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": " Nature inspired neuromorphic architectures are being explored as an\nalternative to imminent limitations of conventional complementary metal-oxide\nsemiconductor (CMOS) architectures. Utilization of such architectures for\npractical applications like advanced pattern recognition tasks will require\nsynaptic connections that are both reconfigurable and stable. Here, we report\nrealization of stable atomic-switch networks (ASN), with inherent complex\nconnectivity, self-assembled from percolating metal nanoparticles (NPs). The\ndevice conductance reflects the configuration of synapses which can be\nmodulated via voltage stimulus. By controlling Relative Humidity (RH) and\noxygen partial-pressure during NP deposition we obtain stochastic conductance\nswitching that is stable over several months. Detailed characterization reveals\nsignatures of electric-field induced atomic-wire formation within the\ntunnel-gaps of the oxidized percolating network. Finally we show that the\nsynaptic structure can be reconfigured by stimulating at different repetition\nrates, which can be utilized as short-term to long-term memory conversion. This\ndemonstration of stable stochastic switching in ASNs provides a promising route\nto hardware implementation of biological neuronal models and, as an example, we\nhighlight possible applications in Reservoir Computing (RC).\n",
"title": "Stable Self-Assembled Atomic-Switch Networks for Neuromorphic Applications"
}
| null | null | null | null | true | null |
14501
| null |
Default
| null | null |
null |
{
"abstract": " An ADE Dynkin diagram gives rise to a family of algebraic curves. In this\npaper, we use arithmetic invariant theory to study the integral points of the\ncurves associated to the exceptional diagrams $E_6, E_7$, $E_8$. These curves\nare non-hyperelliptic of genus 3 or 4. We prove that a positive proportion of\neach family consists of curves with integral points everywhere locally but no\nintegral points globally.\n",
"title": "On the arithmetic of simple singularities of type E"
}
| null | null | null | null | true | null |
14502
| null |
Default
| null | null |
null |
{
"abstract": " We study a network utility maximization (NUM) decomposition in which the set\nof flow rates is grouped by source-destination pairs. We develop theorems for\nboth single-path and multipath cases, which relate an arbitrary NUM problem\ninvolving all flow rates to a simpler problem involving only the aggregate\nrates for each source-destination pair. The optimal aggregate flows are then\napportioned among the constituent flows of each pair. This apportionment is\nsimple for the case of $\\alpha$-fair utility functions. We also show how the\ndecomposition can be implemented with the alternating direction method of\nmultipliers (ADMM) algorithm.\n",
"title": "Centralized Network Utility Maximization over Aggregate Flows"
}
| null | null | null | null | true | null |
14503
| null |
Default
| null | null |
null |
{
"abstract": " With the advent of deep learning, object detection drifted from a bottom-up\nto a top-down recognition problem. State of the art algorithms enumerate a\nnear-exhaustive list of object locations and classify each into: object or not.\nIn this paper, we show that bottom-up approaches still perform competitively.\nWe detect four extreme points (top-most, left-most, bottom-most, right-most)\nand one center point of objects using a standard keypoint estimation network.\nWe group the five keypoints into a bounding box if they are geometrically\naligned. Object detection is then a purely appearance-based keypoint estimation\nproblem, without region classification or implicit feature learning. The\nproposed method performs on-par with the state-of-the-art region based\ndetection methods, with a bounding box AP of 43.2% on COCO test-dev. In\naddition, our estimated extreme points directly span a coarse octagonal mask,\nwith a COCO Mask AP of 18.9%, much better than the Mask AP of vanilla bounding\nboxes. Extreme point guided segmentation further improves this to 34.6% Mask\nAP.\n",
"title": "Bottom-up Object Detection by Grouping Extreme and Center Points"
}
| null | null | null | null | true | null |
14504
| null |
Default
| null | null |
null |
{
"abstract": " This paper examines the cosmic ray He and C nuclei spectra below ~1 GeV/nuc,\nas well as the very rapid increase in the He/C ratio below ~100 MeV/nuc,\nmeasured by Voyager 1 beyond the heliopause. Using a simple Leaky Box Model\n(LBM) for galactic propagation we have not been able to simultaneously\nreproduce the individual He and C nuclei spectra and the large increase in He/C\nratio that is observed at low energies. However, using a truncated LBM with\ndifferent truncation parameters for each nucleus that are related to their rate\nof energy loss by ionization which is ~Z2/A, these different features can be\nmatched. This suggests that we are observing the effects of the source\ndistribution of cosmic rays in the galaxy on the low energy spectra of cosmic\nray nuclei and that there may be a paucity of nearby sources. In this\npropagation model we start very specific source spectra for He and C which are\n~dj/dP = P-2.24, the same for each nucleus and also for all rigidities. These\nsource spectra become spectra with spectral indices ~-2.69 at high rigidities\nfor both charges as a result of a rigidity dependence of the diffusion\ncoefficient governing the propagation which is taken to be ~P-0.45. This\nexponent is determined directly from the B/C ratio measured by AMS-2. These\npropagated P-2.69 spectra, when extended to high energies, predict He and C\nintensities and a He/C ratio that are within +3-5% of the intensities and ratio\nrecently measured by AMS-2 in the energy range from 10 to 1000 GeV/nuc.\n",
"title": "The Different Shapes of the LIS Energy Spectra of Cosmic Ray He and C Nuclei Below ~1 GeV/nuc and The Cosmic Ray He/C Nuclei Ratio vs. Energy -V1 Measurements and LBM Propagation Predictions"
}
| null | null | null | null | true | null |
14505
| null |
Default
| null | null |
null |
{
"abstract": " Symmetric nonnegative matrix factorization (SymNMF) has important\napplications in data analytics problems such as document clustering, community\ndetection and image segmentation. In this paper, we propose a novel nonconvex\nvariable splitting method for solving SymNMF. The proposed algorithm is\nguaranteed to converge to the set of Karush-Kuhn-Tucker (KKT) points of the\nnonconvex SymNMF problem. Furthermore, it achieves a global sublinear\nconvergence rate. We also show that the algorithm can be efficiently\nimplemented in parallel. Further, sufficient conditions are provided which\nguarantee the global and local optimality of the obtained solutions. Extensive\nnumerical results performed on both synthetic and real data sets suggest that\nthe proposed algorithm converges quickly to a local minimum solution.\n",
"title": "A Nonconvex Splitting Method for Symmetric Nonnegative Matrix Factorization: Convergence Analysis and Optimality"
}
| null | null | null | null | true | null |
14506
| null |
Default
| null | null |
null |
{
"abstract": " The total mass M_GCS in the globular cluster (GC) system of a galaxy is\nempirically a near-constant fraction of the total mass M_h = M_bary + M_dark of\nthe galaxy, across a range of 10^5 in galaxy mass. This trend is radically\nunlike the strongly nonlinear behavior of total stellar mass M_star versus M_h.\nWe discuss extensions of this trend to two more extreme situations: (a) entire\nclusters of galaxies, and (b) the Ultra-Diffuse Galaxies (UDGs) recently\ndiscovered in Coma and elsewhere. Our calibration of the ratio \\eta_M = M_GCS /\nM_h from normal galaxies, accounting for new revisions in the adopted\nmass-to-light ratio for GCs, now gives \\eta_M = 2.9 \\times 10^{-5} as the mean\nabsolute mass fraction. We find that the same ratio appears valid for galaxy\nclusters and UDGs. Estimates of \\eta_M in the four clusters we examine tend to\nbe slightly higher than for individual galaxies, butmore data and better\nconstraints on the mean GC mass in such systems are needed to determine if this\ndifference is significant. We use the constancy of \\eta_M to estimate total\nmasses for several individual cases; for example, the total mass of the Milky\nWay is calculated to be M_h = 1.1 \\times 10^{12} M_sun. Physical explanations\nfor the uniformity of \\eta_M are still descriptive, but point to a picture in\nwhich massive, dense star clusters in their formation stages were relatively\nimmune to the feedback that more strongly influenced lower-density regions\nwhere most stars form.\n",
"title": "Galactic Dark Matter Halos and Globular Cluster Populations. III: Extension to Extreme Environments"
}
| null | null |
[
"Physics"
] | null | true | null |
14507
| null |
Validated
| null | null |
null |
{
"abstract": " We present a database of parliamentary debates that contains the complete\nrecord of parliamentary speeches from Dáil Éireann, the lower house and\nprincipal chamber of the Irish parliament, from 1919 to 2013. In addition, the\ndatabase contains background information on all TDs (Teachta Dála, members of\nparliament), such as their party affiliations, constituencies and office\npositions. The current version of the database includes close to 4.5 million\nspeeches from 1,178 TDs. The speeches were downloaded from the official\nparliament website and further processed and parsed with a Python script.\nBackground information on TDs was collected from the member database of the\nparliament website. Data on cabinet positions (ministers and junior ministers)\nwas collected from the official website of the government. A record linkage\nalgorithm and human coders were used to match TDs and ministers.\n",
"title": "Database of Parliamentary Speeches in Ireland, 1919-2013"
}
| null | null | null | null | true | null |
14508
| null |
Default
| null | null |
null |
{
"abstract": " This paper deals with model order reduction of parametrical dynamical\nsystems. We consider the specific setup where the distribution of the system's\ntrajectories is unknown but the following two sources of information are\navailable: \\textit{(i)} some \"rough\" prior knowledge on the system's\nrealisations; \\textit{(ii)} a set of \"incomplete\" observations of the system's\ntrajectories. We propose a Bayesian methodological framework to build\nreduced-order models (ROMs) by exploiting these two sources of information. We\nemphasise that complementing the prior knowledge with the collected data\nprovably enhances the knowledge of the distribution of the system's\ntrajectories. We then propose an implementation of the proposed methodology\nbased on Monte-Carlo methods. In this context, we show that standard ROM\nlearning techniques, such e.g. Proper Orthogonal Decomposition or Dynamic Mode\nDecomposition, can be revisited and recast within the probabilistic framework\nconsidered in this paper.~We illustrate the performance of the proposed\napproach by numerical results obtained for a standard geophysical model.\n",
"title": "Reduced Modeling of Unknown Trajectories"
}
| null | null | null | null | true | null |
14509
| null |
Default
| null | null |
null |
{
"abstract": " Ensembles of nitrogen-vacancy centers in diamond are a highly promising\nplatform for high-sensitivity magnetometry, whose efficacy is often based on\nefficiently generating and monitoring magnetic-field dependent infrared\nfluorescence. Here we report on an increased sensing efficiency with the use of\na 532-nm resonant confocal cavity and a microwave resonator antenna for\nmeasuring the local magnetic noise density using the intrinsic nitrogen-vacancy\nconcentration of a chemical-vapor deposited single-crystal diamond. We measure\na near-shot-noise-limited magnetic noise floor of 200 pT/$\\sqrt{\\text{Hz}}$\nspanning a bandwidth up to 159 Hz, and an extracted sensitivity of\napproximately 3 nT/$\\sqrt{\\text{Hz}}$, with further enhancement limited by the\nnoise floor of the lock-in amplifier and the laser damage threshold of the\noptical components. Exploration of the microwave and optical pump-rate\nparameter space demonstrates a linewidth-narrowing regime reached by virtue of\nusing the optical cavity, allowing an enhanced sensitivity to be achieved,\ndespite an unoptimized collection efficiency of <2 %, and a low\nnitrogen-vacancy concentration of about 0.2 ppb.\n",
"title": "Pump-Enhanced Continuous-Wave Magnetometry using Nitrogen-Vacancy Ensembles"
}
| null | null | null | null | true | null |
14510
| null |
Default
| null | null |
null |
{
"abstract": " Topological spin liquids are robust quantum states of matter with long-range\nentanglement and possess many exotic properties such as the fractional\nstatistics of the elementary excitations. Yet these states, short of local\nparameters like all topological states, are elusive for conventional\nexperimental probes. In this work, we combine theoretical analysis and quantum\nMonte Carlo numerics on a frustrated spin model which hosts a $\\mathbb Z_2$\ntopological spin liquid ground state, and demonstrate that the presence of\nsymmetry-protected gapless edge modes is a characteristic feature of the state,\noriginating from the nontrivial symmetry fractionalization of the elementary\nexcitations. Experimental observation of these modes on the edge would directly\nindicate the existence of the topological spin liquids in the bulk, analogous\nto the fact that the observation of Dirac edge states confirmed the existence\nof topological insulators.\n",
