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": " We provide comments on the article \"High-dimensional simultaneous inference\nwith the bootstrap\" by Ruben Dezeure, Peter Buhlmann and Cun-Hui Zhang.\n",
"title": "Comments on `High-dimensional simultaneous inference with the bootstrap'"
}
| null | null |
[
"Statistics"
] | null | true | null |
13101
| null |
Validated
| null | null |
null |
{
"abstract": " Variable clustering is important for explanatory analysis. However, only few\ndedicated methods for variable clustering with the Gaussian graphical model\nhave been proposed. Even more severe, small insignificant partial correlations\ndue to noise can dramatically change the clustering result when evaluating for\nexample with the Bayesian Information Criteria (BIC). In this work, we try to\naddress this issue by proposing a Bayesian model that accounts for negligible\nsmall, but not necessarily zero, partial correlations. Based on our model, we\npropose to evaluate a variable clustering result using the marginal likelihood.\nTo address the intractable calculation of the marginal likelihood, we propose\ntwo solutions: one based on a variational approximation, and another based on\nMCMC. Experiments on simulated data shows that the proposed method is similarly\naccurate as BIC in the no noise setting, but considerably more accurate when\nthere are noisy partial correlations. Furthermore, on real data the proposed\nmethod provides clustering results that are intuitively sensible, which is not\nalways the case when using BIC or its extensions.\n",
"title": "Robust Bayesian Model Selection for Variable Clustering with the Gaussian Graphical Model"
}
| null | null | null | null | true | null |
13102
| null |
Default
| null | null |
null |
{
"abstract": " In its weak field limit, Scalar-tensor-vector gravity theory introduces a\nYukawa-correction to the gravitational potential. Such a correction depends on\nthe two parameters, $\\alpha$ which accounts for the modification of the\ngravitational constant, and $\\mu^{*-1}$ which represents the scale length on\nwhich the scalar field propagates. These parameters were found to be universal\nwhen the modified gravitational potential was used to fit the galaxy rotation\ncurves and the mass profiles of galaxy clusters, both without Dark Matter. We\ntest the universality of these parameters using the the temperature\nanisotropies due to the thermal Sunyaev-Zeldovich effect. In our model the\nintra-cluster gas is in hydrostatic equilibrium within the modified\ngravitational potential well and it is described by a polytropic equation of\nstate. We predict the thermal Sunyaev-Zeldovich temperature anisotropies\nproduced by Coma cluster, and we compare them with those obtained using the\nPlanck 2013 Nominal maps. In our analysis, we find $\\alpha$ and the scale\nlength, respectively, to be consistent and to depart from their universal\nvalues. Our analysis points out that the assumption of the universality of the\nYukawa-correction to the gravitational potential is ruled out at more than\n$3.5\\sigma$ at galaxy clusters scale, while demonstrating that such a theory of\ngravity is capable to fit the cluster profile if the scale dependence of the\ngravitational potential is restored.\n",
"title": "On the universality of MOG weak field approximation at galaxy cluster scale"
}
| null | null | null | null | true | null |
13103
| null |
Default
| null | null |
null |
{
"abstract": " We develop several efficient algorithms for the classical \\emph{Matrix\nScaling} problem, which is used in many diverse areas, from preconditioning\nlinear systems to approximation of the permanent. On an input $n\\times n$\nmatrix $A$, this problem asks to find diagonal (scaling) matrices $X$ and $Y$\n(if they exist), so that $X A Y$ $\\varepsilon$-approximates a doubly\nstochastic, or more generally a matrix with prescribed row and column sums.\nWe address the general scaling problem as well as some important special\ncases. In particular, if $A$ has $m$ nonzero entries, and if there exist $X$\nand $Y$ with polynomially large entries such that $X A Y$ is doubly stochastic,\nthen we can solve the problem in total complexity $\\tilde{O}(m + n^{4/3})$.\nThis greatly improves on the best known previous results, which were either\n$\\tilde{O}(n^4)$ or $O(m n^{1/2}/\\varepsilon)$.\nOur algorithms are based on tailor-made first and second order techniques,\ncombined with other recent advances in continuous optimization, which may be of\nindependent interest for solving similar problems.\n",
"title": "Much Faster Algorithms for Matrix Scaling"
}
| null | null | null | null | true | null |
13104
| null |
Default
| null | null |
null |
{
"abstract": " In the protein sequence space, natural proteins form clusters of families\nwhich are characterized by their unique native folds whereas the great majority\nof random polypeptides are neither clustered nor foldable to unique structures.\nSince a given polypeptide can be either foldable or unfoldable, a kind of\n\"folding transition\" is expected at the boundary of a protein family in the\nsequence space. By Monte Carlo simulations of a statistical mechanical model of\nprotein sequence alignment that coherently incorporates both short-range and\nlong-range interactions as well as variable-length insertions to reproduce the\nstatistics of the multiple sequence alignment of a given protein family, we\ndemonstrate the existence of such transition between natural-like sequences and\nrandom sequences in the sequence subspaces for 15 domain families of various\nfolds. The transition was found to be highly cooperative and two-state-like.\nFurthermore, enforcing or suppressing consensus residues on a few of the\nwell-conserved sites enhanced or diminished, respectively, the natural-like\npattern formation over the entire sequence. In most families, the key sites\nincluded ligand binding sites. These results suggest some selective pressure on\nthe key residues, such as ligand binding activity, may cooperatively facilitate\nthe emergence of a protein family during evolution. From a more practical\naspect, the present results highlight an essential role of long-range effects\nin precisely defining protein families, which are absent in conventional\nsequence models.\n",
"title": "Cooperative \"folding transition\" in the sequence space facilitates function-driven evolution of protein families"
}
| null | null | null | null | true | null |
13105
| null |
Default
| null | null |
null |
{
"abstract": " The control of solute fluxes through either microscopic phoresis or\nhydrodynamic advection is a fundamental way to transport molecules, which are\nubiquitously present in nature and technology. We study the transport of large\nsolute such as DNA driven by a time-dependent thermal field in a polymer\nsolution. Heat propagation of a single heat spot moving back and forth gives\nrise to the molecular focusing of DNA with frequency-tunable control. We\ndeveloped a theoretical model, where heat conduction, viscoelastic expansion of\nwalls, and the viscosity gradient of a smaller solute are coupled, and that can\nexplain the underlying hydrodynamic focusing and its interplay with phoretic\ntransports. This cross effect may allow one to design a unique miniaturized\npump in microfluidics.\n",
"title": "Thermal Molecular Focusing: Tunable Cross Effect of Phoresis and Advection"
}
| null | null | null | null | true | null |
13106
| null |
Default
| null | null |
null |
{
"abstract": " Magnetohydrodynamically induced interface instability in liquid metal\nbatteries is analyzed. The batteries are represented by a simplified system in\nthe form of a rectangular cell, in which strong vertical electric current flows\nthrough three horizontal layers: the layer of a heavy metal at the bottom, the\nlayer of a light metal at the top, and the layer of electrolyte in the middle.\nA new two-dimensional nonlinear model based on the conservative shallow water\napproximation is derived and utilized in a numerical study. It is found that in\nthe case of small density difference between the electrolyte and one of the\nmetals, the instability closely resembles the rolling pad instability observed\nearlier in the aluminum reduction cells. When the two electrolyte-metal density\ndifferences are comparable, the dynamics of unstable systems is more complex\nand characterized by interaction between two nearly symmetric or antisymmetric\ninterfacial waves.\n",
"title": "Shallow water modeling of rolling pad instability in liquid metal batteries"
}
| null | null | null | null | true | null |
13107
| null |
Default
| null | null |
null |
{
"abstract": " The clustering problem, in its many variants, has numerous applications in\noperations research and computer science (e.g., in applications in\nbioinformatics, image processing, social network analysis, etc.). As sizes of\ndata sets have grown rapidly, researchers have focused on designing algorithms\nfor clustering problems in models of computation suited for large-scale\ncomputation such as MapReduce, Pregel, and streaming models. The $k$-machine\nmodel (Klauck et al., SODA 2015) is a simple, message-passing model for\nlarge-scale distributed graph processing. This paper considers three of the\nmost prominent examples of clustering problems: the uncapacitated facility\nlocation problem, the $p$-median problem, and the $p$-center problem and\npresents $O(1)$-factor approximation algorithms for these problems running in\n$\\tilde{O}(n/k)$ rounds in the $k$-machine model. These algorithms are optimal\nup to polylogarithmic factors because this paper also shows\n$\\tilde{\\Omega}(n/k)$ lower bounds for obtaining polynomial-factor\napproximation algorithms for these problems. These are the first results for\nclustering problems in the $k$-machine model.\nWe assume that the metric provided as input for these clustering problems in\nonly implicitly provided, as an edge-weighted graph and in a nutshell, our main\ntechnical contribution is to show that constant-factor approximation algorithms\nfor all three clustering problems can be obtained by learning only a small\nportion of the input metric.\n",
"title": "Near-Optimal Clustering in the $k$-machine model"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13108
| null |
Validated
| null | null |
null |
{
"abstract": " We develop procedures, based on minimization of the composition $f(x) =\nh(c(x))$ of a convex function $h$ and smooth function $c$, for solving random\ncollections of quadratic equalities, applying our methodology to phase\nretrieval problems. We show that the prox-linear algorithm we develop can solve\nphase retrieval problems---even with adversarially faulty measurements---with\nhigh probability as soon as the number of measurements $m$ is a constant factor\nlarger than the dimension $n$ of the signal to be recovered. The algorithm\nrequires essentially no tuning---it consists of solving a sequence of convex\nproblems---and it is implementable without any particular assumptions on the\nmeasurements taken. We provide substantial experiments investigating our\nmethods, indicating the practical effectiveness of the procedures and showing\nthat they succeed with high probability as soon as $m / n \\ge 2$ when the\nsignal is real-valued.\n",
"title": "Solving (most) of a set of quadratic equalities: Composite optimization for robust phase retrieval"
}
| null | null | null | null | true | null |
13109
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we investigate the problem of grasping novel objects in\nunstructured environments. To address this problem, consideration of the object\ngeometry, reachability and force closure analysis are required. We propose a\nframework for grasping unknown objects by localizing contact regions on the\ncontours formed by a set of depth edges in a single view 2D depth image.\nAccording to the edge geometric features obtained from analyzing the data of\nthe depth map, the contact regions are determined. Finally,We validate the\nperformance of the approach by applying it to the scenes with both single and\nmultiple objects, using Baxter manipulator.\n",
"title": "Edge-Based Recognition of Novel Objects for Robotic Grasping"
}
| null | null | null | null | true | null |
13110
| null |
Default
| null | null |
null |
{
"abstract": " We study the problem of single-image depth estimation for images in the wild.\nWe collect human annotated surface normals and use them to train a neural\nnetwork that directly predicts pixel-wise depth. We propose two novel loss\nfunctions for training with surface normal annotations. Experiments on NYU\nDepth and our own dataset demonstrate that our approach can significantly\nimprove the quality of depth estimation in the wild.\n",
"title": "Surface Normals in the Wild"
}
| null | null | null | null | true | null |
13111
| null |
Default
| null | null |
null |
{
"abstract": " The complexity of a geodesic language has connections to algebraic properties\nof the group. Gilman, Hermiller, Holt, and Rees show that a finitely generated\ngroup is virtually free if and only if its geodesic language is locally\nexcluding for some finite inverse-closed generating set. The existence of such\na correspondence and the result of Hermiller, Holt, and Rees that finitely\ngenerated abelian groups have piecewise excluding geodesic language for all\nfinite inverse-closed generating sets motivated our work. We show that a\nfinitely generated group with piecewise excluding geodesic language need not be\nabelian and give a class of infinite non-abelian groups which have piecewise\nexcluding geodesic languages for certain generating sets. The quaternion group\nis shown to be the only non-abelian 2-generator group with piecewise excluding\ngeodesic language for all finite inverse-closed generating sets. We also show\nthat there are virtually abelian groups with geodesic languages which are not\npiecewise excluding for any finite inverse-closed generating set.\n",
"title": "Piecewise excluding geodesic languages"
}
| null | null | null | null | true | null |
13112
| null |
Default
| null | null |
null |
{
