\nand the square of the magnetic permeability, or ${\\sigma}{{\\mu}^2}$, for\nmeteoritic and metallic samples alike. The results also illustrate that along\nwith electrical conductivity {\\sigma}, the magnetic permeability {\\mu} of a\nsubstance has an important effect on induction heating phenomena for\nparamagnetic ({\\mu}/{\\mu}0 > 1) and especially ferromagnetic materials\n({\\mu}/{\\mu}0 >> 1).\n",
"title": "Developing a Method to Determine Electrical Conductivity in Meteoritic Materials with Applications to Induction Heating Theory (2008 Student Thesis)"
}
| null | null | null | null | true | null |
12099
| null |
Default
| null | null |
null |
{
"abstract": " We investigate possible signatures of halo assembly bias for\nspectroscopically selected galaxy groups from the GAMA survey using weak\nlensing measurements from the spatially overlapping regions of the deeper,\nhigh-imaging-quality photometric KiDS survey. We use GAMA groups with an\napparent richness larger than 4 to identify samples with comparable mean host\nhalo masses but with a different radial distribution of satellite galaxies,\nwhich is a proxy for the formation time of the haloes. We measure the weak\nlensing signal for groups with a steeper than average and with a shallower than\naverage satellite distribution and find no sign of halo assembly bias, with the\nbias ratio of $0.85^{+0.37}_{-0.25}$, which is consistent with the $\\Lambda$CDM\nprediction. Our galaxy groups have typical masses of $10^{13} M_{\\odot}/h$,\nnaturally complementing previous studies of halo assembly bias on galaxy\ncluster scales.\n",
"title": "A KiDS weak lensing analysis of assembly bias in GAMA galaxy groups"
}
| null | null | null | null | true | null |
12100
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.