"title": "Topological Spin Liquid with Symmetry-Protected Edge States"
}
| null | null |
[
"Physics"
] | null | true | null |
14511
| null |
Validated
| null | null |
null |
{
"abstract": " In semi-symbolic (control-explicit data-symbolic) model checking the\nstate-space explosion problem is fought by representing sets of states by\nfirst-order formulas over the bit-vector theory. In this model checking\napproach, most of the verification time is spent in an SMT solver on deciding\nsatisfiability of quantified queries, which represent equality of symbolic\nstates. In this paper, we introduce a new scheme for decomposition of symbolic\nstates, which can be used to significantly improve the performance of any\nsemi-symbolic model checker. Using the decomposition, a model checker can issue\nmuch simpler and smaller queries to the solver when compared to the original\ncase. Some SMT calls may be even avoided completely, as the satisfaction of\nsome of the simplified formulas can be decided syntactically. Moreover, the\ndecomposition allows for an efficient caching scheme for quantified formulas.\nTo support our theoretical contribution, we show the performance gain of our\nmodel checker SymDIVINE on a set of examples from the Software Verification\nCompetition.\n",
"title": "SMT Queries Decomposition and Caching in Semi-Symbolic Model Checking"
}
| null | null | null | null | true | null |
14512
| null |
Default
| null | null |
null |
{
"abstract": " Dimer algebras arise from a particular type of quiver gauge theory. However,\npart of the input to such a theory is the gauge group, and this choice may\nimpose additional constraints on the algebra. If the gauge group of a dimer\ntheory is abelian, then the algebra that arises is not actually the dimer\nalgebra itself, but a particular quotient we introduce called the 'homotopy\nalgebra'. We show that a homotopy algebra $\\Lambda$ on a torus is a dimer\nalgebra if and only if it is noetherian, and otherwise $\\Lambda$ is the\nquotient of a dimer algebra by homotopy relations. Stated in physics terms, a\ndimer theory is superconformal if and only if the corresponding dimer and\nhomotopy algebras coincide. We also give an explicit description of the center\nof a homotopy algebra in terms of a special subset of its perfect matchings. In\nour proofs we introduce formalized notions of Higgsing and the mesonic chiral\nring from quiver gauge theory.\n",
"title": "Homotopy dimer algebras and cyclic contractions"
}
| null | null | null | null | true | null |
14513
| null |
Default
| null | null |
null |
{
"abstract": " The subject is traces of Sobolev spaces with mixed Lebesgue norms on\nEuclidean space. Specifically, restrictions to the hyperplanes given by the\nfirst and last coordinates are applied to functions belonging to\nquasi-homogeneous, mixed-norm Lizorkin--Triebel spaces; Sobolev spaces are\nobtained from these as special cases. Spaces admitting traces in the\ndistribution sense are characterised except for the borderline cases; these are\nalso covered in case of the first variable. With respect to the first variable\nthe trace spaces are proved to be mixed-norm Lizorkin--Triebel spaces with a\nspecific sum exponent. For the last variable they are similarly defined Besov\nspaces. The treatment includes continuous right-inverses and higher order\ntraces. The results rely on a sequence version of Nikolskij's inequality,\nMarschall's inequality for pseudo-differential operators (and Fourier\nmultiplier assertions), as well as dyadic ball criteria.\n",
"title": "On the trace problem for Triebel--Lizorkin spaces with mixed norms"
}
| null | null | null | null | true | null |
14514
| null |
Default
| null | null |
null |
{
"abstract": " Models with many signals, high-dimensional models, often impose structures on\nthe signal strengths. The common assumption is that only a few signals are\nstrong and most of the signals are zero or close (collectively) to zero.\nHowever, such a requirement might not be valid in many real-life applications.\nIn this article, we are interested in conducting large-scale inference in\nmodels that might have signals of mixed strengths. The key challenge is that\nthe signals that are not under testing might be collectively non-negligible\n(although individually small) and cannot be accurately learned. This article\ndevelops a new class of tests that arise from a moment matching formulation. A\nvirtue of these moment-matching statistics is their ability to borrow strength\nacross features, adapt to the sparsity size and exert adjustment for testing\ngrowing number of hypothesis. GRoup-level Inference of Parameter, GRIP, test\nharvests effective sparsity structures with hypothesis formulation for an\nefficient multiple testing procedure. Simulated data showcase that GRIPs error\ncontrol is far better than the alternative methods. We develop a minimax\ntheory, demonstrating optimality of GRIP for a broad range of models, including\nthose where the model is a mixture of a sparse and high-dimensional dense\nsignals.\n",
"title": "Breaking the curse of dimensionality in regression"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
14515
| null |
Validated
| null | null |
null |
{
"abstract": " Training of discrete latent variable models remains challenging because\npassing gradient information through discrete units is difficult. We propose a\nnew class of smoothing transformations based on a mixture of two overlapping\ndistributions, and show that the proposed transformation can be used for\ntraining binary latent models with either directed or undirected priors. We\nderive a new variational bound to efficiently train with Boltzmann machine\npriors. Using this bound, we develop DVAE++, a generative model with a global\ndiscrete prior and a hierarchy of convolutional continuous variables.\nExperiments on several benchmarks show that overlapping transformations\noutperform other recent continuous relaxations of discrete latent variables\nincluding Gumbel-Softmax (Maddison et al., 2016; Jang et al., 2016), and\ndiscrete variational autoencoders (Rolfe 2016).\n",
"title": "DVAE++: Discrete Variational Autoencoders with Overlapping Transformations"
}
| null | null |
[
"Statistics"
] | null | true | null |
14516
| null |
Validated
| null | null |
null |
{
"abstract": " Biological organisms have to cope with stochastic variations in both the\nexternal environment and the internal population dynamics. Theoretical studies\nand laboratory experiments suggest that population diversification could be an\neffective bet-hedging strategy for adaptation to varying environments. Here we\nshow that bet-hedging can also be effective against demographic fluctuations\nthat pose a trade-off between growth and survival for populations even in a\nconstant environment. A species can maximize its overall abundance in the long\nterm by diversifying into coexisting subpopulations of both \"fast-growing\" and\n\"better-surviving\" individuals. Our model generalizes statistical physics\nmodels of birth-death processes to incorporate dispersal, during which new\npopulations are founded, and can further incorporate variations of local\nenvironments. In this way we unify different bet-hedging strategies against\ndemographic and environmental variations as a general means of adaptation to\nboth types of uncertainties in population growth.\n",
"title": "Bet-hedging against demographic fluctuations"
}
| null | null | null | null | true | null |
14517
| null |
Default
| null | null |
null |
{
"abstract": " To improve the performance of Intensive Care Units (ICUs), the field of\nbio-statistics has developed scores which try to predict the likelihood of\nnegative outcomes. These help evaluate the effectiveness of treatments and\nclinical practice, and also help to identify patients with unexpected outcomes.\nHowever, they have been shown by several studies to offer sub-optimal\nperformance. Alternatively, Deep Learning offers state of the art capabilities\nin certain prediction tasks and research suggests deep neural networks are able\nto outperform traditional techniques. Nevertheless, a main impediment for the\nadoption of Deep Learning in healthcare is its reduced interpretability, for in\nthis field it is crucial to gain insight on the why of predictions, to assure\nthat models are actually learning relevant features instead of spurious\ncorrelations. To address this, we propose a deep multi-scale convolutional\narchitecture trained on the Medical Information Mart for Intensive Care III\n(MIMIC-III) for mortality prediction, and the use of concepts from coalitional\ngame theory to construct visual explanations aimed to show how important these\ninputs are deemed by the network. Our results show our model attains state of\nthe art performance while remaining interpretable. Supporting code can be found\nat this https URL.\n",
"title": "ISeeU: Visually interpretable deep learning for mortality prediction inside the ICU"
}
| null | null | null | null | true | null |
14518
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes two low-complexity iterative algorithms to compute the\ncapacity of a single-user multiple-input multiple-output channel with\nper-antenna power constraint. The first method results from manipulating the\noptimality conditions of the considered problem and applying fixed-point\niteration. In the second approach, we transform the considered problem into a\nminimax optimization program using the well-known MAC- BC duality, and then\nsolve it by a novel alternating optimization method. In both proposed iterative\nmethods, each iteration involves an optimization problem which can be\nefficiently solved by the water-filling algorithm. The proposed iterative\nmethods are provably convergent. Complexity analysis and extensive numerical\nexperiments are carried out to demonstrate the superior performance of the\nproposed algorithms over an existing approach known as the mode-dropping\nalgorithm.\n",
"title": "Low-complexity Approaches for MIMO Capacity with Per-antenna Power Constraint"
}
| null | null | null | null | true | null |
14519
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we describe the routine photometric calibration of data taken\nwith the VIRCAM instrument on the ESO VISTA telescope. The broadband ZYJHKs\ndata are directly calibrated from 2MASS point sources visible in every VISTA\nimage. We present the empirical transformations between the 2MASS and VISTA,\nand WFCAM and VISTA, photometric systems for regions of low reddening. We\ninvestigate the long-term performance of VISTA+VIRCAM. An investigation of the\ndependence of the photometric calibration on interstellar reddening leads to\nthese conclusions: (1) For all broadband filters, a linear colour-dependent\ncorrection compensates the gross effects of reddening where $E(B-V)<5.0$. (2)\nFor $Z$ and $Y$, there is a significantly larger scatter above E(B-V)=5.0, and\ninsufficient measurements to adequately constrain the relation beyond this\nvalue. (3) The $JHK\\!s$ filters can be corrected to a few percent up to\nE(B-V)=10.0. We analyse spatial systematics over month-long timescales, both\ninter- and intra-detector and show that these are present only at very low\nlevels in VISTA. We monitor and remove residual detector-to-detector offsets.\nWe compare the calibration of the main pipeline products: pawprints and tiles.\nWe show how variable seeing and transparency affect the final calibration\naccuracy of VISTA tiles, and discuss a technique, {\\it grouting}, for\nmitigating these effects. Comparison between repeated reference fields is used\nto demonstrate that the VISTA photometry is precise to better than $\\simeq2\\%$\nfor the $Y$$J$$H$$Ks$ bands and $3\\%$ for the $Z$ bands. Finally we present\nempirically determined offsets to transform VISTA magnitudes into a true Vega\nsystem.\n",
"title": "The VISTA ZYJHKs Photometric System: Calibration from 2MASS"
}
| null | null | null | null | true | null |
14520
| null |
Default
| null | null |
null |
{
"abstract": " We report on the electronic transport and the impact of spin-filtering in\nmesa-structures made of epitaxial thin films of cuprate superconductor\nYBa2Cu3Ox(YBCO) and the manganite LaMnO3 (LMO) interlayer with the Au/Nb\ncounterelectrode. Ferromagnetic resonance measurements of heterostructure\nAu/LMO/YBCO shows ferromagnetic state at temperatures below 150 K as in the\ncase of reference LMO film grown on the neodymium gallate substrate. The\nheights of the tunneling barrier evaluated from resistive characteristics of\nmesa-structures at different thickness of interlayer showed an exponential\ndecrease from 30 mV down to 5 mV with the increase of manganite interlayer\nthickness. Temperature dependence of the conductivity of mesa-structures could\nbe described taking into account the d-wave superconductivity in YBCO and a\nspin filtering of the electron transport. Spin filtering is supported also by\nmeasurements of magneto-resistance and the high sensitivity of mesa-structure\nconductivity to weak magnetic fields.\n",
"title": "Spin-filtering in superconducting junction with the manganite interlayer"
}
| null | null |
[
"Physics"
] | null | true | null |
14521
| null |
Validated
| null | null |
null |
{
"abstract": " It is shown that for a solvable subgroup $G$ of an almost simple group $S$\nwhich socle is isomorphic to $A_n$ $ (n\\ge5)$ there are $x,y,z,t \\in S$ such\nthat $G \\cap G^x \\cap G^y \\cap G^z \\cap G^t =1.$\n",