"abstract": " Purpose: The radial k-space trajectory is a well-established sampling\ntrajectory used in conjunction with magnetic resonance imaging. However, the\nradial k-space trajectory requires a large number of radial lines for\nhigh-resolution reconstruction. Increasing the number of radial lines causes\nlonger acquisition time, making it more difficult for routine clinical use. On\nthe other hand, if we reduce the number of radial lines, streaking artifact\npatterns are unavoidable. To solve this problem, we propose a novel deep\nlearning approach with domain adaptation to restore high-resolution MR images\nfrom under-sampled k-space data.\nMethods: The proposed deep network removes the streaking artifacts from the\nartifact corrupted images. To address the situation given the limited available\ndata, we propose a domain adaptation scheme that employs a pre-trained network\nusing a large number of x-ray computed tomography (CT) or synthesized radial MR\ndatasets, which is then fine-tuned with only a few radial MR datasets.\nResults: The proposed method outperforms existing compressed sensing\nalgorithms, such as the total variation and PR-FOCUSS methods. In addition, the\ncalculation time is several orders of magnitude faster than the total variation\nand PR-FOCUSS methods.Moreover, we found that pre-training using CT or MR data\nfrom similar organ data is more important than pre-training using data from the\nsame modality for different organ.\nConclusion: We demonstrate the possibility of a domain-adaptation when only a\nlimited amount of MR data is available. The proposed method surpasses the\nexisting compressed sensing algorithms in terms of the image quality and\ncomputation time.\n",
"title": "Deep Learning with Domain Adaptation for Accelerated Projection-Reconstruction MR"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13113
| null |
Validated
| null | null |
null |
{
"abstract": " Efficient human-machine networks require productive interaction between human\nand machine actors. In this study, we address how a strengthening of machine\nagency, for example through increasing levels of automation, affect the human\nactors of the networks. Findings from case studies within air traffic\nmanagement, crisis management, and crowd evacuation are presented, exemplifying\nhow automation may strengthen the agency of human actors in the network through\nresponsibility sharing and task allocation, and serve as a needed prerequisite\nof innovation and change.\n",
"title": "Automation in Human-Machine Networks: How Increasing Machine Agency Affects Human Agency"
}
| null | null | null | null | true | null |
13114
| null |
Default
| null | null |
null |
{
"abstract": " Dynamics reduces the orthorhombicity of magnetic stripes in La_2CuO_4+y. The\nmeasured stripe incommensuration can be used to determine the oxygen content of\nthe sample.\n",
"title": "Comment on \"On the nature of magnetic stripes in cuprate superconductors,\" by H. Jacobsen et al., arXiv:1704.08528v2"
}
| null | null |
[
"Physics"
] | null | true | null |
13115
| null |
Validated
| null | null |
null |
{
"abstract": " Ideally, by enabling multi-tenancy, network virtualization allows to improve\nresource utilization, while providing performance isolation: although the\nunderlying resources are shared, the virtual network appears as a dedicated\nnetwork to the tenant. However, providing such an illusion is challenging in\npractice, and over the last years, many expedient approaches have been proposed\nto provide performance isolation in virtual networks, by enforcing bandwidth\nreservations. We in this paper study another source for overheads and\nunpredictable performance in virtual networks: the hypervisor.\nThe hypervisor is a critical component in multi-tenant environments, but its\noverhead and influence on performance are hardly understood today. In\nparticular, we focus on OpenFlow-based virtualized Software Defined Networks\n(vSDNs). Network virtualization is considered a killer application for SDNs: a\nvSDN allows each tenant to flexibly manage its network from a logically\ncentralized perspective, via a simple API. For the purpose of our study, we\ndeveloped a new benchmarking tool for OpenFlow control and data planes,\nenabling high and consistent OpenFlow message rates. Using our tool, we\nidentify and measure controllable and uncontrollable effects on performance and\noverhead, including the hypervisor technology, the number of tenants as well as\nthe tenant type, as well as the type of OpenFlow messages.\n",
"title": "Logically Isolated, Actually Unpredictable? Measuring Hypervisor Performance in Multi-Tenant SDNs"
}
| null | null | null | null | true | null |
13116
| null |
Default
| null | null |
null |
{
"abstract": " The Bayesian estimation of the unknown parameters of state-space (dynamical)\nsystems has received considerable attention over the past decade, with a\nhandful of powerful algorithms being introduced. In this paper we tackle the\ntheoretical analysis of the recently proposed {\\it nonlinear} population Monte\nCarlo (NPMC). This is an iterative importance sampling scheme whose key\nfeatures, compared to conventional importance samplers, are (i) the approximate\ncomputation of the importance weights (IWs) assigned to the Monte Carlo samples\nand (ii) the nonlinear transformation of these IWs in order to prevent the\ndegeneracy problem that flaws the performance of conventional importance\nsamplers. The contribution of the present paper is a rigorous proof of\nconvergence of the nonlinear IS (NIS) scheme as the number of Monte Carlo\nsamples, $M$, increases. Our analysis reveals that the NIS approximation errors\nconverge to 0 almost surely and with the optimal Monte Carlo rate of\n$M^{-\\frac{1}{2}}$. Moreover, we prove that this is achieved even when the mean\nestimation error of the IWs remains constant, a property that has been termed\n{\\it exact approximation} in the Markov chain Monte Carlo literature. We\nillustrate these theoretical results by means of a computer simulation example\ninvolving the estimation of the parameters of a state-space model typically\nused for target tracking.\n",
"title": "Analysis of a nonlinear importance sampling scheme for Bayesian parameter estimation in state-space models"
}
| null | null | null | null | true | null |
13117
| null |
Default
| null | null |
null |
{
"abstract": " The multiplicative theory of a set of numbers (which could be natural,\ninteger, rational, real or complex numbers) is the first-order theory of the\nstructure of that set with (solely) the multiplication operation (that set is\ntaken to be multiplicative, i.e., closed under multiplication). In this paper\nwe study the multiplicative theories of the complex, real and (positive)\nrational numbers. These theories (and also the multiplicative theories of\nnatural and integer numbers) are known to be decidable (i.e., there exists an\nalgorithm that decides whether a given sentence is derivable form the theory);\nhere we present explicit axiomatizations for them and show that they are not\nfinitely axiomatizable. For each of these sets (of complex, real and [positive]\nrational numbers) a language, including the multiplication operation, is\nintroduced in a way that it allows quantifier elimination (for the theory of\nthat set).\n",
"title": "On Axiomatizability of the Multiplicative Theory of Numbers"
}
| null | null | null | null | true | null |
13118
| null |
Default
| null | null |
null |
{
"abstract": " Recently, Lee and Cha (2015, `On two generalized classes of discrete\nbivariate distributions', {\\it American Statistician}, 221 - 230) proposed two\ngeneral classes of discrete bivariate distributions. They have discussed some\ngeneral properties and some specific cases of their proposed distributions. In\nthis paper we have considered one model, namely bivariate discrete Weibull\ndistribution, which has not been considered in the literature yet. The proposed\nbivariate discrete Weibull distribution is a discrete analogue of the\nMarshall-Olkin bivariate Weibull distribution. We study various properties of\nthe proposed distribution and discuss its interesting physical interpretations.\nThe proposed model has four parameters, and because of that it is a very\nflexible distribution. The maximum likelihood estimators of the parameters\ncannot be obtained in closed forms, and we have proposed a very efficient\nnested EM algorithm which works quite well for discrete data. We have also\nproposed augmented Gibbs sampling procedure to compute Bayes estimates of the\nunknown parameters based on a very flexible set of priors. Two data sets have\nbeen analyzed to show how the proposed model and the method work in practice.\nWe will see that the performances are quite satisfactory. Finally, we conclude\nthe paper.\n",
"title": "On Bivariate Discrete Weibull Distribution"
}
| null | null | null | null | true | null |
13119
| null |
Default
| null | null |
null |
{
"abstract": " Let G be the group of rational points of a reductive connected group over a\nfinite field (resp. nonarchimedean local field of characteristic p) and R a\ncommutative ring. The unipotent (resp. pro-p Iwahori) invariant functor takes a\nsmooth representation of G to a module over the unipotent (resp. pro-p Iwahori)\nHecke R-algebra H of G. We prove that these functors for G and for a Levi\nsubgroup of G commute with the parabolic induction functors, as well as with\nthe right adjoints of the parabolic induction functors. However, they do not\ncommute with the left adjoints of the parabolic induction functors in general;\nthey do if p is invertible in R.\nWhen R is an algebraically closed field of characteristic p, we show in the\nlocal case that an irreducible admissible R-representation V of G is\nsupercuspidal (or equivalently supersingular) if and only if the H-module V^I\nof its invariants by the pro-p Iwahori I admits a supersingular subquotient, if\nand only if V^I is supersingular.\n",
"title": "Parabolic induction in characteristic p"
}
| null | null | null | null | true | null |
13120
| null |
Default
| null | null |
null |
{
"abstract": " We introduce two applications of polygraphs to categorification problems. We\ncompute first, from a coherent presentation of an $n$-category, a coherent\npresentation of its Karoubi envelope. For this, we extend the construction of\nKaroubi envelope to $n$-polygraphs and linear $(n,n-1)$-polygraphs. The second\nproblem treated in this paper is the construction of Grothendieck\ndecategorifications for $(n,n-1)$-polygraphs. This construction yields a\nrewriting system presenting for example algebras categorified by a linear\nmonoidal category. We finally link quasi-convergence of such rewriting systems\nto the uniqueness of direct sum decompositions for linear\n$(n-1,n-1)$-categories.\n",
"title": "Linear polygraphs applied to categorification"
}
| null | null | null | null | true | null |
13121
| null |
Default
| null | null |
null |
{
"abstract": " Matter bounces refer to scenarios wherein the universe contracts at early\ntimes as in a matter dominated epoch until the scale factor reaches a minimum,\nafter which it starts expanding. While such scenarios are known to lead to\nscale invariant spectra of primordial perturbations after the bounce, the\nchallenge has been to construct completely symmetric bounces that lead to a\ntensor-to-scalar ratio which is small enough to be consistent with the recent\ncosmological data. In this work, we construct a model involving two scalar\nfields (a canonical field and a non-canonical ghost field) to drive the\nsymmetric matter bounce and study the evolution of the scalar perturbations in\nthe model. If we consider the scale associated with the bounce to be of the\norder of the Planck scale, the model is completely described in terms of only\none parameter, viz the value of the scale factor at the bounce. We evolve the\nscalar perturbations numerically across the bounce and evaluate the scalar\npower spectra after the bounce. We show that, while the scalar and tensor\nperturbation spectra are scale invariant over scales of cosmological interest,\nthe tensor-to-scalar ratio proves to be much smaller than the current upper\nbound from the observations of the cosmic microwave background anisotropies by\nthe Planck mission. We also support our numerical analysis with analytical\narguments.\n",
"title": "Viable tensor-to-scalar ratio in a symmetric matter bounce"
}
| null | null | null | null | true | null |
13122
| null |
Default
| null | null |
null |
{
"abstract": " Approximate ripple carry adders (RCAs) and carry lookahead adders (CLAs) are\npresented which are compared with accurate RCAs and CLAs for performing a\n32-bit addition. The accurate and approximate RCAs and CLAs are implemented\nusing a 32/28nm CMOS process. Approximations ranging from 4- to 20-bits are\nconsidered for the less significant adder bit positions. The simulation results\nshow that approximate RCAs report reductions in the power-delay product (PDP)\nranging from 19.5% to 82% than the accurate RCA for approximation sizes varying\nfrom 4- to 20-bits. Also, approximate CLAs report reductions in PDP ranging\nfrom 16.7% to 74.2% than the accurate CLA for approximation sizes varying from\n4- to 20-bits. On average, for the approximation sizes considered, it is\nobserved that approximate CLAs achieve a 46.5% reduction in PDP compared to the\napproximate RCAs. Hence, approximate CLAs are preferable over approximate RCAs\nfor the low power implementation of approximate computer arithmetic.\n",
"title": "Approximate Ripple Carry and Carry Lookahead Adders - A Comparative Analysis"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13123
| null |
Validated
| null | null |
null |
{
"abstract": " The normality assumption on data set is very restrictive approach for\nmodelling. The generalized form of normal distribution, named as an exponential\npower (EP) distribution, and its scale mixture form have been considered\nextensively to overcome the problem for modelling non-normal data set since\nlast decades. However, examining the robustness properties of maximum\nlikelihood (ML) estimators of parameters in these distributions, such as the in\nuence function, gross-error sensitivity, breakdown point and\ninformation-standardized sensitivity, has not been considered together. The\nwell-known asymptotic properties of ML estimators of location, scale and added\nskewness parameters in EP and its scale mixture form distributions are studied\nand also these ML estimators for location, scale and scale variant (skewness)\nparameters can be represented as an iterative reweighting algorithm to compute\nthe estimates of these parameters simultaneously.\n",
"title": "On the Robustness and Asymptotic Properties for Maximum Likelihood Estimators of Parameters in Exponential Power and its Scale Mixture Form Distributions"
}
| null | null | null | null | true | null |
13124
| null |
Default
| null | null |
null |
{
"abstract": " This work proposes a novel solution to the problem of internal covariate\nshift and dying neurons using the concept of linked neurons. We define the\nneuron linkage in terms of two constraints: first, all neuron activations in\nthe linkage must have the same operating point. That is to say, all of them\nshare input weights. Secondly, a set of neurons is linked if and only if there\nis at least one member of the linkage that has a non-zero gradient in regard to\nthe input of the activation function. This means that for any input in the\nactivation function, there is at least one member of the linkage that operates\nin a non-flat and non-zero area. This simple change has profound implications\nin the network learning dynamics. In this article we explore the consequences\nof this proposal and show that by using this kind of units, internal covariate\nshift is implicitly solved. As a result of this, the use of linked neurons\nallows to train arbitrarily large networks without any architectural or\nalgorithmic trick, effectively removing the need of using re-normalization\nschemes such as Batch Normalization, which leads to halving the required\ntraining time. It also solves the problem of the need for standarized input\ndata. Results show that the units using the linkage not only do effectively\nsolve the aforementioned problems, but are also a competitive alternative with\nrespect to state-of-the-art with very promising results.\n",
"title": "Solving internal covariate shift in deep learning with linked neurons"
}
| null | null | null | null | true | null |
13125
| null |
Default
| null | null |
null |
{
"abstract": " This work is concerned with Al/Al-oxide(AlO$_{x}$)/Al-layer systems which are\nimportant for Josephson-junction-based superconducting devices such as quantum\nbits. The device performance is limited by noise, which has been to a large\ndegree assigned to the presence and properties of two-level tunneling systems\nin the amorphous AlO$_{x}$ tunnel barrier. The study is focused on the\ncorrelation of the fabrication conditions, nanostructural and nanochemical\nproperties and the occurrence of two-level tunneling systems with particular\nemphasis on the AlO$_{x}$-layer. Electron-beam evaporation with two different\nprocesses and sputter deposition were used for structure fabrication, and the\neffect of illumination by ultraviolet light during Al-oxide formation is\nelucidated. Characterization was performed by analytical transmission electron\nmicroscopy and low-temperature dielectric measurements. We show that the\nfabrication conditions have a strong impact on the nanostructural and\nnanochemical properties of the layer systems and the properties of two-level\ntunneling systems. Based on the understanding of the observed structural\ncharacteristics, routes are derived towards the fabrication of\nAl/AlO$_{x}$/Al-layers systems with improved properties.\n",
"title": "Correlating the nanostructure of Al-oxide with deposition conditions and dielectric contributions of two-level systems in perspective of superconducting quantum circuits"
}
| null | null | null | null | true | null |
13126
| null |
Default
| null | null |
null |
{
"abstract": " Senior project is a typical essential course in computing educational\nprograms. The course involves the selection of a project problem, the\nsubmission of various documents, and intensive communication among the project\nteam members and between them and the course instructors. To facilitate all\nthese tasks, we introduce the senior project management system (SPMS) that\norganizes and manages previous, current, and proposed senior projects in all of\ntheir stages along with proper ways of communication between the students and\ncourse instructors. This paper explains the system requirements and\nspecifications and discusses related design issues. The paper shows the\nimportance of well documenting the specifications and requirements of software\nsystems and paying considerable attention to system design, which has a\npositive impact on implementing high quality systems.\n",
"title": "Senior Project Management System: Requirements, Specification, and Design Issues"
}
| null | null | null | null | true | null |
13127
| null |
Default
| null | null |
null |
{
"abstract": " The study of tensor network theory is an important field and promises a wide\nrange of experimental and quantum information theoretical applications. Matrix\nproduct state is the most well-known example of tensor network states, which\nprovides an effective and efficient representation of one-dimensional quantum\nsystems. Indeed, it lies at the heart of density matrix renormalization group\n(DMRG), a most common method for simulation of one-dimensional strongly\ncorrelated quantum systems. It has got attention from several areas varying\nfrom solid-state systems to quantum computing and quantum simulators. We have\nconsidered maximally entangled matrix product states (GHZ and W). Here, we\ndesigned the quantum circuits for implementing the matrix product states. In\nthis paper, we simulated the matrix product states in customized IBM (2-qubit,\n3-qubit and 4-qubit) quantum systems and determined the probability\ndistribution among the quantum states.\n",
"title": "Simulation of Matrix Product State on a Quantum Computer"
}
| null | null | null | null | true | null |
13128
| null |
Default
| null | null |
null |
{
"abstract": " The broad set of deep generative models (DGMs) has achieved remarkable\nadvances. However, it is often difficult to incorporate rich structured domain\nknowledge with the end-to-end DGMs. Posterior regularization (PR) offers a\nprincipled framework to impose structured constraints on probabilistic models,\nbut has limited applicability to the diverse DGMs that can lack a Bayesian\nformulation or even explicit density evaluation. PR also requires constraints\nto be fully specified a priori, which is impractical or suboptimal for complex\nknowledge with learnable uncertain parts. In this paper, we establish\nmathematical correspondence between PR and reinforcement learning (RL), and,\nbased on the connection, expand PR to learn constraints as the extrinsic reward\nin RL. The resulting algorithm is model-agnostic to apply to any DGMs, and is\nflexible to adapt arbitrary constraints with the model jointly. Experiments on\nhuman image generation and templated sentence generation show models with\nlearned knowledge constraints by our algorithm greatly improve over base\ngenerative models.\n",
"title": "Deep Generative Models with Learnable Knowledge Constraints"
}
| null | null | null | null | true | null |
13129
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we propose an unsupervised face clustering algorithm called\n\"Proximity-Aware Hierarchical Clustering\" (PAHC) that exploits the local\nstructure of deep representations. In the proposed method, a similarity measure\nbetween deep features is computed by evaluating linear SVM margins. SVMs are\ntrained using nearest neighbors of sample data, and thus do not require any\nexternal training data. Clusters are then formed by thresholding the similarity\nscores. We evaluate the clustering performance using three challenging\nunconstrained face datasets, including Celebrity in Frontal-Profile (CFP),\nIARPA JANUS Benchmark A (IJB-A), and JANUS Challenge Set 3 (JANUS CS3)\ndatasets. Experimental results demonstrate that the proposed approach can\nachieve significant improvements over state-of-the-art methods. Moreover, we\nalso show that the proposed clustering algorithm can be applied to curate a set\nof large-scale and noisy training dataset while maintaining sufficient amount\nof images and their variations due to nuisance factors. The face verification\nperformance on JANUS CS3 improves significantly by finetuning a DCNN model with\nthe curated MS-Celeb-1M dataset which contains over three million face images.\n",
"title": "A Proximity-Aware Hierarchical Clustering of Faces"
}
| null | null | null | null | true | null |
13130
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a new acoustic emission (AE) source localization for\nisotropic plates with reflecting boundaries. This approach that has no blind\nspot leverages multimodal edge reflections to identify AE sources with only a\nsingle sensor. The implementation of the proposed approach involves three main\nsteps. First, the continuous wavelet transform (CWT) and the dispersion curves\nof the fundamental Lamb wave modes are utilized to estimate the distance\nbetween an AE source and a sensor. This step uses a modal acoustic emission\napproach. Then, an analytical model is proposed that uses the estimated\ndistances to simulate the edge-reflected waves. Finally, the correlation\nbetween the experimental and the simulated waveforms is used to estimate the\nlocation of AE sources. Hsu-Nielson pencil lead break (PLB) tests were\nperformed on an aluminum plate to validate this algorithm and promising results\nwere achieved. Based on these results, the paper reports the statistics of the\nlocalization errors.\n",
"title": "Acoustic emission source localization in thin metallic plates: a single-sensor approach based on multimodal edge reflections"
}
| null | null | null | null | true | null |
13131
| null |
Default
| null | null |
null |
{
"abstract": " Anderson's paving conjecture, now known to hold due to the resolution of the\nKadison-Singer problem asserts that every zero diagonal Hermitian matrix admits\nnon-trivial pavings with dimension independent bounds. In this paper, we\ndevelop a technique extending the arguments of Marcus, Spielman and Srivastava\nin their solution of the Kadison-Singer problem to show the existence of\nnon-trivial pavings for collections of matrices. We show that given zero\ndiagonal Hermitian contractions $A^{(1)}, \\cdots, A^{(k)} \\in M_n(\\mathbb{C})$\nand $\\epsilon > 0$, one may find a paving $X_1 \\amalg \\cdots \\amalg X_r = [n]$\nwhere $r \\leq 18k\\epsilon^{-2}$ such that, \\[\\lambda_{max} (P_{X_i} A^{(j)}\nP_{X_i}) < \\epsilon, \\quad i \\in [r], \\, j \\in [k].\\] As a consequence, we get\nthe correct asymptotic estimates for paving general zero diagonal matrices;\nzero diagonal contractions can be $(O(\\epsilon^{-2}),\\epsilon)$ paved. As an\napplication, we give a simplified proof wth slightly better estimates of a\ntheorem of Johnson, Ozawa and Schechtman concerning commutator representations\nof zero trace matrices.\n",
"title": "Asymptotically Optimal Multi-Paving"
}
| null | null | null | null | true | null |
13132
| null |
Default
| null | null |
null |
{
"abstract": " We describe the AMReX suite of astrophysics codes and their application to\nmodeling problems in stellar astrophysics. Maestro is tuned to efficiently\nmodel subsonic convective flows while Castro models the highly compressible\nflows associated with stellar explosions. Both are built on the\nblock-structured adaptive mesh refinement library AMReX. Together, these codes\nenable a thorough investigation of stellar phenomena, including Type Ia\nsupernovae and X-ray bursts. We describe these science applications and the\napproach we are taking to make these codes performant on current and future\nmany-core and GPU-based architectures.\n",
"title": "Meeting the Challenges of Modeling Astrophysical Thermonuclear Explosions: Castro, Maestro, and the AMReX Astrophysics Suite"
}
| null | null | null | null | true | null |
13133
| null |
Default
| null | null |
null |
{
"abstract": " We present a translation of the Lambek calculus with brackets and the unit\nconstant, $\\mathbf{Lb}^{\\boldsymbol{*}}_{\\mathbf{1}}$, into the Lambek calculus\nwith brackets allowing empty antecedents, but without the unit constant,\n$\\mathbf{Lb}^{\\boldsymbol{*}}$. Using this translation, we extend previously\nknown results for $\\mathbf{Lb}^{\\boldsymbol{*}}$ to\n$\\mathbf{Lb}^{\\boldsymbol{*}}_{\\mathbf{1}}$: (1) languages generated by\ncategorial grammars based on the Lambek calculus with brackets are context-free\n(Kanazawa 2017); (2) the polynomial-time algorithm for deciding derivability of\nbounded depth sequents (Kanovich et al. 2017).\n",
"title": "Eliminating the unit constant in the Lambek calculus with brackets"
}
| null | null | null | null | true | null |
13134
| null |
Default
| null | null |
null |
{
"abstract": " We present a distributed formation control strategy for agents with a variety\nof dynamics to achieve a desired planar formation. The proposed strategy is\nfully distributed, does not require inter-agent communication or a common sense\nof orientation, and can be implemented using relative position measurements\nacquired by agents in their local coordinate frames. We show how the control\ndesigned for agents with the simplest dynamical model, i.e., the\nsingle-integrator dynamics, can be extended to holonomic agents with\nhigher-order dynamics such as quadrotors, and nonholonomic agents such as\nunicycles and cars. We prove that the proposed strategy is robust to\nsaturations in the input, unmodeled dynamics, and switches in the sensing\ntopology. We further show that the control is relaxed in the sense that agents\ncan move along a rotated and scaled control direction without affecting the\nconvergence to the desired formation. This observation is used to design a\ndistributed collision avoidance strategy. We demonstrate the proposed approach\nin simulations and further present a distributed robotic platform to test the\nstrategy experimentally. Our experimental platform consists of off-the-shelf\nequipment that can be used to test and validate other multi-agent algorithms.\nThe code and implementation instructions for this platform are available online\nand free.\n",
"title": "Robust Distributed Planar Formation Control for Higher-Order Holonomic and Nonholonomic Agents"
}
| null | null | null | null | true | null |
13135
| null |
Default
| null | null |
null |
{
"abstract": " In recent years, RTB(Real Time Bidding) becomes a popular online\nadvertisement trading method. During the auction, each DSP(Demand Side\nPlatform) is supposed to evaluate current opportunity and respond with an ad\nand corresponding bid price. It's essential for DSP to find an optimal ad\nselection and bid price determination strategy which maximizes revenue or\nperformance under budget and ROI(Return On Investment) constraints in P4P(Pay\nFor Performance) or P4U(Pay For Usage) mode. We solve this problem by 1)\nformalizing the DSP problem as a constrained optimization problem, 2) proposing\nthe augmented MMKP(Multi-choice Multi-dimensional Knapsack Problem) with\ngeneral solution, 3) and demonstrating the DSP problem is a special case of the\naugmented MMKP and deriving specialized strategy. Our strategy is verified\nthrough simulation and outperforms state-of-the-art strategies in real\napplication. To the best of our knowledge, our solution is the first dual based\nDSP bidding framework that is derived from strict second price auction\nassumption and generally applicable to the multiple ads scenario with various\nobjectives and constraints.\n",