"title": "Intersection of conjugate solvable subgroups in symmetric groups"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14522
| null |
Validated
| null | null |
null |
{
"abstract": " Gene expression (GE) data capture valuable condition-specific information\n(\"condition\" can mean a biological process, disease stage, age, patient, etc.)\nHowever, GE analyses ignore physical interactions between gene products, i.e.,\nproteins. Since proteins function by interacting with each other, and since\nbiological networks (BNs) capture these interactions, BN analyses are\npromising. However, current BN data fail to capture condition-specific\ninformation. Recently, GE and BN data have been integrated using network\npropagation (NP) to infer condition-specific BNs. However, existing NP-based\nstudies result in a static condition-specific network, even though cellular\nprocesses are dynamic. A dynamic process of our interest is aging. We use\nprominent existing NP methods in a new task of inferring a dynamic rather than\nstatic condition-specific (aging-related) network. Then, we study evolution of\nnetwork structure with age - we identify proteins whose network positions\nsignificantly change with age and predict them as new aging-related candidates.\nWe validate the predictions via e.g., functional enrichment analyses and\nliterature search. Dynamic network inference via NP yields higher prediction\nquality than the only existing method for inferring a dynamic aging-related BN,\nwhich does not use NP.\n",
"title": "Improving inference of the dynamic biological network underlying aging via network propagation"
}
| null | null | null | null | true | null |
14523
| null |
Default
| null | null |
null |
{
"abstract": " A problem faced by many instructors is that of designing exams that\naccurately assess the abilities of the students. Typically these exams are\nprepared several days in advance, and generic question scores are used based on\nrough approximation of the question difficulty and length. For example, for a\nrecent class taught by the author, there were 30 multiple choice questions\nworth 3 points, 15 true/false with explanation questions worth 4 points, and 5\nanalytical exercises worth 10 points. We describe a novel framework where\nalgorithms from machine learning are used to modify the exam question weights\nin order to optimize the exam scores, using the overall class grade as a proxy\nfor a student's true ability. We show that significant error reduction can be\nobtained by our approach over standard weighting schemes, and we make several\nnew observations regarding the properties of the \"good\" and \"bad\" exam\nquestions that can have impact on the design of improved future evaluation\nmethods.\n",
"title": "Optimal Weighting for Exam Composition"
}
| null | null | null | null | true | null |
14524
| null |
Default
| null | null |
null |
{
"abstract": " Most recent CNN architectures use average pooling as a final feature encoding\nstep. In the field of fine-grained recognition, however, recent global\nrepresentations like bilinear pooling offer improved performance. In this\npaper, we generalize average and bilinear pooling to \"alpha-pooling\", allowing\nfor learning the pooling strategy during training. In addition, we present a\nnovel way to visualize decisions made by these approaches. We identify parts of\ntraining images having the highest influence on the prediction of a given test\nimage. It allows for justifying decisions to users and also for analyzing the\ninfluence of semantic parts. For example, we can show that the higher capacity\nVGG16 model focuses much more on the bird's head than, e.g., the lower-capacity\nVGG-M model when recognizing fine-grained bird categories. Both contributions\nallow us to analyze the difference when moving between average and bilinear\npooling. In addition, experiments show that our generalized approach can\noutperform both across a variety of standard datasets.\n",
"title": "Generalized orderless pooling performs implicit salient matching"
}
| null | null | null | null | true | null |
14525
| null |
Default
| null | null |
null |
{
"abstract": " Abridged: We used the fourth internal data release of the Gaia-ESO survey to\ncharacterize the bulge chemistry, spatial distribution, kinematics, and to\ncompare it chemically with the thin and thick disks. The sample consist on\n~2500 red clump stars in 11 bulge fields ($-10^\\circ\\leq l\\leq+8^\\circ$ and\n$-10^\\circ\\leq b\\leq-4^\\circ$), and a set of ~6300 disk stars selected for\ncomparison. The bulge MDF is confirmed to be bimodal across the whole sampled\narea, with metal-poor stars dominating at high latitudes. The metal-rich stars\nexhibit bar-like kinematics and display a bimodality in their magnitude\ndistribution, a feature which is tightly associated with the X-shape bulge.\nThey overlap with the metal-rich end of the thin disk sequence in the [Mg/Fe]\nvs. [Fe/H] plane. Metal-poor bulge stars have a more isotropic hot kinematics\nand do not participate in the X-shape bulge. With similar Mg-enhancement\nlevels, the position of the metal-poor bulge sequence \"knee\" is observed at\n[Fe/H]$_{knee}=-0.37\\pm0.09$, being 0.06 dex higher than that of the thick\ndisk. It suggests a higher SFR for the bulge than for the thick disk. Finally,\nwe present a chemical evolution model that suitably fits the whole bulge\nsequence by assuming a fast ($<1$ Gyr) intense burst of stellar formation at\nearly epochs. We associate metal-rich stars with the B/P bulge formed from the\nsecular evolution of the early thin disk. On the other hand, the metal-poor\nsubpopulation might be the product of an early prompt dissipative collapse\ndominated by massive stars. Nevertheless, our results do not allow us to firmly\nrule out the possibility that these stars come from the secular evolution of\nthe early thick disk. This is the first time that an analysis of the bulge MDF\nand $\\alpha$-abundances has been performed in a large area on the basis of a\nhomogeneous, fully spectroscopic analysis of high-resolution, high S/N data.\n",
"title": "The Gaia-ESO Survey: Exploring the complex nature and origins of the Galactic bulge populations"
}
| null | null | null | null | true | null |
14526
| null |
Default
| null | null |
null |
{
"abstract": " Robust Stable Marriage (RSM) is a variant of the classical Stable Marriage\nproblem, where the robustness of a given stable matching is measured by the\nnumber of modifications required for repairing it in case an unforeseen event\noccurs. We focus on the complexity of finding an (a,b)-supermatch. An\n(a,b)-supermatch is defined as a stable matching in which if any 'a'\n(non-fixed) men/women break up it is possible to find another stable matching\nby changing the partners of those 'a' men/women and also the partners of at\nmost 'b' other couples. In order to show deciding if there exists an\n(a,b)-supermatch is NP-Complete, we first introduce a SAT formulation that is\nNP-Complete by using Schaefer's Dichotomy Theorem. Then, we show the\nequivalence between the SAT formulation and finding a (1,1)-supermatch on a\nspecific family of instances.\n",
"title": "On the Complexity of Robust Stable Marriage"
}
| null | null | null | null | true | null |
14527
| null |
Default
| null | null |
null |
{
"abstract": " We know that in empty space there is no preferred state of rest. This is true\nboth in special relativity but also in Newtonian mechanics with its associated\nGalilean relativity. It comes as something of a surprise, therefore, to\ndiscover the existence a friction force associated with spontaneous emission.\nhe resolution of this paradox relies on a central idea from special relativity\neven though our derivation of it is non-relativistic. We examine the\npossibility that the physics underlying this effect might be explored in an ion\ntrap, via the observation of a superposition of different mass states.\n",
"title": "Vacuum Friction"
}
| null | null | null | null | true | null |
14528
| null |
Default
| null | null |
null |
{
"abstract": " An Intelligent Personal Agent (IPA) is an agent that has the purpose of\nhelping the user to gain information through reliable resources with the help\nof knowledge navigation techniques and saving time to search the best content.\nThe agent is also responsible for responding to the chat-based queries with the\nhelp of Conversation Corpus. We will be testing different methods for optimal\nquery generation. To felicitate the ease of usage of the application, the agent\nwill be able to accept the input through Text (Keyboard), Voice (Speech\nRecognition) and Server (Facebook) and output responses using the same method.\nExisting chat bots reply by making changes in the input, but we will give\nresponses based on multiple SRT files. The model will learn using the human\ndialogs dataset and will be able respond human-like. Responses to queries about\nfamous things (places, people, and words) can be provided using web scraping\nwhich will enable the bot to have knowledge navigation features. The agent will\neven learn from its past experiences supporting semi-supervised learning.\n",
"title": "Intelligent Personal Assistant with Knowledge Navigation"
}
| null | null | null | null | true | null |
14529
| null |
Default
| null | null |
null |
{
"abstract": " With the increased use of Internet, governments and large companies store and\nshare massive amounts of personal data in such a way that leaves no space for\ntransparency. When a user needs to achieve a simple task like applying for\ncollege or a driving license, he needs to visit a lot of institutions and\norganizations, thus leaving a lot of private data in many places. The same\nhappens when using the Internet. These privacy issues raised by the centralized\narchitectures along with the recent developments in the area of serverless\napplications demand a decentralized private data layer under user control. We\nintroduce the Private Data System (PDS), a distributed approach which enables\nself-sovereign storage and sharing of private data. The system is composed of\nnodes spread across the entire Internet managing local key-value databases. The\ncommunication between nodes is achieved through executable choreographies,\nwhich are capable of preventing information leakage when executing across\ndifferent organizations with different regulations in place. The user has full\ncontrol over his private data and is able to share and revoke access to\norganizations at any time. Even more, the updates are propagated instantly to\nall the parties which have access to the data thanks to the system design.\nSpecifically, the processing organizations may retrieve and process the shared\ninformation, but are not allowed under any circumstances to store it on long\nterm. PDS offers an alternative to systems that aim to ensure self-sovereignty\nof specific types of data through blockchain inspired techniques but face\nvarious problems, such as low performance. Both approaches propose a\ndistributed database, but with different characteristics. While the\nblockchain-based systems are built to solve consensus problems, PDS's purpose\nis to solve the self-sovereignty aspects raised by the privacy laws, rules and\nprinciples.\n",
"title": "Private Data System Enabling Self-Sovereign Storage Managed by Executable Choreographies"
}
| null | null | null | null | true | null |
14530
| null |
Default
| null | null |
null |
{
"abstract": " We evaluate a curious determinant, first mentioned by George Andrews in 1980\nin the context of descending plane partitions. Our strategy is to combine the\nfamous Desnanot-Jacobi-Dodgson identity with automated proof techniques. More\nprecisely, we follow the holonomic ansatz that was proposed by Doron Zeilberger\nin 2007. We derive a compact and nice formula for Andrews's determinant, and\nuse it to solve a challenge problem that we posed in a previous paper. By\nnoting that Andrews's determinant is a special case of a two-parameter family\nof determinants, we find closed forms for several one-parameter subfamilies.\nThe interest in these determinants arises because they count cyclically\nsymmetric rhombus tilings of a hexagon with several triangular holes inside.\n",
"title": "A Curious Family of Binomial Determinants That Count Rhombus Tilings of a Holey Hexagon"
}
| null | null | null | null | true | null |
14531
| null |
Default
| null | null |
null |
{
"abstract": " This entry discusses the problem of describing some communities identified in\na complex network of interest, in a way allowing to interpret them. We suppose\nthe community structure has already been detected through one of the many\nmethods proposed in the literature. The question is then to know how to extract\nvaluable information from this first result, in order to allow human\ninterpretation. This requires subsequent processing, which we describe in the\nrest of this entry.\n",
"title": "Community Structure Characterization"
}
| null | null | null | null | true | null |
14532
| null |
Default
| null | null |
null |
{