"title": "Dual Based DSP Bidding Strategy and its Application"
}
| null | null | null | null | true | null |
13136
| null |
Default
| null | null |
null |
{
"abstract": " Graph-structured data such as social networks, functional brain networks,\ngene regulatory networks, communications networks have brought the interest in\ngeneralizing deep learning techniques to graph domains. In this paper, we are\ninterested to design neural networks for graphs with variable length in order\nto solve learning problems such as vertex classification, graph classification,\ngraph regression, and graph generative tasks. Most existing works have focused\non recurrent neural networks (RNNs) to learn meaningful representations of\ngraphs, and more recently new convolutional neural networks (ConvNets) have\nbeen introduced. In this work, we want to compare rigorously these two\nfundamental families of architectures to solve graph learning tasks. We review\nexisting graph RNN and ConvNet architectures, and propose natural extension of\nLSTM and ConvNet to graphs with arbitrary size. Then, we design a set of\nanalytically controlled experiments on two basic graph problems, i.e. subgraph\nmatching and graph clustering, to test the different architectures. Numerical\nresults show that the proposed graph ConvNets are 3-17% more accurate and\n1.5-4x faster than graph RNNs. Graph ConvNets are also 36% more accurate than\nvariational (non-learning) techniques. Finally, the most effective graph\nConvNet architecture uses gated edges and residuality. Residuality plays an\nessential role to learn multi-layer architectures as they provide a 10% gain of\nperformance.\n",
"title": "Residual Gated Graph ConvNets"
}
| null | null | null | null | true | null |
13137
| null |
Default
| null | null |
null |
{
"abstract": " This paper describes a novel storyboarding scheme that uses a model trained\non pairwise image comparisons to identify images likely to be of interest to a\nmobile robot user. Traditional storyboarding schemes typically attempt to\nsummarise robot observations using predefined novelty or image quality\nobjectives, but we propose a user training stage that allows the incorporation\nof user interest when storyboarding. Our approach dramatically reduces the\nnumber of image comparisons required to infer image interest by applying a\nGaussian process smoothing algorithm on image features extracted using a\npre-trained convolutional neural network. As a particularly valuable\nby-product, the proposed approach allows the generation of user-specific\nsaliency or attention maps.\n",
"title": "User-driven mobile robot storyboarding: Learning image interest and saliency from pairwise image comparisons"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13138
| null |
Validated
| null | null |
null |
{
"abstract": " The aim of my dissertation is to investigate habitability in extra-Solar\nSystems. Most of the time, only planets are considered as possible places where\nextraterrestrial life can emerge and evolve, however, their moons could be\ninhabited, too. I present a comprehensive study, which considers habitability\nnot only on planets, but on satellites, as well. My research focuses on three\nclosely related topics. The first one is the circumstellar habitable zone,\nwhich is usually used as a first proxy for determining the habitability of a\nplanet around the host star. The word habitability is used in the sense that\nliquid water, which is essential for life as we know it, may be present on the\nplanetary surface. Whether the planet is habitable or not, its moon might have\na suitable surface temperature for holding water reservoirs, providing that\ntidal heating is in action. Tidal heating is generated inside the satellite and\nits source is the strong gravitational force of the nearby planet. The second\ntopic of my research explores tidal heating and the habitability of extra-solar\nmoons with and without stellar radiation and other related energy sources. Life\nis possible to form even on icy planetary bodies, inside tidally heated\nsubsurface oceans. The third topic probes the possibility of identifying an\nice-covered satellite from photometric observations. A strong indication of\nsurface ice is the high reflectance of the body, which may be measured when the\nmoon disappears behind the host star, so its reflected light is blocked out by\nthe star.\n",
"title": "Habitability of Exoplanetary Systems"
}
| null | null | null | null | true | null |
13139
| null |
Default
| null | null |
null |
{
"abstract": " As part of a large investigation with Hubble Space Telescope to study the\nfaintest stars within the globular cluster Omega Centauri, in this work we\npresent early results on the multiplicity of its main sequence (MS) stars,\nbased on deep optical and near-infrared observations. By using appropriate\ncolor-magnitude diagrams we have identified, for the first time, the two main\nstellar populations I, and II along the entire MS, from the turn-off towards\nthe hydrogen-burning limit. We have compared the observations with suitable\nsynthetic spectra of MS stars and conclude that the two MSs are consistent with\nstellar populations with different metallicity, helium, and light-element\nabundance. Specifically, MS-I corresponds to a metal-poor stellar population\n([Fe/H]~-1.7) with Y~ 0.25 and [O/Fe]~0.30. The MS-II hosts helium-rich\n(Y~0.37-0.40) stars with metallicity ranging from [Fe/H]~-1.7 to -1.4. Below\nthe MS knee (mF160W~19.5, our photometry reveals that each of the two main MSs\nhosts stellar subpopulations with different oxygen abundances, with very O-poor\nstars ([O/Fe]~-0.5) populating the MS-II. Such a complexity has never been\nobserved in previous studies of M-dwarfs in globular clusters. A few months\nbefore the lunch of the James Webb Space Telescope, these results demonstrate\nthe power of optical and near-infrared photometry in the study of multiple\nstellar populations in globular clusters.\n",
"title": "The HST Large Program on Omega Centauri. I. Multiple stellar populations at the bottom of the main sequence probed in NIR-Optical"
}
| null | null | null | null | true | null |
13140
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we provide a first analysis of the research questions that\narise when dealing with the problem of communicating pieces of formal\nargumentation through natural language interfaces. It is a generally held\nopinion that formal models of argumentation naturally capture human argument,\nand some preliminary studies have focused on justifying this view.\nUnfortunately, the results are not only inconclusive, but seem to suggest that\nexplaining formal argumentation to humans is a rather articulated task.\nGraphical models for expressing argumentation-based reasoning are appealing,\nbut often humans require significant training to use these tools effectively.\nWe claim that natural language interfaces to formal argumentation systems offer\na real alternative, and may be the way forward for systems that capture human\nargument.\n",
"title": "On Natural Language Generation of Formal Argumentation"
}
| null | null | null | null | true | null |
13141
| null |
Default
| null | null |
null |
{
"abstract": " Many recent deep learning platforms rely on third-party libraries (such as\ncuBLAS) to utilize the computing power of modern hardware accelerators (such as\nGPUs). However, we observe that they may achieve suboptimal performance because\nthe library functions are not used appropriately. In this paper, we target at\noptimizing the operations of multiplying a matrix with the transpose of another\nmatrix (referred to as NT operation hereafter), which contribute about half of\nthe training time of fully connected deep neural networks. Rather than directly\ncalling the library function, we propose a supervised learning based algorithm\nselection approach named MTNN, which uses a gradient boosted decision tree to\nselect one from two alternative NT implementations intelligently: (1) calling\nthe cuBLAS library function; (2) calling our proposed algorithm TNN that uses\nan efficient out-of-place matrix transpose. We evaluate the performance of MTNN\non two modern GPUs: NVIDIA GTX 1080 and NVIDIA Titan X Pascal. MTNN can achieve\n96\\% of prediction accuracy with very low computational overhead, which results\nin an average of 54\\% performance improvement on a range of NT operations. To\nfurther evaluate the impact of MTNN on the training process of deep neural\nnetworks, we have integrated MTNN into a popular deep learning platform Caffe.\nOur experimental results show that the revised Caffe can outperform the\noriginal one by an average of 28\\%. Both MTNN and the revised Caffe are\nopen-source.\n",
"title": "Supervised Learning Based Algorithm Selection for Deep Neural Networks"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13142
| null |
Validated
| null | null |
null |
{
"abstract": " A significant amount of search queries originate from some real world\ninformation need or tasks. In order to improve the search experience of the end\nusers, it is important to have accurate representations of tasks. As a result,\nsignificant amount of research has been devoted to extracting proper\nrepresentations of tasks in order to enable search systems to help users\ncomplete their tasks, as well as providing the end user with better query\nsuggestions, for better recommendations, for satisfaction prediction, and for\nimproved personalization in terms of tasks. Most existing task extraction\nmethodologies focus on representing tasks as flat structures. However, tasks\noften tend to have multiple subtasks associated with them and a more\nnaturalistic representation of tasks would be in terms of a hierarchy, where\neach task can be composed of multiple (sub)tasks. To this end, we propose an\nefficient Bayesian nonparametric model for extracting hierarchies of such tasks\n\\& subtasks. We evaluate our method based on real world query log data both\nthrough quantitative and crowdsourced experiments and highlight the importance\nof considering task/subtask hierarchies.\n",
"title": "Extracting Hierarchies of Search Tasks & Subtasks via a Bayesian Nonparametric Approach"
}
| null | null | null | null | true | null |
13143
| null |
Default
| null | null |
null |
{
"abstract": " We prove the unpolarized Shafarevich conjecture for K3 surfaces: the set of\nisomorphism classes of K3 surfaces over a fixed number field with good\nreduction away from a fixed and finite set of places is finite. Our proof is\nbased on the theorems of Faltings and André, as well as the Kuga-Satake\nconstruction.\n",
"title": "The unpolarized Shafarevich Conjecture for K3 Surfaces"
}
| null | null | null | null | true | null |
13144
| null |
Default
| null | null |
null |
{
"abstract": " Computing steady-state distributions in infinite-state stochastic systems is\nin general a very dificult task. Product-form Petri nets are those Petri nets\nfor which the steady-state distribution can be described as a natural product\ncorresponding, up to a normalising constant, to an exponentiation of the\nmarkings. However, even though some classes of nets are known to have a\nproduct-form distribution, computing the normalising constant can be hard. The\nclass of (closed) {\\Pi}3-nets has been proposed in an earlier work, for which\nit is shown that one can compute the steady-state distribution efficiently.\nHowever these nets are bounded. In this paper, we generalise queuing Markovian\nnetworks and closed {\\Pi}3-nets to obtain the class of open {\\Pi}3-nets, that\ngenerate infinite-state systems. We show interesting properties of these nets:\n(1) we prove that liveness can be decided in polynomial time, and that\nreachability in live {\\Pi}3-nets can be decided in polynomial time; (2) we show\nthat we can decide ergodicity of such nets in polynomial time as well; (3) we\nprovide a pseudo-polynomial time algorithm to compute the normalising constant.\n",
"title": "Unbounded product-form Petri nets"
}
| null | null | null | null | true | null |
13145
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a discrete-time option pricing model that is rooted in\nReinforcement Learning (RL), and more specifically in the famous Q-Learning\nmethod of RL. We construct a risk-adjusted Markov Decision Process for a\ndiscrete-time version of the classical Black-Scholes-Merton (BSM) model, where\nthe option price is an optimal Q-function. Pricing is done by learning to\ndynamically optimize risk-adjusted returns for an option replicating portfolio,\nas in the Markowitz portfolio theory. Using Q-Learning and related methods,\nonce created in a parametric setting, the model is able to go model-free and\nlearn to price and hedge an option directly from data generated from a dynamic\nreplicating portfolio which is rebalanced at discrete times. If the world is\naccording to BSM, our risk-averse Q-Learner converges, given enough training\ndata, to the true BSM price and hedge ratio of the option in the continuous\ntime limit, even if hedges applied at the stage of data generation are\ncompletely random (i.e. it can learn the BSM model itself, too!), because\nQ-Learning is an off-policy algorithm. If the world is different from a BSM\nworld, the Q-Learner will find it out as well, because Q-Learning is a\nmodel-free algorithm. For finite time steps, the Q-Learner is able to\nefficiently calculate both the optimal hedge and optimal price for the option\ndirectly from trading data, and without an explicit model of the world. This\nsuggests that RL may provide efficient data-driven and model-free methods for\noptimal pricing and hedging of options, once we depart from the academic\ncontinuous-time limit, and vice versa, option pricing methods developed in\nMathematical Finance may be viewed as special cases of model-based\nReinforcement Learning. Our model only needs basic linear algebra (plus Monte\nCarlo simulation, if we work with synthetic data).\n",
"title": "QLBS: Q-Learner in the Black-Scholes(-Merton) Worlds"
}
| null | null | null | null | true | null |
13146
| null |