"abstract": " This paper introduces deep neural networks (DNNs) as add-on blocks to\nbaseline feedback control systems to enhance tracking performance of arbitrary\ndesired trajectories. The DNNs are trained to adapt the reference signals to\nthe feedback control loop. The goal is to achieve a unity map between the\ndesired and the actual outputs. In previous work, the efficacy of this approach\nwas demonstrated on quadrotors; on 30 unseen test trajectories, the proposed\nDNN approach achieved an average impromptu tracking error reduction of 43% as\ncompared to the baseline feedback controller. Motivated by these results, this\nwork aims to provide platform-independent design guidelines for the proposed\nDNN-enhanced control architecture. In particular, we provide specific\nguidelines for the DNN feature selection, derive conditions for when the\nproposed approach is effective, and show in which cases the training efficiency\ncan be further increased.\n",
"title": "Design of Deep Neural Networks as Add-on Blocks for Improving Impromptu Trajectory Tracking"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14533
| null |
Validated
| null | null |
null |
{
"abstract": " We apply three data science techniques, Nonnegative Matrix Factorization\n(NMF), Principal Component Analysis (PCA) and Independent Component Analysis\n(ICA), to simulated X-ray energy spectra of a particular class of super-massive\nblack holes. Two competing physical models, one whose variable components are\nadditive and the other whose variable components are multiplicative, are known\nto successfully describe X-ray spectral variation of these super-massive black\nholes, within accuracy of the contemporary observation. We hope to utilize\nthese techniques to compare the viability of the models by probing the\nmathematical structure of the observed spectra, while comparing advantages and\ndisadvantages of each technique. We find that PCA is best to determine the\ndimensionality of a dataset, while NMF is better suited for interpreting\nspectral components and comparing them in terms of the physical models in\nquestion. ICA is able to reconstruct the parameters responsible for spectral\nvariation. In addition, we find that the results of these techniques are\nsufficiently different that applying them to observed data may be a useful test\nin comparing the accuracy of the two spectral models.\n",
"title": "Application of data science techniques to disentangle X-ray spectral variation of super-massive black holes"
}
| null | null | null | null | true | null |
14534
| null |
Default
| null | null |
null |
{
"abstract": " Motivated by truncated EM method introduced by Mao (2015), a new explicit\nnumerical method named modified truncated Euler-Maruyama method is developed in\nthis paper. Strong convergence rates of the given numerical scheme to the exact\nsolutions to stochastic differential equations are investigated under given\nconditions in this paper. Compared with truncated EM method, the given\nnumerical simulation strongly converges to the exact solution at fixed time $T$\nand over a time interval $[0,T]$ under weaker sufficient conditions. Meanwhile,\nthe convergence rates are also obtained for both cases. Two examples are\nprovided to support our conclusions.\n",
"title": "Strong convergence rates of modified truncated EM method for stochastic differential equations"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14535
| null |
Validated
| null | null |
null |
{
"abstract": " This expository paper is concerned with the properties of proper holomorphic\nmappings between domains in complex affine spaces. We discuss some of the main\ngeometric methods of this theory, such as the Reflection Principle, the scaling\nmethod, and the Kobayashi-Royden metric. We sketch the proofs of certain\nprincipal results and discuss some recent achievements. Several open problems\nare also stated.\n",
"title": "Some aspects of holomorphic mappings: a survey"
}
| null | null | null | null | true | null |
14536
| null |
Default
| null | null |
null |
{
"abstract": " Dual spectral computed tomography (DSCT) can achieve energy- and\nmaterial-selective images, and has a superior distinguishability of some\nmaterials than conventional single spectral computed tomography (SSCT).\nHowever, the decomposition process is illposed, which is sensitive with noise,\nthus the quality of decomposed images are usually degraded, especially the\nsignal-to-noise ratio (SNR) is much lower than single spectra based directly\nreconstructions. In this work, we first establish a local linear relationship\nbetween dual spectra based decomposed results and single spectra based directly\nreconstructed images. Then, based on this constraint, we propose an\noptimization model for DSCT and develop a guided image filtering based\niterative solution method. Both simulated and real experiments are provided to\nvalidate the effectiveness of the proposed approach.\n",
"title": "Local Linear Constraint based Optimization Model for Dual Spectral CT"
}
| null | null | null | null | true | null |
14537
| null |
Default
| null | null |
null |
{
"abstract": " The timed pattern matching problem is an actively studied topic because of\nits relevance in monitoring of real-time systems. There one is given a log $w$\nand a specification $\\mathcal{A}$ (given by a timed word and a timed automaton\nin this paper), and one wishes to return the set of intervals for which the log\n$w$, when restricted to the interval, satisfies the specification\n$\\mathcal{A}$. In our previous work we presented an efficient timed pattern\nmatching algorithm: it adopts a skipping mechanism inspired by the classic\nBoyer--Moore (BM) string matching algorithm. In this work we tackle the problem\nof online timed pattern matching, towards embedded applications where it is\nvital to process a vast amount of incoming data in a timely manner.\nSpecifically, we start with the Franek-Jennings-Smyth (FJS) string matching\nalgorithm---a recent variant of the BM algorithm---and extend it to timed\npattern matching. Our experiments indicate the efficiency of our FJS-type\nalgorithm in online and offline timed pattern matching.\n",
"title": "Efficient Online Timed Pattern Matching by Automata-Based Skipping"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14538
| null |
Validated
| null | null |
null |
{
"abstract": " We report on $g$, $r$ and $i$ band observations of the Interstellar Object\n'Oumuamua (1I) taken on 2017 October 29 from 04:28 to 08:40 UTC by the Apache\nPoint Observatory (APO) 3.5m telescope's ARCTIC camera. We find that 1I's\ncolors are $g-r=0.41\\pm0.24$ and $r-i=0.23\\pm0.25$, consistent with the visible\nspectra of Masiero (2017), Ye et al. (2017) and Fitzsimmons et al. (2017), and\nmost comparable to the population of Solar System C/D asteroids, Trojans, or\ncomets. We find no evidence of any cometary activity at a heliocentric distance\nof 1.46 au, approximately 1.5 months after 1I's closest approach distance to\nthe Sun. Significant brightness variability was seen in the $r$ observations,\nwith the object becoming notably brighter towards the end of the run. By\ncombining our APO photometric time series data with the Discovery Channel\nTelescope (DCT) data of Knight et al. (2017), taken 20 h later on 2017 October\n30, we construct an almost complete light curve with a most probable lightcurve\nperiod of $P \\simeq 4~{\\rm h}$. Our results imply a double peaked rotation\nperiod of 8.1 $\\pm$ 0.02 h, with a peak-to-peak amplitude of 1.5 - 2.1 mags.\nAssuming that 1I's shape can be approximated by an ellipsoid, the amplitude\nconstraint implies that 1I has an axial ratio of 3.5 to 10.3, which is\nstrikingly elongated. Assuming that 1I is rotating above its critical break up\nlimit, our results are compatible with 1I having having modest cohesive\nstrength and may have obtained its elongated shape during a tidal disruption\nevent before being ejected from its home system. Astrometry useful for\nconstraining 1I's orbit was also obtained and published in Weaver et al.\n(2017).\n",
"title": "APO Time Resolved Color Photometry of Highly-Elongated Interstellar Object 1I/'Oumuamua"
}
| null | null | null | null | true | null |
14539
| null |
Default
| null | null |
null |
{
"abstract": " We propose a method for variable selection in discriminant analysis with\nmixed categorical and continuous variables. This method is based on a criterion\nthat permits to reduce the variable selection problem to a problem of\nestimating suitable permutation and dimensionality. Then, estimators for these\nparameters are proposed and the resulting method for selecting variables is\nshown to be consistent. A simulation study that permits to study several\npoperties of the proposed approach and to compare it with an existing method is\ngiven.\n",
"title": "Variable selection in discriminant analysis for mixed variables and several groups"
}
| null | null | null | null | true | null |
14540
| null |
Default
| null | null |
null |
{
"abstract": " Generative Adversarial Networks (GAN) (Goodfellow et al., 2014) are an\neffective method for training generative models of complex data such as natural\nimages. However, they are notoriously hard to train and can suffer from the\nproblem of missing modes where the model is not able to produce examples in\ncertain regions of the space. We propose an iterative procedure, called AdaGAN,\nwhere at every step we add a new component into a mixture model by running a\nGAN algorithm on a reweighted sample. This is inspired by boosting algorithms,\nwhere many potentially weak individual predictors are greedily aggregated to\nform a strong composite predictor. We prove that such an incremental procedure\nleads to convergence to the true distribution in a finite number of steps if\neach step is optimal, and convergence at an exponential rate otherwise. We also\nillustrate experimentally that this procedure addresses the problem of missing\nmodes.\n",
"title": "AdaGAN: Boosting Generative Models"
}
| null | null | null | null | true | null |
14541
| null |
Default
| null | null |
null |
{
"abstract": " A basic, and still largely unanswered, question in the context of Generative\nAdversarial Networks (GANs) is whether they are truly able to capture all the\nfundamental characteristics of the distributions they are trained on. In\nparticular, evaluating the diversity of GAN distributions is challenging and\nexisting methods provide only a partial understanding of this issue. In this\npaper, we develop quantitative and scalable tools for assessing the diversity\nof GAN distributions. Specifically, we take a classification-based perspective\nand view loss of diversity as a form of covariate shift introduced by GANs. We\nexamine two specific forms of such shift: mode collapse and boundary\ndistortion. In contrast to prior work, our methods need only minimal human\nsupervision and can be readily applied to state-of-the-art GANs on large,\ncanonical datasets. Examining popular GANs using our tools indicates that these\nGANs have significant problems in reproducing the more distributional\nproperties of their training dataset.\n",
"title": "A Classification-Based Study of Covariate Shift in GAN Distributions"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14542
| null |
Validated
| null | null |
null |
{
"abstract": " We present a recurrent encoder-decoder deep neural network architecture that\ndirectly translates speech in one language into text in another. The model does\nnot explicitly transcribe the speech into text in the source language, nor does\nit require supervision from the ground truth source language transcription\nduring training. We apply a slightly modified sequence-to-sequence with\nattention architecture that has previously been used for speech recognition and\nshow that it can be repurposed for this more complex task, illustrating the\npower of attention-based models. A single model trained end-to-end obtains\nstate-of-the-art performance on the Fisher Callhome Spanish-English speech\ntranslation task, outperforming a cascade of independently trained\nsequence-to-sequence speech recognition and machine translation models by 1.8\nBLEU points on the Fisher test set. In addition, we find that making use of the\ntraining data in both languages by multi-task training sequence-to-sequence\nspeech translation and recognition models with a shared encoder network can\nimprove performance by a further 1.4 BLEU points.\n",
"title": "Sequence-to-Sequence Models Can Directly Translate Foreign Speech"
}
| null | null | null | null | true | null |
14543
| null |
Default
| null | null |
null |
{
"abstract": " Personalized search has been a hot research topic for many years and has been\nwidely used in e-commerce. This paper describes our solution to tackle the\nchallenge of personalized e-commerce search at CIKM Cup 2016. The goal of this\ncompetition is to predict search relevance and re-rank the result items in SERP\naccording to the personalized search, browsing and purchasing preferences.\nBased on a detailed analysis of the provided data, we extract three different\ntypes of features, i.e., statistic features, query-item features and session\nfeatures. Different models are used on these features, including logistic\nregression, gradient boosted decision trees, rank svm and a novel deep match\nmodel. With the blending of multiple models, a stacking ensemble model is built\nto integrate the output of individual models and produce a more accurate\nprediction result. Based on these efforts, our solution won the champion of the\ncompetition on all the evaluation metrics.\n",
"title": "Ensemble Methods for Personalized E-Commerce Search Challenge at CIKM Cup 2016"
}
| null | null | null | null | true | null |
14544
| null |
Default
| null | null |
null |
{