Default
| null | null |
null |
{
"abstract": " The performance of single channel source separation algorithms has improved\ngreatly in recent times with the development and deployment of neural networks.\nHowever, many such networks continue to operate on the magnitude spectrogram of\na mixture, and produce an estimate of source magnitude spectrograms, to perform\nsource separation. In this paper, we interpret these steps as additional neural\nnetwork layers and propose an end-to-end source separation network that allows\nus to estimate the separated speech waveform by operating directly on the raw\nwaveform of the mixture. Furthermore, we also propose the use of masking based\nend-to-end separation networks that jointly optimize the mask and the latent\nrepresentations of the mixture waveforms. These networks show a significant\nimprovement in separation performance compared to existing architectures in our\nexperiments. To train these end-to-end models, we investigate the use of\ncomposite cost functions that are derived from objective evaluation metrics as\nmeasured on waveforms. We present subjective listening test results that\ndemonstrate the improvement attained by using masking based end-to-end networks\nand also reveal insights into the performance of these cost functions for\nend-to-end source separation.\n",
"title": "End-to-end Networks for Supervised Single-channel Speech Separation"
}
| null | null | null | null | true | null |
13147
| null |
Default
| null | null |
null |
{
"abstract": " We examine the problem of searching sequentially for a desired real value (a\nkey) within a sorted unbalanced three-dimensional finite real array. This\nclassic problem can be viewed as determining the correct dimensional threshold\nfunction from a finite class of such functions within the array, based on\nsequential queries that take the form of point samples. This note addresses the\nchallenge of constructing algorithms that require the minimum number of queries\nnecessary in the worst case, to search for a given key in arrays that have\nthree dimensions with sizes that are not necessarily equal.\n",
"title": "A note on searching sorted unbalanced three-dimensional arrays"
}
| null | null | null | null | true | null |
13148
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we classify the isomorphism classes of four dimensional\nnilpotent associative algebras over a field F, studying regular subgroups of\nthe affine group AGL_4(F). In particular we provide explicit representatives\nfor such classes when F is a finite field, the real field R or an algebraically\nclosed field.\n",
"title": "Isomorphism classes of four dimensional nilpotent associative algebras over a field"
}
| null | null | null | null | true | null |
13149
| null |
Default
| null | null |
null |
{
"abstract": " The current article unveils and analyzes some important factors, influencing\ndiversity in strategic decision-making approaches in local companies.\nResearcher's attention is oriented to survey important characteristics of the\nstrategic moves, undertaken by leading companies in Bulgaria.\n",
"title": "Contemporary facets of business successes among leading companies, operating in Bulgaria"
}
| null | null |
[
"Quantitative Finance"
] | null | true | null |
13150
| null |
Validated
| null | null |
null |
{
"abstract": " Part-based representation has been proven to be effective for a variety of\nvisual applications. However, automatic discovery of discriminative parts\nwithout object/part-level annotations is challenging. This paper proposes a\ndiscriminative mid-level representation paradigm based on the responses of a\ncollection of part detectors, which only requires the image-level labels.\nTowards this goal, we first develop a detector-based spectral clustering method\nto mine the representative and discriminative mid-level patterns for detector\ninitialization. The advantage of the proposed pattern mining technology is that\nthe distance metric based on detectors only focuses on discriminative details,\nand a set of such grouped detectors offer an effective way for consistent\npattern mining. Relying on the discovered patterns, we further formulate the\ndetector learning process as a confidence-loss sparse Multiple Instance\nLearning (cls-MIL) task, which considers the diversity of the positive samples,\nwhile avoid drifting away the well localized ones by assigning a confidence\nvalue to each positive sample. The responses of the learned detectors can form\nan effective mid-level image representation for both image classification and\nobject localization. Experiments conducted on benchmark datasets demonstrate\nthe superiority of our method over existing approaches.\n",
"title": "Ensemble of Part Detectors for Simultaneous Classification and Localization"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13151
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we present a novel approach based on the random walk process\nfor finding meaningful representations of a graph model. Our approach leverages\nthe transient behavior of many short random walks with novel initialization\nmechanisms to generate model discriminative features. These features are able\nto capture a more comprehensive structural signature of the underlying graph\nmodel. The resulting representation is invariant to both node permutation and\nthe size of the graph, allowing direct comparison between large classes of\ngraphs. We test our approach on two challenging model selection problems: the\ndiscrimination in the sparse regime of an Erdös-Renyi model from a\nstochastic block model and the planted clique problem. Our representation\napproach achieves performance that closely matches known theoretical limits in\naddition to being computationally simple and scalable to large graphs.\n",
"title": "Graph Model Selection via Random Walks"
}
| null | null | null | null | true | null |
13152
| null |
Default
| null | null |
null |
{
"abstract": " We provide a non-trivial measure of irrationality for a class of Mahler\nnumbers defined with infinite products which cover the Thue-Morse constant.\n",
"title": "On irrationality measure of Thue-Morse constant"
}
| null | null |
[
"Mathematics"
] | null | true | null |
13153
| null |
Validated
| null | null |
null |
{
"abstract": " We consider slowly evolving, i.e. ADIABATIC, operational regime within a\ntransmission level (continental scale) natural gas pipeline system. This allows\nus to introduce a set of nodal equations of reduced complexity describing gas\ntransients in injection/consumption UNBALANCED (so-called line-pack) cases. We\ndiscuss, in details, construction of the UNBALANCED ADIABATIC (UA)\napproximation on the basic example of a single pipe. The UA approximation is\nexpected to play a significant \"model reduction\" role in solving control,\noptimization and planning problems relevant for flawless functioning of modern\nnatural gas networks.\n",
"title": "Adiabatic approach for natural gas pipeline computations"
}
| null | null | null | null | true | null |
13154
| null |
Default
| null | null |
null |
{
"abstract": " Although shill bidding is a common auction fraud, it is however very tough to\ndetect. Due to the unavailability and lack of training data, in this study, we\nbuild a high-quality labeled shill bidding dataset based on recently collected\nauctions from eBay. Labeling shill biding instances with multidimensional\nfeatures is a critical phase for the fraud classification task. For this\npurpose, we introduce a new approach to systematically label the fraud data\nwith the help of the hierarchical clustering CURE that returns remarkable\nresults as illustrated in the experiments.\n",
"title": "Clustering and Labelling Auction Fraud Data"
}
| null | null | null | null | true | null |
13155
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the evolution of decorrelation bandwidth of inter core\ncrosstalk (IC-XT) in homogeneous weakly coupled multicore fibers (WC-MCFs). The\nmodified mode-coupled equations (MCEs) are numerically solved by combining the\nfourth order Runge-Kutta method and compound Simpson integral method. It can be\ntheoretically and numerically observed that the decorrelation bandwidth of\nIC-XT decreases with transmission distance by fractional linear function. The\nevolution rule of IC-XT's decorrelation bandwidth is further confirmed by\nexperiments, which can be used as an evaluation criterion for channel model.\nFinally, we propose a new channel model with the coupling matrix of IC-XT\ngenerated automatically by phase transfer function (PTF), which is in good\nagreement with the above evaluation criterion. We believe the proposed channel\nmodel can provide a good simulation platform for homogeneous WC-MCF based\ncommunication systems.\n",
"title": "Investigation of channel model for weakly coupled multicore fiber"
}
| null | null | null | null | true | null |
13156
| null |
Default
| null | null |
null |
{
"abstract": " In 1999 Lyngs{\\o} and Pedersen proposed a conjecture stating that every\nbinary circular word of length $n$ with equal number of zeros and ones has an\nantipalindromic linear subsequence of length at least $\\frac{2}{3}n$. No\nprogress over a trivial $\\frac{1}{2}n$ bound has been achieved since then. We\nsuggest a palindromic counterpart to this conjecture and provide a non-trivial\ninfinite series of circular words which prove the upper bound of $\\frac{2}{3}n$\nfor both conjectures at the same time. The construction also works for words\nover an alphabet of size $k$ and gives rise to a generalization of the\nconjecture by Lyngs{\\o} and Pedersen. Moreover, we discuss some possible\nstrengthenings and weakenings of the named conjectures. We also propose two\nsimilar conjectures for linear words and provide some evidences for them.\n",
"title": "Palindromic Subsequences in Finite Words"
}
| null | null | null | null | true | null |
13157
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we study the asymptotic properties of Bayesian multiple testing\nprocedures for a large class of Gaussian scale mixture pri- ors. We study two\ntypes of multiple testing risks: a Bayesian risk proposed in Bogdan et al.\n(2011) where the data are assume to come from a mixture of normal, and a\nfrequentist risk similar to the one proposed by Arias-Castro and Chen (2017).\nFollowing the work of van der Pas et al. (2016), we give general conditions on\nthe prior such that both risks can be bounded. For the Bayesian risk, the bound\nis almost sharp. This result show that under these conditions, the considered\nclass of continuous prior can be competitive with the usual two-group model\n(e.g. spike and slab priors). We also show that if the non-zeros component of\nthe parameter are large enough, the minimax risk can be made asymptotically\nnull. The separation rates obtained are consistent with the one that could be\nguessed from the existing literature (see van der Pas et al., 2017b). For both\nproblems, we then give conditions under which an adaptive version of the result\ncan be obtained.\n",
"title": "Risk quantification for the thresholding rule for multiple testing using Gaussian scale mixtures"
}
| null | null | null | null | true | null |
13158
| null |
Default
| null | null |
null |
{
"abstract": " A short account of recent existence and multiplicity theorems on the\nDirichlet problem for an elliptic equation with $(p,q)$-Laplacian in a bounded\ndomain is performed. Both eigenvalue problems and different types of\nperturbation terms are briefly discussed. Special attention is paid to possibly\ncoercive, resonant, subcritical, critical, or asymmetric right-hand sides.\n",
"title": "Some recent results on the Dirichlet problem for (p,q)-Laplace equations"
}
| null | null | null | null | true | null |
13159
| null |
Default
| null | null |
null |
{
"abstract": " A convex penalty for promoting switching controls for partial differential\nequations is introduced; such controls consist of an arbitrary number of\ncomponents of which at most one should be simultaneously active. Using a\nMoreau-Yosida approximation, a family of approximating problems is obtained\nthat is amenable to solution by a semismooth Newton method. The efficiency of\nthis approach and the structure of the obtained controls are demonstrated by\nnumerical examples.\n",
"title": "A convex penalty for switching control of partial differential equations"
}
| null | null | null | null | true | null |
13160
| null |
Default
| null | null |
null |
{
"abstract": " Blue Waters is a Petascale-level supercomputer whose mission is to enable the\nnational scientific and research community to solve \"grand challenge\" problems\nthat are orders of magnitude more complex than can be carried out on other high\nperformance computing systems. Given the important and unique role that Blue\nWaters plays in the U.S. research portfolio, it is important to have a detailed\nunderstanding of its workload in order to guide performance optimization both\nat the software and system configuration level as well as inform architectural\nbalance tradeoffs. Furthermore, understanding the computing requirements of the\nBlue Water's workload (memory access, IO, communication, etc.), which is\ncomprised of some of the most computationally demanding scientific problems,\nwill help drive changes in future computing architectures, especially at the\nleading edge. With this objective in mind, the project team carried out a\ndetailed workload analysis of Blue Waters.\n",
"title": "Workload Analysis of Blue Waters"
}
| null | null | null | null | true | null |
13161
| null |
Default
| null | null |
null |
{
"abstract": " Consider a time series of measurements of the state of an evolving system,\nx(t), where x has two or more components. This paper shows how to perform\nnonlinear blind source separation; i.e., how to determine if these signals are\nequal to linear or nonlinear mixtures of the state variables of two or more\nstatistically independent subsystems. First, the local distributions of\nmeasurement velocities are processed in order to derive vectors at each point\nin x-space. If the data are separable, each of these vectors must be directed\nalong a subspace of x-space that is traversed by varying the state variable of\none subsystem, while all other subsystems are kept constant. Because of this\nproperty, these vectors can be used to construct a small set of mappings, which\nmust contain the unmixing function, if it exists. Therefore, nonlinear blind\nsource separation can be performed by examining the separability of the data\nafter it has been transformed by each of these mappings. The method is\nanalytic, constructive, and model-independent. It is illustrated by blindly\nrecovering the separate utterances of two speakers from nonlinear combinations\nof their audio waveforms.\n",