"abstract": " Sobol' sequences are widely used for quasi-Monte Carlo methods that arise in\nfinancial applications. Sobol' sequences have parameter values called direction\nnumbers, which are freely chosen by the user, so there are several\nimplementations of Sobol' sequence generators. The aim of this paper is to\nprovide a comparative study of (non-commercial) high-dimensional Sobol'\nsequences by calculating financial models. Additionally, we implement the\nNiederreiter sequence (in base 2) with a slight modification, that is, we\nreorder the rows of the generating matrices, and analyze and compare it with\nthe Sobol' sequences.\n",
"title": "Comparison of Sobol' sequences in financial applications"
}
| null | null | null | null | true | null |
14545
| null |
Default
| null | null |
null |
{
"abstract": " A high order wavelet integral collocation method (WICM) is developed for\ngeneral nonlinear boundary value problems in physics. This method is\nestablished based on Coiflet approximation of multiple integrals of interval\nbounded functions combined with an accurate and adjustable boundary extension\ntechnique. The convergence order of this approximation has been proven to be N\nas long as the Coiflet with N-1 vanishing moment is adopted, which can be any\npositive even integers. Before the conventional collocation method is applied\nto the general problems, the original differential equation is changed into its\nequivalent form by denoting derivatives of the unknown function as new\nfunctions and constructing relations between the low and high order\nderivatives. For the linear cases, error analysis has proven that the proposed\nWICM is order N, and condition numbers of relevant matrices are almost\nindependent of the number of collocation points. Numerical examples of a wide\nrange of nonlinear differential equations in physics demonstrate that accuracy\nof the proposed WICM is even greater than N, and most interestingly, such\naccuracy is independent of the order of the differential equation to be solved.\nComparison to existing numerical methods further justifies the accuracy and\nefficiency of the proposed method.\n",
"title": "A wavelet integral collocation method for nonlinear boundary value problems in Physics"
}
| null | null | null | null | true | null |
14546
| null |
Default
| null | null |
null |
{
"abstract": " Weakly-coupled TeV-scale particles may mediate the interactions between\nnormal matter and dark matter. If so, the LHC would produce dark matter through\nthese mediators, leading to the familiar \"mono-X\" search signatures, but the\nmediators would also produce signals without missing momentum via the same\nvertices involved in their production. This document from the LHC Dark Matter\nWorking Group suggests how to compare searches for these two types of signals\nin case of vector and axial-vector mediators, based on a workshop that took\nplace on September 19/20, 2016 and subsequent discussions. These suggestions\ninclude how to extend the spin-1 mediated simplified models already in\nwidespread use to include lepton couplings. This document also provides\nanalytic calculations of the relic density in the simplified models and reports\nan issue that arose when ATLAS and CMS first began to use preliminary numerical\ncalculations of the dark matter relic density in these models.\n",
"title": "Recommendations of the LHC Dark Matter Working Group: Comparing LHC searches for heavy mediators of dark matter production in visible and invisible decay channels"
}
| null | null | null | null | true | null |
14547
| null |
Default
| null | null |
null |
{
"abstract": " We present generalized versions of the concepts of seniority number and\nionicity. These generalized numbers count respectively the partially occupied\nand fully occupied shells for any partition of the orbital space into shells.\nThe Hermitian operators whose eigenspaces correspond to wave functions of\ndefinite generalized seniority or ionicity values are introduced. The\ngeneralized seniority numbers (GSNs) afford to establish refined hierarchies of\nconfiguration interaction (CI) spaces within those of fixed ordinary seniority.\nSuch a hierarchy is illustrated on the buckminsterfullerene molecule.\n",
"title": "Generalization of the concepts of seniority number and ionicity"
}
| null | null | null | null | true | null |
14548
| null |
Default
| null | null |
null |
{
"abstract": " As the size and complexity of software systems increase, the number and\nsophistication of software security flaws increase as well. The analysis of\nthese flaws began as a manual approach, but it soon became apparent that tools\nwere necessary to assist human experts in this task, resulting in a number of\ntechniques and approaches that automated aspects of the vulnerability analysis\nprocess.\nRecently, DARPA carried out the Cyber Grand Challenge, a competition among\nautonomous vulnerability analysis systems designed to push the tool-assisted\nhuman-centered paradigm into the territory of complete automation. However,\nwhen the autonomous systems were pitted against human experts it became clear\nthat certain tasks, albeit simple, could not be carried out by an autonomous\nsystem, as they require an understanding of the logic of the application under\nanalysis.\nBased on this observation, we propose a shift in the vulnerability analysis\nparadigm, from tool-assisted human-centered to human-assisted tool-centered. In\nthis paradigm, the automated system orchestrates the vulnerability analysis\nprocess, and leverages humans (with different levels of expertise) to perform\nwell-defined sub-tasks, whose results are integrated in the analysis. As a\nresult, it is possible to scale the analysis to a larger number of programs,\nand, at the same time, optimize the use of expensive human resources.\nIn this paper, we detail our design for a human-assisted automated\nvulnerability analysis system, describe its implementation atop an open-sourced\nautonomous vulnerability analysis system that participated in the Cyber Grand\nChallenge, and evaluate and discuss the significant improvements that\nnon-expert human assistance can offer to automated analysis approaches.\n",
"title": "Rise of the HaCRS: Augmenting Autonomous Cyber Reasoning Systems with Human Assistance"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14549
| null |
Validated
| null | null |
null |
{
"abstract": " We propose two coded schemes for the distributed computing problem of\nmultiplying a matrix by a set of vectors. The first scheme is based on\npartitioning the matrix into submatrices and applying maximum distance\nseparable (MDS) codes to each submatrix. For this scheme, we prove that up to a\ngiven number of partitions the communication load and the computational delay\n(not including the encoding and decoding delay) are identical to those of the\nscheme recently proposed by Li et al., based on a single, long MDS code.\nHowever, due to the use of shorter MDS codes, our scheme yields a significantly\nlower overall computational delay when the delay incurred by encoding and\ndecoding is also considered. We further propose a second coded scheme based on\nLuby Transform (LT) codes under inactivation decoding. Interestingly, LT codes\nmay reduce the delay over the partitioned scheme at the expense of an increased\ncommunication load. We also consider distributed computing under a deadline and\nshow numerically that the proposed schemes outperform other schemes in the\nliterature, with the LT code-based scheme yielding the best performance for the\nscenarios considered.\n",
"title": "Block-Diagonal and LT Codes for Distributed Computing With Straggling Servers"
}
| null | null | null | null | true | null |
14550
| null |
Default
| null | null |
null |
{
"abstract": " We study the Gevrey character of a natural parameterization of one\ndimensional invariant manifolds associated to a parabolic direction of fixed\npoints of analytic maps, that is, a direction associated with an eigenvalue\nequal to $1$. We show that, under general hypotheses, these invariant manifolds\nare Gevrey with type related to some explicit constants. We provide examples of\nthe optimality of our results as well as some applications to celestial\nmechanics, namely, the Sitnikov problem and the restricted planar three body\nproblem.\n",
"title": "Gevrey estimates for one dimensional parabolic invariant manifolds of non-hyperbolic fixed points"
}
| null | null | null | null | true | null |
14551
| null |
Default
| null | null |
null |
{
"abstract": " On a periodic basis, publicly traded companies are required to report\nfundamentals: financial data such as revenue, operating income, debt, among\nothers. These data points provide some insight into the financial health of a\ncompany. Academic research has identified some factors, i.e. computed features\nof the reported data, that are known through retrospective analysis to\noutperform the market average. Two popular factors are the book value\nnormalized by market capitalization (book-to-market) and the operating income\nnormalized by the enterprise value (EBIT/EV). In this paper: we first show\nthrough simulation that if we could (clairvoyantly) select stocks using factors\ncalculated on future fundamentals (via oracle), then our portfolios would far\noutperform a standard factor approach. Motivated by this analysis, we train\ndeep neural networks to forecast future fundamentals based on a trailing\n5-years window. Quantitative analysis demonstrates a significant improvement in\nMSE over a naive strategy. Moreover, in retrospective analysis using an\nindustry-grade stock portfolio simulator (backtester), we show an improvement\nin compounded annual return to 17.1% (MLP) vs 14.4% for a standard factor\nmodel.\n",
"title": "Improving Factor-Based Quantitative Investing by Forecasting Company Fundamentals"
}
| null | null | null | null | true | null |
14552
| null |
Default
| null | null |
null |
{
"abstract": " A new pathway to nuclear magnetic resonance spectroscopy in high pressure\ndiamond anvil cells is introduced, using inductively coupled broadband passive\nelectro-magnetic lenses to locally amplify the magnetic flux at the isolated\nsample, leading to an increase in sensitivity. The lenses are adopted for the\ngeometrical restrictions imposed by a toroidal diamond indenter cell, and yield\nhigh signal-to-noise ratios at pressures as high as 72 GPa, at initial sample\nvolumes of only 230 pl. The corresponding levels of detection, LODt, are found\nto be up to four orders of magnitude lower compared to formerly used solenoidal\nmicro-coils in diamond anvil cells, as shown by Proton-NMR measurements on\nparaffin oil. This approach opens up the field of ultra-high pressure sciences\nfor one of the most versatile spectroscopic methods available in a pressure\nrange unprecedended up to now.\n",
"title": "Magnetic Flux Tailoring through Lenz Lenses in Toroidal Diamond Indenter Cells: A New Pathway to High Pressure Nuclear Magnetic Resonance"
}
| null | null |
[
"Physics"
] | null | true | null |
14553
| null |
Validated
| null | null |
null |
{
"abstract": " A meticulous assessment of the risk of extreme environmental events is of\ngreat necessity for populations, civil authorities as well as the\ninsurance/reinsurance industry. Koch (2017, 2018) introduced a concept of\nspatial risk measure and a related set of axioms which are well-suited to\nanalyse and quantify the risk due to events having a spatial extent, precisely\nsuch as natural disasters. In this paper, we first carry out a detailed study\nof the correlation (and covariance) structure of powers of the Smith and\nBrown-Resnick max-stable random fields. Then, using the latter results, we\nthoroughly investigate spatial risk measures associated with variance and\ninduced by powers of max-stable random fields. In addition, we show that\nspatial risk measures associated with several classical risk measures and\ninduced by such cost fields satisfy (at least) part of the previously mentioned\naxioms under appropriate conditions on the max-stable fields. Considering such\ncost fields is particularly relevant when studying the impact of extreme wind\nspeeds on buildings and infrastructure.\n",
"title": "Spatial risk measures induced by powers of max-stable random fields"
}
| null | null | null | null | true | null |
14554
| null |
Default
| null | null |
null |
{
"abstract": " This paper deals with the homogenization problem of one-dimensional\npseudo-elliptic equations with a rapidly varying random potential. The main\npurpose is to characterize the homogenization error (random fluctuations),\ni.e., the difference between the random solution and the homogenized solution,\nwhich strongly depends on the autocovariance property of the underlying random\npotential. It is well known that when the random potential has short-range\ndependence, the rescaled homogenization error converges in distribution to a\nstochastic integral with respect to standard Brownian motion. Here, we are\ninterested in potentials with long-range dependence and we prove convergence to\nstochastic integrals with respect to Hermite process.\n",
"title": "Fluctuations in 1D stochastic homogenization of pseudo-elliptic equations with long-range dependent potentials"
}
| null | null | null | null | true | null |
14555