"title": "Model-Independent Analytic Nonlinear Blind Source Separation"
}
| null | null |
[
"Statistics"
] | null | true | null |
13162
| null |
Validated
| null | null |
null |
{
"abstract": " We study inflation in models with many interacting fields subject to randomly\ngenerated scalar potentials. We use methods from non-equilibrium random matrix\ntheory to construct the potentials and an adaption of the 'transport method' to\nevolve the two-point correlators during inflation. This construction allows,\nfor the first time, for an explicit study of models with up to 100 interacting\nfields supporting a period of 'approximately saddle-point' inflation. We\ndetermine the statistical predictions for observables by generating over 30,000\nmodels with 2-100 fields supporting at least 60 efolds of inflation. These\nstudies lead us to seven lessons: i) Manyfield inflation is not single-field\ninflation, ii) The larger the number of fields, the simpler and sharper the\npredictions, iii) Planck compatibility is not rare, but future experiments may\nrule out this class of models, iv) The smoother the potentials, the sharper the\npredictions, v) Hyperparameters can transition from stiff to sloppy, vi)\nDespite tachyons, isocurvature can decay, vii) Eigenvalue repulsion drives the\npredictions. We conclude that many of the 'generic predictions' of single-field\ninflation can be emergent features of complex inflation models.\n",
"title": "Seven Lessons from Manyfield Inflation in Random Potentials"
}
| null | null | null | null | true | null |
13163
| null |
Default
| null | null |
null |
{
"abstract": " User engagement in online social networking depends critically on the level\nof social activity in the corresponding platform--the number of online actions,\nsuch as posts, shares or replies, taken by their users. Can we design\ndata-driven algorithms to increase social activity? At a user level, such\nalgorithms may increase activity by helping users decide when to take an action\nto be more likely to be noticed by their peers. At a network level, they may\nincrease activity by incentivizing a few influential users to take more\nactions, which in turn will trigger additional actions by other users. In this\npaper, we model social activity using the framework of marked temporal point\nprocesses, derive an alternate representation of these processes using\nstochastic differential equations (SDEs) with jumps and, exploiting this\nalternate representation, develop two efficient online algorithms with provable\nguarantees to steer social activity both at a user and at a network level. In\ndoing so, we establish a previously unexplored connection between optimal\ncontrol of jump SDEs and doubly stochastic marked temporal point processes,\nwhich is of independent interest. Finally, we experiment both with synthetic\nand real data gathered from Twitter and show that our algorithms consistently\nsteer social activity more effectively than the state of the art.\n",
"title": "Steering Social Activity: A Stochastic Optimal Control Point Of View"
}
| null | null | null | null | true | null |
13164
| null |
Default
| null | null |
null |
{
"abstract": " We investigate statistical properties of a clustering algorithm that receives\nlevel set estimates from a kernel density estimator and then estimates the\nfirst split in the density level cluster tree if such a split is present or\ndetects the absence of such a split. Key aspects of our analysis include finite\nsample guarantees, consistency, rates of convergence, and an adaptive\ndata-driven strategy for chosing the kernel bandwidth. For the rates and the\nadaptivity we do not need continuity assumptions on the density such as\nHölder continuity, but only require intuitive geometric assumptions of\nnon-parametric nature.\n",
"title": "Adaptive Clustering Using Kernel Density Estimators"
}
| null | null |
[
"Statistics"
] | null | true | null |
13165
| null |
Validated
| null | null |
null |
{
"abstract": " We prove a realization formula and a model formula for analytic functions\nwith modulus bounded by $1$ on the symmetrized bidisc \\[ G\\stackrel{\\rm def}{=}\n\\{(z+w,zw): |z|<1, \\, |w| < 1\\}. \\] As an application we prove a Pick-type\ntheorem giving a criterion for the existence of such a function satisfying a\nfinite set of interpolation conditions.\n",
"title": "Realization of functions on the symmetrized bidisc"
}
| null | null | null | null | true | null |
13166
| null |
Default
| null | null |
null |
{
"abstract": " Self consistent GW approach (scGW) has been applied to calculate the ground\nstate properties (equilibrium Wigner-Seitz radius $S_{WZ}$ and bulk modulus\n$B$) of 3d transition metals Sc, Ti, V, Fe, Co, Ni, and Cu. The approach\nsystematically underestimates $S_{WZ}$ with average relative deviation from the\nexperimental data about 1% and it overestimates the calculated bulk modulus\nwith relative error about 25%. It is shown that scGW is superior in accuracy as\ncompared to the local density approximation (LDA) but it is less accurate than\nthe generalized gradient approach (GGA) for the materials studied. If compared\nto the random phase approximation (RPA), scGW is slightly less accurate, but\nits error for the 3d metals looks more systematic. The systematic nature of the\ndeviation from the experimental data suggests that the next order of the\nperturbation theory should allow one to reduce the error.\n",
"title": "Ground state properties of 3d metals from self-consistent GW approach"
}
| null | null | null | null | true | null |
13167
| null |
Default
| null | null |
null |
{
"abstract": " Interval-censored data, in which the event time is only known to lie in some\ntime interval, arise commonly in practice; for example, in a medical study in\nwhich patients visit clinics or hospitals at pre-scheduled times, and the\nevents of interest occur between visits. Such data are appropriately analyzed\nusing methods that account for this uncertainty in event time measurement. In\nthis paper we propose a survival tree method for interval-censored data based\non the conditional inference framework. Using Monte Carlo simulations we find\nthat the tree is effective in uncovering underlying tree structure, performs\nsimilarly to an interval-censored Cox proportional hazards model fit when the\ntrue relationship is linear, and performs at least as well as (and in the\npresence of right-censoring outperforms) the Cox model when the true\nrelationship is not linear. Further, the interval-censored tree outperforms\nsurvival trees based on imputing the event time as an endpoint or the midpoint\nof the censoring interval. We illustrate the application of the method on tooth\nemergence data.\n",
"title": "Survival Trees for Interval-Censored Survival data"
}
| null | null | null | null | true | null |
13168
| null |
Default
| null | null |
null |
{
"abstract": " The medical field stands to see significant benefits from the recent advances\nin deep learning. Knowing the uncertainty in the decision made by any machine\nlearning algorithm is of utmost importance for medical practitioners. This\nstudy demonstrates the utility of using Bayesian LSTMs for classification of\nmedical time series. Four medical time series datasets are used to show the\naccuracy improvement Bayesian LSTMs provide over standard LSTMs. Moreover, we\nshow cherry-picked examples of confident and uncertain classifications of the\nmedical time series. With simple modifications of the common practice for deep\nlearning, significant improvements can be made for the medical practitioner and\npatient.\n",
"title": "Bayesian LSTMs in medicine"
}
| null | null | null | null | true | null |
13169
| null |
Default
| null | null |
null |
{
"abstract": " This paper develops and compares two motion planning algorithms for a\ntethered UAV with and without the possibility of the tether contacting the\nconfined and cluttered environment. Tethered aerial vehicles have been studied\ndue to their advantages such as power duration, stability, and safety. However,\nthe disadvantages brought in by the extra tether have not been well\ninvestigated by the robotic locomotion community, especially when the tethered\nagent is locomoting in a non-free space occupied with obstacles. In this work,\nwe propose two motion planning frameworks that (1) reduce the reachable\nconfiguration space by taking into account the tether and (2) deliberately plan\n(and relax) the contact point(s) of the tether with the environment and enable\nan equivalent reachable configuration space as the non-tethered counterpart\nwould have. Both methods are tested on a physical robot, Fotokite Pro. With our\napproaches, tethered aerial vehicles could find their applications in confined\nand cluttered environments with obstacles as opposed to ideal free space, while\nstill maintaining the advantages from the usage of a tether. The motion\nplanning strategies are particularly suitable for marsupial heterogeneous\nrobotic teams, such as visual servoing/assisting for another mobile,\ntele-operated primary robot.\n",
"title": "Motion Planning for a UAV with a Straight or Kinked Tether"
}
| null | null | null | null | true | null |
13170
| null |
Default
| null | null |
null |
{
"abstract": " We study the galactic wind in the edge-on spiral galaxy UGC 10043 with the\ncombination of the CALIFA integral field spectroscopy data, scanning\nFabry-Perot interferometry (FPI), and multiband photometry. We detect ionized\ngas in the extraplanar regions reaching a relatively high distance, up to ~ 4\nkpc above the galactic disk. The ionized gas line ratios ([N ii]/Ha, [S ii]/Ha\nand [O i]/Ha) present an enhancement along the semi minor axis, in contrast\nwith the values found at the disk, where they are compatible with ionization\ndue to H ii-regions. These differences, together with the biconic symmetry of\nthe extra-planar ionized structure, makes UGC 10043 a clear candidate for a\ngalaxy with gas outflows ionizated by shocks. From the comparison of shock\nmodels with the observed line ratios, and the kinematics observed from the FPI\ndata, we constrain the physical properties of the observed outflow. The data\nare compatible with a velocity increase of the gas along the extraplanar\ndistances up to < 400 km/s and the preshock density decreasing in the same\ndirection. We also observe a discrepancy in the SFR estimated based on Ha (0.36\nMsun/yr ) and the estimated with the CIGALE code, being the latter 5 times\nlarger. Nevertheless, this SFR is still not enough to drive the observed\ngalactic wind if we do not take into account the filling factor. We stress that\nthe combination of the three techniques of observation with models is a\npowerful tool to explore galactic winds in the Local Universe.\n",
"title": "Star formation driven galactic winds in UGC 10043"
}
| null | null | null | null | true | null |
13171
| null |
Default
| null | null |
null |
{
"abstract": " Objective: To establish the performance of several drive and measurement\npatterns in EIT imaging of neural activity in peripheral nerve, which involves\nlarge impedance change in the nerve's anisotropic length axis. Approach: Eight\ndrive and measurement electrode patterns are compared using a finite element\n(FE) four cylindrical shell model of a peripheral nerve and a 32 channel\ndual-ring nerve cuff. The central layer of the FE model contains impedance\nchanges representative of neural activity of -0.3 in the length axis and -8.8 x\n10-4 in the radial axis. Four of the electrode patterns generate longitudinal\ndrive current, which runs perpendicular to the anisotropic axis. Main results:\nTransverse current patterns produce higher resolution than longitudinal\npatterns but are also more susceptible to noise and errors, and exhibit poorer\nsensitivity to impedance changes in central sample locations. Three of the four\nlongitudinal current patterns considered can reconstruct fascicle level\nimpedance changes with up to 0.2 mV noise and error, which corresponds to\nbetween -5.5 and +0.18 dB of the normalised signal standard deviation. Reducing\nthe spacing between the two electrode rings in all longitudinal current\npatterns reduced the signal to error ratio across all depth locations of the\nsample. Significance: Electrode patterns which target the large impedance\nchange in the anisotropic length axis can provide improved robustness against\nnoise and errors, which is a critical step towards real time EIT imaging of\nneural activity in peripheral nerve.\n",
"title": "Drive and measurement electrode patterns for electrode impedance tomography (EIT) imaging of neural activity in peripheral nerve"
}
| null | null | null | null | true | null |
13172
| null |
Default
| null | null |
null |
{
"abstract": " Liouville type theorems for the stationary Navier-Stokes equations are proven\nunder certain assumptions. These assumptions are motivated by conditions that\nappear in Liouvile type theorems for the heat equations with a given divergence\nfree drift.\n",
"title": "Remarks on Liouville Type Theorems for Steady-State Navier-Stokes Equations"
}
| null | null | null | null | true | null |
13173
| null |
Default
| null | null |
null |
{
"abstract": " Let $P$ be a set of nodes in a wireless network, where each node is modeled\nas a point in the plane, and let $s\\in P$ be a given source node. Each node $p$\ncan transmit information to all other nodes within unit distance, provided $p$\nis activated. The (homogeneous) broadcast problem is to activate a minimum\nnumber of nodes such that in the resulting directed communication graph, the\nsource $s$ can reach any other node. We study the complexity of the regular and\nthe hop-bounded version of the problem (in the latter, $s$ must be able to\nreach every node within a specified number of hops), with the restriction that\nall points lie inside a strip of width $w$. We almost completely characterize\nthe complexity of both the regular and the hop-bounded versions as a function\nof the strip width $w$.\n",
"title": "The Homogeneous Broadcast Problem in Narrow and Wide Strips"
}
| null | null | null | null | true | null |
13174
| null |
Default
| null | null |
null |
{