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, Runge-Kutta-Gegenbauer (RKG) stability polynomials of\narbitrarily high order of accuracy are introduced in closed form. The stability\ndomain of RKG polynomials extends in the the real direction with the square of\npolynomial degree, and in the imaginary direction as an increasing function of\nGegenbauer parameter. Consequently, the polynomials are naturally suited to the\nconstruction of high order stabilized Runge-Kutta (SRK) methods for systems of\nPDEs of mixed hyperbolic-parabolic type.\nWe present SRK methods composed of $L$ ordered forward Euler stages, with\ncomplex-valued stepsizes derived from the roots of RKG stability polynomials of\ndegree $L$. Internal stability is maintained at large stage number through an\nordering algorithm which limits internal amplification factors to $10 L^2$.\nTest results for mildly stiff nonlinear advection-diffusion-reaction problems\nwith moderate ($\\lesssim 1$) mesh Péclet numbers are provided at second,\nfourth, and sixth orders, with nonlinear reaction terms treated by complex\nsplitting techniques above second order.\n",
"title": "Runge-Kutta-Gegenbauer methods for advection-diffusion problems"
}
| null | null | null | null | true | null |
14556
| null |
Default
| null | null |
null |
{
"abstract": " We develop a tensor network technique that can solve universal reversible\nclassical computational problems, formulated as vertex models on a square\nlattice [Nat. Commun. 8, 15303 (2017)]. By encoding the truth table of each\nvertex constraint in a tensor, the total number of solutions compatible with\npartial inputs/outputs at the boundary can be represented as the full\ncontraction of a tensor network. We introduce an iterative\ncompression-decimation (ICD) scheme that performs this contraction efficiently.\nThe ICD algorithm first propagates local constraints to longer ranges via\nrepeated contraction-decomposition sweeps over all lattice bonds, thus\nachieving compression on a given length scale. It then decimates the lattice\nvia coarse-graining tensor contractions. Repeated iterations of these two steps\ngradually collapse the tensor network and ultimately yield the exact tensor\ntrace for large systems, without the need for manual control of tensor\ndimensions. Our protocol allows us to obtain the exact number of solutions for\ncomputations where a naive enumeration would take astronomically long times.\n",
"title": "Tensor network method for reversible classical computation"
}
| null | null | null | null | true | null |
14557
| null |
Default
| null | null |
null |
{
"abstract": " Recent advances in ultrafast measurement in cold atoms, as well as pump-probe\nspectroscopy of $K_3 C_{60}$ films, have opened the possibility of rapidly\nquenching systems of interacting fermions to, and across, a finite temperature\nsuperfluid transition. However, determining that a transient state has\napproached a second-order critical point is difficult, as standard equilibrium\ntechniques are inapplicable. We show that the approach to the superfluid\ncritical point in a transient state may be detected via time-resolved transport\nmeasurements, such as the optical conductivity. We leverage the fact that\nquenching to the vicinity of the critical point produces a highly time\ndependent density of superfluid fluctuations, which affect the conductivity in\ntwo ways. First, by inelastic scattering between the fermions and the\nfluctuations, and second by direct conduction through the fluctuations, with\nthe latter providing a lower resistance current carrying channel. The\ncompetition between these two effects leads to nonmonotonic behavior in the\ntime- resolved optical conductivity, providing a signature of the critical\ntransient state.\n",
"title": "Model Predictions for Time-Resolved Transport Measurements Made near the Superfluid Critical Points of Cold Atoms and $K_3C_{60}$ Films"
}
| null | null | null | null | true | null |
14558
| null |
Default
| null | null |
null |
{
"abstract": " Phonetic segmentation is the process of splitting speech into distinct\nphonetic units. Human experts routinely perform this task manually by analyzing\nauditory and visual cues using analysis software, which is an extremely\ntime-consuming process. Methods exist for automatic segmentation, but these are\nnot always accurate enough. In order to improve automatic segmentation, we need\nto model it as close to the manual segmentation as possible. This corpus is an\neffort to capture the human segmentation behavior by recording experts\nperforming a segmentation task. We believe that this data will enable us to\nhighlight the important aspects of manual segmentation, which can be used in\nautomatic segmentation to improve its accuracy.\n",
"title": "A Multimodal Corpus of Expert Gaze and Behavior during Phonetic Segmentation Tasks"
}
| null | null | null | null | true | null |
14559
| null |
Default
| null | null |
null |
{
"abstract": " Implicit models, which allow for the generation of samples but not for\npoint-wise evaluation of probabilities, are omnipresent in real-world problems\ntackled by machine learning and a hot topic of current research. Some examples\ninclude data simulators that are widely used in engineering and scientific\nresearch, generative adversarial networks (GANs) for image synthesis, and\nhot-off-the-press approximate inference techniques relying on implicit\ndistributions. The majority of existing approaches to learning implicit models\nrely on approximating the intractable distribution or optimisation objective\nfor gradient-based optimisation, which is liable to produce inaccurate updates\nand thus poor models. This paper alleviates the need for such approximations by\nproposing the Stein gradient estimator, which directly estimates the score\nfunction of the implicitly defined distribution. The efficacy of the proposed\nestimator is empirically demonstrated by examples that include meta-learning\nfor approximate inference, and entropy regularised GANs that provide improved\nsample diversity.\n",
"title": "Gradient Estimators for Implicit Models"
}
| null | null | null | null | true | null |
14560
| null |
Default
| null | null |
null |
{
"abstract": " Traditional centralized energy systems have the disadvantages of difficult\nmanagement and insufficient incentives. Blockchain is an emerging technology,\nwhich can be utilized in energy systems to enhance their management and\ncontrol. Integrating token economy and blockchain technology, token economic\nsystems in energy possess the characteristics of strong incentives and low\ncost, facilitating integrating renewable energy and demand side management, and\nproviding guarantees for improving energy efficiency and reducing emission.\nThis article describes the concept and functionality of token economics, and\nthen analyzes the feasibility of applying token economics in the energy\nsystems, and finally discuss the applications of token economics with an\nexample in integrated energy systems.\n",
"title": "Token Economics in Energy Systems: Concept, Functionality and Applications"
}
| null | null | null | null | true | null |
14561
| null |
Default
| null | null |
null |
{
"abstract": " Generative Adversarial Networks (GANs) are a powerful framework for deep\ngenerative modeling. Posed as a two-player minimax problem, GANs are typically\ntrained end-to-end on real-valued data and can be used to train a generator of\nhigh-dimensional and realistic images. However, a major limitation of GANs is\nthat training relies on passing gradients from the discriminator through the\ngenerator via back-propagation. This makes it fundamentally difficult to train\nGANs with discrete data, as generation in this case typically involves a\nnon-differentiable function. These difficulties extend to the reinforcement\nlearning setting when the action space is composed of discrete decisions. We\naddress these issues by reframing the GAN framework so that the generator is no\nlonger trained using gradients through the discriminator, but is instead\ntrained using a learned critic in the actor-critic framework with a Temporal\nDifference (TD) objective. This is a natural fit for sequence modeling and we\nuse it to achieve improvements on language modeling tasks over the standard\nTeacher-Forcing methods.\n",
"title": "ACtuAL: Actor-Critic Under Adversarial Learning"
}
| null | null | null | null | true | null |
14562
| null |
Default
| null | null |
null |
{
"abstract": " The introduction of serious games as pedagogical supports in the field of\neducation is a process gaining in popularity amongst the teaching community.\nThis article creates a link between the integration of new pedagogical\nsolutions in first-year primary class and the fundamental research on the\nmotivation of the players/learners, detailing an experiment based on a game\nspecifically developed, named QCM. QCM considers the learning worksheets issued\nfrom the Freinet pedagogy using various gameplay mechanisms. The main\ncontribution of QCM in relation to more traditional games is the dissociation\nof immersion mechanisms, in order to improve the understanding of the user\nexperience. This game also contains a system of gameplay metrics, the analysis\nof which shows a relative increase in the motivation of students using QCM\ninstead of paper worksheets, while revealing large differences in students\nbehavior in conjunction with the mechanisms of gamification employed. Keywords\n: Serious games, learning analytics, gamification, flow.\n",
"title": "Influence des mécanismes dissociés de ludifications sur l'apprentissage en support numérique de la lecture en classe primaire"
}
| null | null | null | null | true | null |
14563
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we investigate the complexity of one-dimensional dynamic\nprogramming, or more specifically, of the Least-Weight Subsequence (LWS)\nproblem: Given a sequence of $n$ data items together with weights for every\npair of the items, the task is to determine a subsequence $S$ minimizing the\ntotal weight of the pairs adjacent in $S$. A large number of natural problems\ncan be formulated as LWS problems, yielding obvious $O(n^2)$-time solutions.\nIn many interesting instances, the $O(n^2)$-many weights can be succinctly\nrepresented. Yet except for near-linear time algorithms for some specific\nspecial cases, little is known about when an LWS instantiation admits a\nsubquadratic-time algorithm and when it does not. In particular, no lower\nbounds for LWS instantiations have been known before. In an attempt to remedy\nthis situation, we provide a general approach to study the fine-grained\ncomplexity of succinct instantiations of the LWS problem. In particular, given\nan LWS instantiation we identify a highly parallel core problem that is\nsubquadratically equivalent. This provides either an explanation for the\napparent hardness of the problem or an avenue to find improved algorithms as\nthe case may be.\nMore specifically, we prove subquadratic equivalences between the following\npairs (an LWS instantiation and the corresponding core problem) of problems: a\nlow-rank version of LWS and minimum inner product, finding the longest chain of\nnested boxes and vector domination, and a coin change problem which is closely\nrelated to the knapsack problem and (min,+)-convolution. Using these\nequivalences and known SETH-hardness results for some of the core problems, we\ndeduce tight conditional lower bounds for the corresponding LWS instantiations.\nWe also establish the (min,+)-convolution-hardness of the knapsack problem.\n",
"title": "On the Fine-grained Complexity of One-Dimensional Dynamic Programming"
}
| null | null | null | null | true | null |
14564
| null |
Default
| null | null |
null |
{
"abstract": " It has recently been found that bosonic excitations of ordered media, such as\nphonons or spinons, can exhibit topologically nontrivial band structures. Of\nparticular interest are magnon and triplon excitations in quantum magnets, as\nthey can easily be manipulated by an applied field. Here we study triplon\nexcitations in an S=1/2 quantum spin ladder and show that they exhibit\nnontrivial topology, even in the quantum-disordered paramagnetic phase. Our\nanalysis reveals that the paramagnetic phase actually consists of two separate\nregions with topologically distinct triplon excitations. We demonstrate that\nthe topological transition between these two regions can be tuned by an\nexternal magnetic field. The winding number that characterizes the topology of\nthe triplons is derived and evaluated. By the bulk-boundary correspondence, we\nfind that the non-zero winding number implies the presence of localized triplon\nend states. Experimental signatures and possible physical realizations of the\ntopological paramagnetic phase are discussed.\n",
"title": "Topological quantum paramagnet in a quantum spin ladder"
}
| null | null | null | null | true | null |
14565
| null |
Default
| null | null |
null |
{
"abstract": " The dynamic dipole polarizabilities of the low-lying states of Ca$^{+}$ for\nlinearly and circularly polarized light are calculated by using relativistic\nconfiguration interaction plus core polarization (RCICP) approach. The magic\nwavelengths, at which the two levels of the transitions have the same ac Stark\nshifts, for $4s$-$4p_{j,m}$ and $4s$-$3d_{j,m}$ magnetic sublevels transitions\nare determined. The present magic wavelengths for linearly polarized light\nagree with the available results excellently. The polarizability for the\ncircularly polarized light has the scalar, vector and tensor components. The\ndynamic polarizability is different for each of magnetic sublevels of the\natomic state. Additional magic wavelengths have been found for the circularly\npolarized light. We recommend that the measurement of the magic wavelength near\n850 nm for $4s-4p_{\\frac32,m=\\pm\\frac32,\\pm\\frac12}$ could be able to determine\nthe oscillator strength ratio of $f_{4p_{\\frac32} \\to 3d_{\\frac32}}$ and\n$f_{4p_{\\frac32} \\to 3d_{\\frac52}}$.\n",
"title": "Magic wavelengths of Ca$^{+}$ ion for linearly and circularly polarized light"
}
| null | null | null | null | true | null |
14566
| null |
Default
| null | null |
null |
{
"abstract": " The fifth generation of mobile communications is anticipated to open up\ninnovation opportunities for new industries such as vertical markets. However,\nthese verticals originate myriad use cases with diverging requirements that\nfuture 5G networks have to efficiently support. Network slicing may be a\nnatural solution to simultaneously accommodate over a common network\ninfrastructure the wide range of services that vertical-specific use cases will\ndemand. In this article, we present the network slicing concept, with a\nparticular focus on its application to 5G systems. We start by summarizing the\nkey aspects that enable the realization of so-called network slices. Then, we\ngive a brief overview on the SDN architecture proposed by the ONF and show that\nit provides tools to support slicing. We argue that although such architecture\npaves the way for network slicing implementation, it lacks some essential\ncapabilities that can be supplied by NFV. Hence, we analyze a proposal from the\nETSI to incorporate the capabilities of SDN into the NFV architecture.\nAdditionally, we present an example scenario that combines SDN and NFV\ntechnologies to address the realization of network slices. Finally, we\nsummarize the open research issues with the purpose of motivating new advances\nin this field.\n",
"title": "Network Slicing for 5G with SDN/NFV: Concepts, Architectures and Challenges"
}
| null | null | null | null | true | null |
14567
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the structural, electronic, transport, and thermoelectric\nproperties of LaNiO$_3$/SrTiO$_3(001)$ superlattices containing either\nexclusively $n$- or $p$-type interfaces or coupled interfaces of opposite\npolarity by using density functional theory calculations with an on-site\nCoulomb repulsion term. The results show that significant octahedral tilts are\ninduced in the SrTiO$_3$ part of the superlattice. Moreover, the La-Sr\ndistances and Ni-O out-of-plane bond lengths at the interfaces exhibit a\ndistinct variation by about $7\\,\\%$ with the sign of the electrostatic doping.\nIn contrast to the much studied LaAlO$_3$/SrTiO$_3$ system, the charge mismatch\nat the interfaces is exclusively accommodated within the LaNiO$_3$ layers,\nwhereas the interface polarity leads to a band offset and to the formation of\nan electric field within the coupled superlattice. Features of the electronic\nstructure indicate an orbital-selective quantization of quantum well states.\nThe potential- and confinement-induced multiband splitting results in complex\ncylindrical Fermi surfaces with a tendency towards nesting that depends on the\ninterface polarity. The analysis of the thermoelectric response reveals a\nparticularly large positive Seebeck coefficient ($135~\\mu$V/K) and a high\nfigure of merit ($0.35$) for room-temperature cross-plane transport in the\n$p$-type superlattice that is attributed to the participation of the SrTiO$_3$\nvalence band. Superlattices with either $n$- or $p$-type interfaces show\ncross-plane Seebeck coefficients of opposite sign and thus emerge as a platform\nto construct an oxide-based thermoelectric generator with structurally and\nelectronically compatible $n$- and $p$-type oxide thermoelectrics.\n",
"title": "Design of $n$- and $p$-type oxide thermoelectrics in LaNiO$_3$/SrTiO$_3(001)$ superlattices exploiting interface polarity"
}
| null | null |
[
"Physics"
] | null | true | null |
14568
| null |
Validated
| null | null |
null |
{
"abstract": " In this work exact solutions for the equation that describes anomalous heat\npropagation in 1D harmonic lattices are obtained. Rectangular, triangular, and\nsawtooth initial perturbations of the temperature field are considered. The\nsolution for an initially rectangular temperature profile is investigated in\ndetail. It is shown that the decay of the solution near the wavefront is\nproportional to $1/ \\sqrt{t}$. In the center of the perturbation zone the decay\nis proportional to $1/t$. Thus the solution decays slower near the wavefront,\nleaving clearly visible peaks that can be detected experimentally.\n",
"title": "Localized heat perturbation in harmonic 1D crystals. Solutions for an equation of anomalous heat conduction"
}
| null | null | null | null | true | null |
14569
| null |
Default
| null | null |
null |
{
"abstract": " Recent research in psycholinguistics has provided increasing evidence that\nhumans predict upcoming content. Prediction also affects perception and might\nbe a key to robustness in human language processing. In this paper, we\ninvestigate the factors that affect human prediction by building a\ncomputational model that can predict upcoming discourse referents based on\nlinguistic knowledge alone vs. linguistic knowledge jointly with common-sense\nknowledge in the form of scripts. We find that script knowledge significantly\nimproves model estimates of human predictions. In a second study, we test the\nhighly controversial hypothesis that predictability influences referring\nexpression type but do not find evidence for such an effect.\n",
"title": "Modeling Semantic Expectation: Using Script Knowledge for Referent Prediction"
}
| null | null | null | null | true | null |
14570
| null |
Default
| null | null |
null |
{
"abstract": " The standard cosmographic approach consists in performing a series expansion\nof a cosmological observable around $z=0$ and then using the data to constrain\nthe cosmographic (or kinematic) parameters at present time. Such a procedure\nworks well if applied to redshift ranges inside the $z$-series convergence\nradius ($z<1$), but can be problematic if we want to cover redshift intervals\nthat fall outside the $z-$series convergence radius. This problem can be\ncircumvented if we work with the $y-$redshift, $y=z/(1+z)$, or the scale\nfactor, $a=1/(1+z)=1-y$, for example. In this paper, we use the scale factor\n$a$ as the variable of expansion. We expand the luminosity distance and the\nHubble parameter around an arbitrary $\\tilde{a}$ and use the Supernovae Ia (SNe\nIa) and the Hubble parameter data to estimate $H$, $q$, $j$ and $s$ at $z\\ne0$\n($\\tilde{a}\\neq1$). We show that the last relevant term for both expansions is\nthe third. Since the third order expansion of $d_L(z)$ has one parameter less\nthan the third order expansion of $H(z)$, we also consider, for completeness, a\nfourth order expansion of $d_L(z)$. For the third order expansions, the results\nobtained from both SNe Ia and $H(z)$ data are incompatible with the\n$\\Lambda$CDM model at $2\\sigma$ confidence level, but also incompatible with\neach other. When the fourth order expansion of $d_L(z)$ is taken into account,\nthe results obtained from SNe Ia data are compatible with the $\\Lambda$CDM\nmodel at $2\\sigma$ confidence level, but still remains incompatible with\nresults obtained from $H(z)$ data. These conflicting results may indicate a\ntension between the current SNe Ia and $H(z)$ data sets.\n",
"title": "Constraints on kinematic parameters at $z\\ne0$"
}
| null | null | null | null | true | null |
14571
| null |
Default
| null | null |
null |
{
"abstract": " Discourse parsing has long been treated as a stand-alone problem independent\nfrom constituency or dependency parsing. Most attempts at this problem are\npipelined rather than end-to-end, sophisticated, and not self-contained: they\nassume gold-standard text segmentations (Elementary Discourse Units), and use\nexternal parsers for syntactic features. In this paper we propose the first\nend-to-end discourse parser that jointly parses in both syntax and discourse\nlevels, as well as the first syntacto-discourse treebank by integrating the\nPenn Treebank with the RST Treebank. Built upon our recent span-based\nconstituency parser, this joint syntacto-discourse parser requires no\npreprocessing whatsoever (such as segmentation or feature extraction), achieves\nthe state-of-the-art end-to-end discourse parsing accuracy.\n",
"title": "Joint Syntacto-Discourse Parsing and the Syntacto-Discourse Treebank"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14572
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we study the combinatorial multi-armed bandit problem (CMAB)\nwith probabilistically triggered arms (PTAs). Under the assumption that the arm\ntriggering probabilities (ATPs) are positive for all arms, we prove that a\nclass of upper confidence bound (UCB) policies, named Combinatorial UCB with\nexploration rate $\\kappa$ (CUCB-$\\kappa$), and Combinatorial Thompson Sampling\n(CTS), which estimates the expected states of the arms via Thompson sampling,\nachieve bounded regret. In addition, we prove that CUCB-$0$ and CTS incur\n$O(\\sqrt{T})$ gap-independent regret. These results improve the results in\nprevious works, which show $O(\\log T)$ gap-dependent and $O(\\sqrt{T\\log T})$\ngap-independent regrets, respectively, under no assumptions on the ATPs. Then,\nwe numerically evaluate the performance of CUCB-$\\kappa$ and CTS in a\nreal-world movie recommendation problem, where the actions correspond to\nrecommending a set of movies, the arms correspond to the edges between the\nmovies and the users, and the goal is to maximize the total number of users\nthat are attracted by at least one movie. Our numerical results complement our\ntheoretical findings on bounded regret. Apart from this problem, our results\nalso directly apply to the online influence maximization (OIM) problem studied\nin numerous prior works.\n",
"title": "Combinatorial Multi-armed Bandit with Probabilistically Triggered Arms: A Case with Bounded Regret"
}
| null | null | null | null | true | null |
14573
| null |
Default
| null | null |
null |
{
"abstract": " Deep neural networks achieve unprecedented performance levels over many tasks\nand scale well with large quantities of data, but performance in the low-data\nregime and tasks like one shot learning still lags behind. While recent work\nsuggests many hypotheses from better optimization to more complicated network\nstructures, in this work we hypothesize that having a learnable and more\nexpressive similarity objective is an essential missing component. Towards\novercoming that, we propose a network design inspired by deep residual networks\nthat allows the efficient computation of this more expressive pairwise\nsimilarity objective. Further, we argue that regularization is key in learning\nwith small amounts of data, and propose an additional generator network based\non the Generative Adversarial Networks where the discriminator is our residual\npairwise network. This provides a strong regularizer by leveraging the\ngenerated data samples. The proposed model can generate plausible variations of\nexemplars over unseen classes and outperforms strong discriminative baselines\nfor few shot classification tasks. Notably, our residual pairwise network\ndesign outperforms previous state-of-theart on the challenging mini-Imagenet\ndataset for one shot learning by getting over 55% accuracy for the 5-way\nclassification task over unseen classes.\n",
"title": "Generative Adversarial Residual Pairwise Networks for One Shot Learning"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14574
| null |
Validated
| null | null |
null |
{
"abstract": " Droplet evaporation in turbulent sprays involves unsteady, multiscale and\nmultiphase processes which make its comprehension and model capabilities still\nlimited. The present work aims to investigate droplet vaporization dynamics\nwithin a turbulent spatial developing jet in dilute, non-reacting conditions.\nWe address the problem using a Direct Numerical Simulation of jet laden with\nacetone droplets using an hybrid Eulerian/Lagrangian approach based on the\npoint droplet approximation. A detailed statistical analysis of both phases is\npresented. In particular, we show how crucial is the preferential sampling of\nthe vapour phase induced by the inhomogeneous localization of the droplets\nthrough the flow. The preferential segregation of droplets develops suddenly\ndownstream the inlet both within the turbulent core and in the mixing layer.\nTwo distinct mechanisms have been found to drive these phenomena, the inertial\nsmall-scale clustering in the jet core and the intermittent dynamics of\ndroplets across the turbulent/non-turbulent interface in the mixing layer where\ndry air entrainment occurs. These phenomenologies strongly affect the overall\nvaporization process and lead to a spectacular widening of droplets size and\nvaporization rate distributions in the downstream evolution of the turbulent\nspray.\n",
"title": "Evaporation of dilute droplets in a turbulent jet: clustering and entrainment effects"
}
| null | null | null | null | true | null |
14575
| null |
Default
| null | null |
null |
{