"abstract": " We present ALMA $^{12}$CO (J=1-0, 3-2 and 6-5), $^{13}$CO (J=1-0) and\nC$^{18}$O (J=1-0) observations of the local Ultra Luminous Infrared Galaxy,\nIRAS 13120-5453 (dubbed \"The Yo-yo\"). The morphologies of the three isotopic\nspecies differ, where $^{13}$CO shows a hole in emission towards the center. We\nmeasure integrated brightness temperature line ratios of $^{12}$CO/$^{13}$CO\n$\\geq$ 60 (exceeding 200) and $^{13}$CO/C$^{18}$O $\\leq$ 1 in the central\nregion. Assuming optical thin emission, C$^{18}$O is more abundant than\n$^{13}$CO in several regions. The abundances within the central 500 pc are\nconsistent with enrichment of the ISM via a young starburst ($<$7Myr), a\ntop-heavy initial mass function or a combination of both.\n",
"title": "Extreme CO Isotopic Abundances in the ULIRG IRAS 13120-5453: An Extremely Young Starburst or Top-Heavy Initial Mass Function"
}
| null | null |
[
"Physics"
] | null | true | null |
13175
| null |
Validated
| null | null |
null |
{
"abstract": " We consider the problem of computing the nearest matrix polynomial with a\nnon-trivial Smith Normal Form. We show that computing the Smith form of a\nmatrix polynomial is amenable to numeric computation as an optimization\nproblem. Furthermore, we describe an effective optimization technique to find a\nnearby matrix polynomial with a non-trivial Smith form. The results are then\ngeneralized to include the computation of a matrix polynomial having a maximum\nspecified number of ones in the Smith Form (i.e., with a maximum specified\nMcCoy rank).\nWe discuss the geometry and existence of solutions and how our results can\nused for an error analysis. We develop an optimization-based approach and\ndemonstrate an iterative numerical method for computing a nearby matrix\npolynomial with the desired spectral properties. We also describe an\nimplementation of our algorithms and demonstrate the robustness with examples\nin Maple.\n",
"title": "Computing Nearby Non-trivial Smith Forms"
}
| null | null |
[
"Computer Science"
] | null | true | null |
13176
| null |
Validated
| null | null |
null |
{
"abstract": " Fixed point iterations play a central role in the design and the analysis of\na large number of optimization algorithms. We study a new iterative scheme in\nwhich the update is obtained by applying a composition of quasinonexpansive\noperators to a point in the affine hull of the orbit generated up to the\ncurrent iterate. This investigation unifies several algorithmic constructs,\nincluding Mann's mean value method, inertial methods, and multi-layer\nmemoryless methods. It also provides a framework for the development of new\nalgorithms, such as those we propose for solving monotone inclusion and\nminimization problems.\n",
"title": "Quasinonexpansive Iterations on the Affine Hull of Orbits: From Mann's Mean Value Algorithm to Inertial Methods"
}
| null | null |
[
"Mathematics"
] | null | true | null |
13177
| null |
Validated
| null | null |
null |
{
"abstract": " In prior work, we addressed the problem of optimally controlling on line\nconnected and automated vehicles crossing two adjacent intersections in an\nurban area to minimize fuel consumption while achieving maximal throughput\nwithout any explicit traffic signaling and without considering left and right\nturns. In this paper, we extend the solution of this problem to account for\nleft and right turns under hard safety constraints. Furthermore, we formulate\nand solve another optimization problem to minimize a measure of passenger\ndiscomfort while the vehicle turns at the intersection and we investigate the\nassociated tradeoff between minimizing fuel consumption and passenger\ndiscomfort.\n",
"title": "Decentralized Optimal Control for Connected Automated Vehicles at Intersections Including Left and Right Turns"
}
| null | null | null | null | true | null |
13178
| null |
Default
| null | null |
null |
{
"abstract": " Let (P) denote the problem of existence of a point in the plane of a given\ntriangle T, that is at rational distance from all the vertices of T. In this\narticle, we provide a complete solution to (P) for all equilateral triangles.\n",
"title": "The Rational Distance Problem for Equilateral Triangles"
}
| null | null | null | null | true | null |
13179
| null |
Default
| null | null |
null |
{
"abstract": " It is proved that the category $\\mathbb{EM}$ of extended multisets is dually\nequivalent to the category $\\mathbb{CHMV}$ of compact Hausdorff MV-algebras\nwith continuous homomorphisms, which is in turn equivalent to the category of\ncomplete and completely distributive MV-algebras with homomorphisms that\nreflect principal maximal ideals. Urysohn-Strauss's Lemma, Gleason's Theorem,\nand projective objects are also investigated for topological MV-algebras.\n",
"title": "Compact Hausdorff MV-algebras: Structure, Duality and Projectivity"
}
| null | null | null | null | true | null |
13180
| null |
Default
| null | null |
null |
{
"abstract": " We prove the nonvanishing of the twisted central critical values of a class\nof automorphic L-functions for twists by all but finitely many unitary\ncharacters in particular infinite families. The methods build on a\nnon-archimedean approach introduced by Greenberg in the context of the Birch\nand Swinnerton-Dyer Conjecture. While this paper focuses on L-functions\nassociated to certain automorphic representations of unitary groups, it\nillustrates how decades-old methods from Iwasawa theory can be combined with\nthe output of new machinery to achieve broader nonvanishing results.\n",
"title": "Nonvanishing theorems for twisted L-functions on unitary groups"
}
| null | null | null | null | true | null |
13181
| null |
Default
| null | null |
null |
{
"abstract": " We consider an elastic composite material containing particulate inclusions\nin a soft elastic matrix that is bounded by a rigid wall, e.g., the substrate.\nIf such a composite serves as a soft actuator, forces are imposed on or induced\nbetween the embedded particles. We investigate how the presence of the rigid\nwall affects the interactions between the inclusions in the elastic matrix. For\nno-slip boundary conditions, we transfer Blake's derivation of a corresponding\nGreen's function from low-Reynolds-number hydrodynamics to the linearly elastic\ncase. Results for no-slip and free-slip surface conditions are compared to each\nother and to the bulk behavior. Our results suggest that walls with free-slip\nsurface conditions are preferred when they serve as substrates for soft\nactuators made from elastic composite materials. As we further demonstrate, the\npresence of a rigid wall can qualitatively change the interactions between the\ninclusions. In effect, it can switch attractive interactions into repulsive\nones (and vice versa). It should be straightforward to observe the effects in\nfuture experiments and to combine our results, e.g., with the modeling of\nbiological cells and tissue on rigid surfaces.\n",
"title": "Force-induced elastic matrix-mediated interactions in the presence of a rigid wall"
}
| null | null | null | null | true | null |
13182
| null |
Default
| null | null |
null |
{
"abstract": " We propose a method for the implementation of one-way quantum computing in\nsuperconducting circuits. Measurement-based quantum computing is a universal\nquantum computation paradigm in which an initial cluster-state provides the\nquantum resource, while the iteration of sequential measurements and local\nrotations encodes the quantum algorithm. Up to now, technical constraints have\nlimited a scalable approach to this quantum computing alternative. The initial\ncluster state can be generated with available controlled-phase gates, while the\nquantum algorithm makes use of high-fidelity readout and coherent feedforward.\nWith current technology, we estimate that quantum algorithms with above 20\nqubits may be implemented in the path towards quantum supremacy. Moreover, we\npropose an alternative initial state with properties of maximal persistence and\nmaximal connectedness, reducing the required resources of one-way quantum\ncomputing protocols.\n",
"title": "One-way quantum computing in superconducting circuits"
}
| null | null | null | null | true | null |
13183
| null |
Default
| null | null |
null |
{
"abstract": " Let $X$ be a compact connected strongly pseudoconvex $CR$ manifold of real\ndimension $2n-1$ in $\\mathbb{C}^{N}$. For $n\\ge 3$, Yau solved the complex\nPlateau problem of hypersurface type by checking a bunch of Kohn-Rossi\ncohomology groups in 1981. In this paper, we generalize Yau's conjecture on\nsome numerical invariant of every isolated surface singularity defined by Yau\nand the author to any dimension and prove that the conjecture is true for local\ncomplete intersection singularities of dimension $n\\ge 3$. As a direct\napplication, we solved complex Plateau problem of hypersurface type for any\ndimension $n\\ge 3$ by checking only one numerical invariant.\n",
"title": "Smooth solution to higher dimensional complex Plateau problem"
}
| null | null | null | null | true | null |
13184
| null |
Default
| null | null |
null |
{
"abstract": " We study the Steiner Tree problem, in which a set of terminal vertices needs\nto be connected in the cheapest possible way in an edge-weighted graph. This\nproblem has been extensively studied from the viewpoint of approximation and\nalso parametrization. In particular, on one hand Steiner Tree is known to be\nAPX-hard, and W[2]-hard on the other, if parameterized by the number of\nnon-terminals (Steiner vertices) in the optimum solution. In contrast to this\nwe give an efficient parameterized approximation scheme (EPAS), which\ncircumvents both hardness results. Moreover, our methods imply the existence of\na polynomial size approximate kernelization scheme (PSAKS) for the considered\nparameter.\nWe further study the parameterized approximability of other variants of\nSteiner Tree, such as Directed Steiner Tree and Steiner Forest. For neither of\nthese an EPAS is likely to exist for the studied parameter: for Steiner Forest\nan easy observation shows that the problem is APX-hard, even if the input graph\ncontains no Steiner vertices. For Directed Steiner Tree we prove that\napproximating within any function of the studied parameter is W[1]-hard.\nNevertheless, we show that an EPAS exists for Unweighted Directed Steiner Tree,\nbut a PSAKS does not. We also prove that there is an EPAS and a PSAKS for\nSteiner Forest if in addition to the number of Steiner vertices, the number of\nconnected components of an optimal solution is considered to be a parameter.\n",
"title": "Parameterized Approximation Schemes for Steiner Trees with Small Number of Steiner Vertices"
}
| null | null | null | null | true | null |
13185
| null |
Default
| null | null |
null |
{
"abstract": " A luminous stimulus which penetrates in a retina is converted to a nerve\nmessage. Ganglion cells give a response that may be approximated by a wavelet.\nWe determine a function PSI which is associated with the propagation of nerve\nimpulses along an axon. Each kind of channel (inward and outward) may be open\nor closed, depending on the transmembrane potential. The transition between\nthese states is a random event. Using quantum relations, we estimate the number\nof channels susceptible to switch between the closed and open states. Our\nquantum approach was first to calculate the energy level distribution in a\nchannel. We obtain, for each kind of channel, the empty level density and the\nfilled level density of the open and closed conformations. The joint density of\nlevels provides the transition number between the closed and open\nconformations. The algebraic sum of inward and outward open channels is a\nfunction PSI of the normalized energy E. The function PSI verifies the major\nproperties of a wavelet. We calculate the functional dependence of the axon\nmembrane conductance with the transmembrane energy.\n",
"title": "Nerve impulse propagation and wavelet theory"
}
| null | null |
[
"Quantitative Biology"
] | null | true | null |
13186
| null |
Validated
| null | null |
null |
{
"abstract": " We present a significantly different reflection process from an optically\nthin flat metallic or dielectric layer and propose a strikingly simple method\nto form approximately unipolar half-cycle optical pulses via reflection of a\nsingle-cycle optical pulse. Unipolar pulses in reflection arise due to\nspecifics of effectively one-dimensional pulse propagation. Namely, we show\nthat in considered system the field emitted by a flat medium layer is\nproportional to the velocity of oscillating medium charges instead of their\nacceleration as it is usually the case. When the single-cycle pulse interacts\nwith linear optical medium, the oscillation velocity of medium charges can be\nthen forced to keep constant sign throughout the pulse duration. Our results\nessentially differ from the direct mirror reflection and suggest a possibility\nof unusual transformations of the few-cycle light pulses in linear optical\nsystems.\n",
"title": "Generation of unipolar half-cycle pulse via unusual reflection of a single-cycle pulse from an optically thin metallic or dielectric layer"
}
| null | null | null | null | true | null |
13187
| null |
Default
| null | null |
null |
{
"abstract": " We examine gradient descent on unregularized logistic regression problems,\nwith homogeneous linear predictors on linearly separable datasets. We show the\npredictor converges to the direction of the max-margin (hard margin SVM)\nsolution. The result also generalizes to other monotone decreasing loss\nfunctions with an infimum at infinity, to multi-class problems, and to training\na weight layer in a deep network in a certain restricted setting. Furthermore,\nwe show this convergence is very slow, and only logarithmic in the convergence\nof the loss itself. This can help explain the benefit of continuing to optimize\nthe logistic or cross-entropy loss even after the training error is zero and\nthe training loss is extremely small, and, as we show, even if the validation\nloss increases. Our methodology can also aid in understanding implicit\nregularization n more complex models and with other optimization methods.\n",
"title": "The Implicit Bias of Gradient Descent on Separable Data"
}
| null | null | null | null | true | null |
13188
| null |
Default
| null | null |
null |
{
"abstract": " We demonstrate the existence of long-lived prethermalized states in the Mott\ninsulating Hubbard model driven by periodic electric fields. These states,\nwhich also exist in the resonantly driven case with a large density of\nphoto-induced doublons and holons, are characterized by a nonzero current and\nan effective temperature of the doublons and holons which depends sensitively\non the driving condition. Focusing on the specific case of resonantly driven\nmodels whose effective time-independent Hamiltonian in the high-frequency\ndriving limit corresponds to noninteracting fermions, we show that the time\nevolution of the double occupation can be reproduced by the effective\nHamiltonian, and that the prethermalization plateaus at finite driving\nfrequency are controlled by the next-to-leading order correction in the\nhigh-frequency expansion of the effective Hamiltonian. We propose a numerical\nprocedure to determine an effective Hubbard interaction that mimics the\ncorrelation effects induced by these higher order terms.\n",
"title": "Floquet prethermalization in the resonantly driven Hubbard model"
}
| null | null | null | null | true | null |
13189
| null |
Default
| null | null |
null |
{
"abstract": " We develop a systematic study of Jahn-Teller (JT) models with continuous\nsymmetries by explor- ing their algebraic properties. The compact symmetric\nspaces corresponding to JT models carrying a Lie group symmetry are identified,\nand their invariants used to reduce their adiabatic potential energy surfaces\ninto orbit spaces. Each orbit consists of a set of JT distorted molecular\nstructures with equal adiabatic electronic spectrum. Molecular motion may be\ndecomposed into pseudorota- tional and radial. The former preserves the orbit,\nwhile the latter maps an orbit into another. The dimensionality and topology of\nthe internal space of each orbit depends on the number of degener- ate states\nin its adiabatic electronic spectra. Furthermore, qualitatively different\npseudorotational modes occur in orbits of different types. We also provide a\nsimple proof that the electronic spectrum for the space of JT minimum-energy\nstructures (trough) displays a universality predicted by the epikernel\nprinciple. This result is in turn used to prove the topological equivalence\nbetween bosonic (fermionic) JT troughs and real (quaternionic) projective\nspaces, a conclusion which has outstanding physical consequences, as explained\nin our work. The relevance of our study for the more common case of JT systems\nwith only discrete point group symmetry, and for generic asymmetric molecular\nsystems with conical intersections involving more than two states is likewise\ndiscussed. In particular, we show that JT models with continuous symmetries\npresent the simplest models of conical intersections among an arbitrary number\nof electronic state crossings.\n",
"title": "Continuous vibronic symmetries in Jahn-Teller models"
}
| null | null |
[
"Physics"
] | null | true | null |
13190
| null |
Validated
| null | null |
null |
{
"abstract": " In the article we present a general theory of augmented Lagrangian functions\nfor cone constrained optimization problems that allows one to study almost all\nknown augmented Lagrangians for cone constrained programs within a unified\nframework. We develop a new general method for proving the existence of global\nsaddle points of augmented Lagrangian functions, called the localization\nprinciple. The localization principle unifies, generalizes and sharpens most of\nthe known results on existence of global saddle points, and, in essence,\nreduces the problem of the existence of saddle points to a local analysis of\noptimality conditions. With the use of the localization principle we obtain\nfirst necessary and sufficient conditions for the existence of a global saddle\npoint of an augmented Lagrangian for cone constrained minimax problems via both\nsecond and first order optimality conditions. In the second part of the paper,\nwe present a general approach to the construction of globally exact augmented\nLagrangian functions. The general approach developed in this paper allowed us\nnot only to sharpen most of the existing results on globally exact augmented\nLagrangians, but also to construct first globally exact augmented Lagrangian\nfunctions for equality constrained optimization problems, for nonlinear second\norder cone programs and for nonlinear semidefinite programs. These globally\nexact augmented Lagrangians can be utilized in order to design new\nsuperlinearly (or even quadratically) convergent optimization methods for cone\nconstrained optimization problems.\n",
"title": "Augmented Lagrangian Functions for Cone Constrained Optimization: the Existence of Global Saddle Points and Exact Penalty Property"
}
| null | null | null | null | true | null |
13191
| null |
Default
| null | null |
null |
{
"abstract": " We introduce and study a one-parameter generalization of the q-Whittaker\nsymmetric functions. This is a family of multivariate symmetric polynomials,\nwhose construction may be viewed as an application of the procedure of fusion\nfrom integrable lattice models to a vertex model interpretation of a\none-parameter generalization of Hall-Littlewood polynomials from [Bor17, BP16a,\nBP16b].\nWe prove branching and Pieri rules, standard and dual (skew) Cauchy summation\nidentities, and an integral representation for the new polynomials.\n",
"title": "Spin $q$-Whittaker polynomials"
}
| null | null | null | null | true | null |
13192
| null |
Default
| null | null |
null |
{
"abstract": " Frankl and Füredi conjectured in 1989 that the maximum Lagrangian of all\n$r$-uniform hypergraphs of fixed size $m$ is realised by the initial segment of\nthe colexicographic order. In particular, in the principal case\n$m=\\binom{t}{r}$ their conjecture states that every $H\\subseteq\n\\mathbb{N}^{(r)}$ of size $\\binom{t}{r}$ satisfies \\begin{align*} \\max\n\\{\\sum_{A \\in H}\\prod_{i\\in A} y_i \\ \\colon \\ y_1,y_2,\\ldots \\geq 0; \\sum_{i\\in\n\\mathbb{N}} y_i=1 \\}&\\leq \\frac{1}{t^r}\\binom{t}{r}. \\end{align*}\nWe prove the above statement for all $r\\geq 4$ and large values of $t$ (the\ncase $r=3$ was settled by Talbot in 2002). More generally, we show for any\n$r\\geq 4$ that the Frankl-Füredi conjecture holds whenever $\\binom{t-1}{r}\n\\leq m \\leq \\binom{t}{r}- \\gamma_r t^{r-2}$ for a constant $\\gamma_r>0$,\nthereby verifying it for `most' $m\\in \\mathbb{N}$.\nFurthermore, for $r=3$ we make an improvement on the results of\nTalbot~\\cite{Tb} and Tang, Peng, Zhang and Zhao~\\cite{TPZZ}.\n",
"title": "Lagrangians of hypergraphs: The Frankl-Füredi conjecture holds almost everywhere"
}
| null | null | null | null | true | null |
13193
| null |
Default
| null | null |
null |
{
"abstract": " To accelerate research on adversarial examples and robustness of machine\nlearning classifiers, Google Brain organized a NIPS 2017 competition that\nencouraged researchers to develop new methods to generate adversarial examples\nas well as to develop new ways to defend against them. In this chapter, we\ndescribe the structure and organization of the competition and the solutions\ndeveloped by several of the top-placing teams.\n",
"title": "Adversarial Attacks and Defences Competition"
}
| null | null | null | null | true | null |
13194
| null |
Default
| null | null |
null |
{
"abstract": " Complex networks analyses of many physical, biological and social phenomena\nshow remarkable structural regularities, yet, their application in studying\nhuman past interaction remains underdeveloped. Here, we present an innovative\nmethod for identifying community structures in the archaeological record that\nallow for independent evaluation of the copper using societies in the Balkans,\nfrom c. 6200 to c. 3200 BC. We achieve this by exploring modularity of\nnetworked systems of these societies across an estimated 3000 years. We employ\nchemical data of copper-based objects from 79 archaeological sites as the\nindependent variable for detecting most densely interconnected sets of nodes\nwith a modularity maximization method. Our results reveal three dominant\nmodular structures across the entire period, which exhibit strong spatial and\ntemporal significance. We interpret patterns of copper supply among prehistoric\nsocieties as reflective of social relations, which emerge as equally important\nas physical proximity. Although designed on a variable isolated from any\narchaeological and spatiotemporal information, our method provides\narchaeologically and spatiotemporally meaningful results. It produces models of\nhuman interaction and cooperation that can be evaluated independently of\nestablished archaeological systematics, and can find wide application on any\nquantitative data from archaeological and historical record.\n",
"title": "Community structure of copper supply networks in the prehistoric Balkans: An independent evaluation of the archaeological record from the 7th to the 4th millennium BC"
}
| null | null | null | null | true | null |
13195
| null |
Default
| null | null |
null |
{
"abstract": " This paper describes a method for clustering data that are spread out over\nlarge regions and which dimensions are on different scales of measurement. Such\nan algorithm was developed to implement a robotics application consisting in\nsorting and storing objects in an unsupervised way. The toy dataset used to\nvalidate such application consists of Lego bricks of different shapes and\ncolors. The uncontrolled lighting conditions together with the use of RGB color\nfeatures, respectively involve data with a large spread and different levels of\nmeasurement between data dimensions. To overcome the combination of these two\ncharacteristics in the data, we have developed a new weighted K-means\nalgorithm, called gap-ratio K-means, which consists in weighting each dimension\nof the feature space before running the K-means algorithm. The weight\nassociated with a feature is proportional to the ratio of the biggest gap\nbetween two consecutive data points, and the average of all the other gaps.\nThis method is compared with two other variants of K-means on the Lego bricks\nclustering problem as well as two other common classification datasets.\n",
"title": "Clustering for Different Scales of Measurement - the Gap-Ratio Weighted K-means Algorithm"
}
| null | null | null | null | true | null |
13196
| null |
Default
| null | null |
null |
{
"abstract": " The goal of the paper is to investigate the dynamics of the eigenvalues of\nthe Sturm-Liouville operator with summable PT-symmetric potential on the finite\ninterval. It turns out that the case of a complex Airy operator presents an\nexactly solvable model which allows us to trace the dynamics of the movement of\nthe eigenvalues in all details and to find explicitly the critical parameter\nvalues, in particular, to specify precisely the number $\\varepsilon_1$ such\nthat for $0<\\varepsilon<\\varepsilon_1$ the operator has a real spectrum and is\nsimilar to a self-adjoint operator.\n",
"title": "Eigenvalue Dynamics of a PT-symmetric Sturm-Liouville Operator. Criteria of the Similarity to a Self-adjoint or Normal Operator"
}
| null | null | null | null | true | null |
13197
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we study algorithmic problems for automaton semigroups and\nautomaton groups related to freeness and finiteness. In the course of this\nstudy, we also exhibit some connections between the algebraic structure of\nautomaton (semi)groups and their dynamics on the boundary.\nFirst, we show that it is undecidable to check whether the group generated by\na given invertible automaton has a positive relation, i. e. a relation p = 1\nsuch that p only contains positive generators. Besides its obvious relation to\nthe freeness of the group, the absence of positive relations has previously\nbeen studied and is connected to the triviality of some stabilizers of the\nboundary. We show that the emptiness of the set of positive relations is\nequivalent to the dynamical property that all (directed positive) orbital\ngraphs centered at non-singular points are acyclic. Our approach also works to\nshow undecidability of the freeness problem for automaton semigroups; in fact,\nit shows undecidability of a strengthened version where the input automaton is\ncomplete and invertible.\nGillibert showed that the finiteness problem for automaton semigroups is\nundecidable. In the second part of the paper, we show that this undecidability\nresult also holds if the input is restricted to be bi-reversible and invertible\n(but, in general, not complete). As an immediate consequence, we obtain that\nthe finiteness problem for automaton subsemigroups of semigroups generated by\ninvertible, yet partial automata, so called automaton-inverse semigroups, is\nalso undecidable.\n",
"title": "Automaton Semigroups and Groups: on the Undecidability of Problems Related to Freeness and Finiteness"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
13198
| null |
Validated
| null | null |
null |
{
"abstract": " Recent data indicate one or more moderately nearby supernovae in the early\nPleistocene, with additional events likely in the Miocene. This has motivated\nmore detailed computations, using new information about the nature of\nsupernovae and the distances of these events to describe in more detail the\nsorts of effects that are indicated at the Earth. This short\ncommunication/review is designed to describe some of these effects so that they\nmay possibly be related to changes in the biota around these times.\n",
"title": "Terrestrial effects of moderately nearby supernovae"
}
| null | null | null | null | true | null |
13199
| null |
Default
| null | null |
null |
{
"abstract": " Generative adversarial networks (GANs) form a generative modeling approach\nknown for producing appealing samples, but they are notably difficult to train.\nOne common way to tackle this issue has been to propose new formulations of the\nGAN objective. Yet, surprisingly few studies have looked at optimization\nmethods designed for this adversarial training. In this work, we cast GAN\noptimization problems in the general variational inequality framework. Tapping\ninto the mathematical programming literature, we counter some common\nmisconceptions about the difficulties of saddle point optimization and propose\nto extend techniques designed for variational inequalities to the training of\nGANs. We apply averaging, extrapolation and a novel computationally cheaper\nvariant that we call extrapolation from the past to the stochastic gradient\nmethod (SGD) and Adam.\n",
"title": "A Variational Inequality Perspective on Generative Adversarial Networks"
}
| null | null | null | null | true | null |
13200
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.