"abstract": " Let $(\\mathbb{X} , d, \\mu )$ be a proper metric measure space and let $\\Omega\n\\subset \\mathbb{X}$ be a bounded domain. For each $x\\in \\Omega$, we choose a\nradius $0< \\varrho (x) \\leq \\mathrm{dist}(x, \\partial \\Omega ) $ and let $B_x$\nbe the closed ball centered at $x$ with radius $\\varrho (x)$. If $\\alpha \\in\n\\mathbb{R}$, consider the following operator in $C( \\overline{\\Omega} )$, $$\n\\mathcal{T}_{\\alpha}u(x)=\\frac{\\alpha}{2}\\left(\\sup_{B_x } u+\\inf_{B_x }\nu\\right)+(1-\\alpha)\\,\\frac{1}{\\mu(B_x)}\\int_{B_x}\\hspace{-0.1cm} u\\ d\\mu. $$\nUnder appropriate assumptions on $\\alpha$, $\\mathbb{X}$, $\\mu$ and the radius\nfunction $\\varrho$ we show that solutions $u\\in C( \\overline{\\Omega} )$ of the\nfunctional equation $\\mathcal{T}_{\\alpha}u = u$ satisfy a local Hölder or\nLipschitz condition in $\\Omega$. The motivation comes from the so called\n$p$-harmonious functions in euclidean domains.\n",
"title": "A priori Hölder and Lipschitz regularity for generalized $p$-harmonious functions in metric measure spaces"
}
| null | null | null | null | true | null |
14576
| null |
Default
| null | null |
null |
{
"abstract": " We show how simultaneous, back-action evading tracking of non-commuting\nobservables can be achieved in a widely-used sensing technology, atomic\ninterferometry. Using high-dynamic-range dynamically-decoupled quantum\nnon-demolition (QND) measurements on a precessing atomic spin ensemble, we\ntrack the collective spin angle and amplitude with negligible effects from back\naction, giving steady-state tracking sensitivity 2.9 dB beyond the standard\nquantum limit and 7.0 dB beyond Poisson statistics.\n",
"title": "Simultaneous tracking of spin angle and amplitude beyond classical limits"
}
| null | null | null | null | true | null |
14577
| null |
Default
| null | null |
null |
{
"abstract": " Though theoretically expected, the charge exchange emission from galaxy\nclusters has not yet been confidently detected. Accumulating hints were\nreported recently, including a rather marginal detection with the Hitomi data\nof the Perseus cluster. As suggested in Gu et al. (2015), a detection of charge\nexchange line emission from galaxy clusters would not only impact the\ninterpretation of the newly-discovered 3.5 keV line, but also open up a new\nresearch topic on the interaction between hot and cold matter in clusters. We\naim to perform the most systematic search for the O VIII charge exchange line\nin cluster spectra using the RGS on board XMM. We introduce a sample of 21\nclusters observed with the RGS. The dominating thermal plasma emission is\nmodeled and subtracted with a two-temperature CIE component, and the residuals\nare stacked for the line search. The systematic uncertainties in the fits are\nquantified by refitting the spectra with a varying continuum and line\nbroadening. By the residual stacking, we do find a hint of a line-like feature\nat 14.82 A, the characteristic wavelength expected for oxygen charge exchange.\nThis feature has a marginal significance of 2.8 sigma, and the average\nequivalent width is 2.5E-4 keV. We further demonstrate that the putative\nfeature can be hardly affected by the systematic errors from continuum\nmodelling and instrumental effects, or the atomic uncertainties of the\nneighbouring thermal lines. Assuming a realistic temperature and abundance\npattern, the physical model implied by the possible oxygen line agrees well\nwith the theoretical model proposed previously to explain the reported 3.5 keV\nline. If the charge exchange source indeed exists, we would expect that the\noxygen abundance is potentially overestimated by 8-22% in previous X-ray\nmeasurements which assumed pure thermal lines.\n",
"title": "Charge exchange in galaxy clusters"
}
| null | null | null | null | true | null |
14578
| null |
Default
| null | null |
null |
{
"abstract": " We consider Friedlander's wave equation in two space dimensions in the\nhalf-space x > 0 with the boundary condition u(x,y,t)=0 when x=0. For a\nGaussian beam w(x,y,t;k) concentrated on a ray path that is tangent to x=0 at\n(x,y,t)=(0,0,0) we calculate the \"reflected\" wave z(x,y,t;k) in t > 0 such that\nw(x,y,t;k)+z(x,y,t;k) satisfies Friedlander's wave equation and vanishes on\nx=0. These computations are done to leading order in k on the ray path. The\ninteraction of beams with boundaries has been studied for non-tangential beams\nand for beams gliding along the boundary. We find that the amplitude of the\nsolution on the central ray for large k after leaving the boundary is very\nnearly one half of that of the incoming beam.\n",
"title": "A Grazing Gaussian Beam"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14579
| null |
Validated
| null | null |
null |
{
"abstract": " This paper represents a systematic way for generation of Aaria, a simulated\nmodel for serial manipulators for the purpose of kinematic or dynamic analysis\nwith a vast variety of structures based on Simulink SimMechanics. The proposed\nmodel can receive configuration parameters, for instance in accordance with\nmodified Denavit-Hartenberg convention, or trajectories for its base or joints\nfor structures with 1 to 6 degrees of freedom (DOF). The manipulator is\nequipped with artificial joint sensors as well as simulated Inertial\nMeasurement Units (IMUs) on each link. The simulation output can be positions,\nvelocities, torques, in the joint space or IMU outputs; angular velocity,\nlinear acceleration, tool coordinates with respect to the inertial frame. This\nsimulation model is a source of a dataset for virtual multimodal sensory data\nfor automation of robot modeling and control designed for machine learning and\ndeep learning approaches based on big data.\n",
"title": "Reconfigurable Manipulator Simulation for Robotics and Multimodal Machine Learning Application: Aaria"
}
| null | null | null | null | true | null |
14580
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we study the following multi-parameter variant of the\ncelebrated Falconer distance problem. Given ${\\textbf{d}}=(d_1,d_2, \\dots,\nd_{\\ell})\\in \\mathbb{N}^{\\ell}$ with $d_1+d_2+\\dots+d_{\\ell}=d$ and $E\n\\subseteq \\mathbb{R}^d$, we define $$ \\Delta_{\\textbf{d}}(E) = \\left\\{\n\\left(|x^{(1)}-y^{(1)}|,\\ldots,|x^{(\\ell)}-y^{(\\ell)}|\\right) : x,y \\in E\n\\right\\} \\subseteq \\mathbb{R}^{\\ell}, $$ where for $x\\in \\mathbb{R}^d$ we write\n$x=\\left( x^{(1)},\\dots, x^{(\\ell)} \\right)$ with $x^{(i)} \\in\n\\mathbb{R}^{d_i}$.\nWe ask how large does the Hausdorff dimension of $E$ need to be to ensure\nthat the $\\ell$-dimensional Lebesgue measure of $\\Delta_{\\textbf{d}}(E)$ is\npositive? We prove that if $2 \\leq d_i$ for $1 \\leq i \\leq \\ell$, then the\nconclusion holds provided $$ \\dim(E)>d-\\frac{\\min d_i}{2}+\\frac{1}{3}.$$ We\nalso note that, by previous constructions, the conclusion does not in general\nhold if $$\\dim(E)<d-\\frac{\\min d_i}{2}.$$ A group action derivation of a\nsuitable Mattila integral plays an important role in the argument.\n",
"title": "Group actions and a multi-parameter Falconer distance problem"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14581
| null |
Validated
| null | null |
null |
{
"abstract": " Recent years witnessed an extensive development of the theory of the critical\npoint in two-dimensional statistical systems, which allowed to prove {\\it\nexistence} and {\\it conformal invariance} of the {\\it scaling limit} for\ntwo-dimensional Ising model and dimers in planar graphs. Unfortunately, we are\nstill far from a full understanding of the subject: so far, exact solutions at\nthe lattice level, in particular determinant structure and exact discrete\nholomorphicity, play a cucial role in the rigorous control of the scaling\nlimit. The few results about not-integrable (interacting) systems at\ncriticality are still unable to deal with {\\it finite domains} and {\\it\nboundary corrections}, which are of course crucial for getting informations\nabout conformal covariance. In this thesis, we address the question of adapting\nconstructive Renormalization Group methods to non-integrable critical systems\nin $d= 1+1$ dimensions. We study a system of interacting spinless fermions on a\none-dimensional semi-infinite lattice, which can be considered as a prototype\nof the Luttinger universality class with Dirichlet Boundary Conditions. We\ndevelop a convergent renormalized expression for the thermodynamic observables\nin the presence of a quadratic {\\it boundary defect} counterterm, polynomially\nlocalized at the boundary. In particular, we get explicit bounds on the\nboundary corrections to the specific ground state energy.\n",
"title": "Interacting fermions on the half-line: boundary counterterms and boundary corrections"
}
| null | null | null | null | true | null |
14582
| null |
Default
| null | null |
null |
{
"abstract": " In today's era of big data, robust least-squares regression becomes a more\nchallenging problem when considering the adversarial corruption along with\nexplosive growth of datasets. Traditional robust methods can handle the noise\nbut suffer from several challenges when applied in huge dataset including 1)\ncomputational infeasibility of handling an entire dataset at once, 2) existence\nof heterogeneously distributed corruption, and 3) difficulty in corruption\nestimation when data cannot be entirely loaded. This paper proposes online and\ndistributed robust regression approaches, both of which can concurrently\naddress all the above challenges. Specifically, the distributed algorithm\noptimizes the regression coefficients of each data block via heuristic hard\nthresholding and combines all the estimates in a distributed robust\nconsolidation. Furthermore, an online version of the distributed algorithm is\nproposed to incrementally update the existing estimates with new incoming data.\nWe also prove that our algorithms benefit from strong robustness guarantees in\nterms of regression coefficient recovery with a constant upper bound on the\nerror of state-of-the-art batch methods. Extensive experiments on synthetic and\nreal datasets demonstrate that our approaches are superior to those of existing\nmethods in effectiveness, with competitive efficiency.\n",
"title": "Online and Distributed Robust Regressions under Adversarial Data Corruption"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14583
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper we prove a global result for the Schrödinger map problem with\ninitial data with small Besov norm at critical regularity.\n",
"title": "Global well-posedness for the Schrödinger map problem with small Besov norm"
}
| null | null | null | null | true | null |
14584
| null |
Default
| null | null |
null |
{
"abstract": " Implicit schemes have been extensively used in building physics to compute\nthe solution of moisture diffusion problems in porous materials for improving\nstability conditions. Nevertheless, these schemes require important\nsub-iterations when treating non-linear problems. To overcome this\ndisadvantage, this paper explores the use of improved explicit schemes, such as\nDufort-Frankel, Crank-Nicolson and hyperbolisation approaches. A first case\nstudy has been considered with the hypothesis of linear transfer. The\nDufort-Frankel, Crank-Nicolson and hyperbolisation schemes were compared to the\nclassical Euler explicit scheme and to a reference solution. Results have shown\nthat the hyperbolisation scheme has a stability condition higher than the\nstandard Courant-Friedrichs-Lewy (CFL) condition. The error of this schemes\ndepends on the parameter \\tau representing the hyperbolicity magnitude added\ninto the equation. The Dufort-Frankel scheme has the advantages of being\nunconditionally stable and is preferable for non-linear transfer, which is the\nsecond case study. Results have shown the error is proportional to O(\\Delta t).\nA modified Crank-Nicolson scheme has been proposed in order to avoid\nsub-iterations to treat the non-linearities at each time step. The main\nadvantages of the Dufort-Frankel scheme are (i) to be twice faster than the\nCrank-Nicolson approach; (ii) to compute explicitly the solution at each time\nstep; (iii) to be unconditionally stable and (iv) easier to parallelise on\nhigh-performance computer systems. Although the approach is unconditionally\nstable, the choice of the time discretisation $\\Delta t$ remains an important\nissue to accurately represent the physical phenomena.\n",
"title": "Stable explicit schemes for simulation of nonlinear moisture transfer in porous materials"
}
| null | null | null | null | true | null |
14585
| null |
Default
| null | null |
null |
{
"abstract": " We give a complete description of the congruence lattices of the following\nfinite diagram monoids: the partition monoid, the planar partition monoid, the\nBrauer monoid, the Jones monoid (also known as the Temperley-Lieb monoid), the\nMotzkin monoid, and the partial Brauer monoid. All the congruences under\ndiscussion arise as special instances of a new construction, involving an ideal\nI, a retraction I->M onto the minimal ideal, a congruence on M, and a normal\nsubgroup of a maximal subgroup outside I.\n",
"title": "Congruence lattices of finite diagram monoids"
}
| null | null | null | null | true | null |
14586
| null |
Default
| null | null |
null |
{
"abstract": " Let $\\sf X$ be a symplectic orbifold groupoid with $\\sf S$ being a symplectic\nsub-orbifold groupoid, and $\\sf X_{\\mathfrak a}$ be the weight-$\\mathfrak a$\nblowup of $\\sf X$ along $\\sf S$ with $\\sf Z$ being the corresponding\nexceptional divisor. We show that there is a weighted blowup correspondence\nbetween some certain absolute orbifold Gromov--Witten invariants of $\\sf X$\nrelative to $\\sf S$ and some certain relative orbifold Gromov--Witten\ninvariants of the pair $(\\sf X_{\\mathfrak a}|Z)$. As an application, we prove\nthat the symplectic uniruledness of symplectic orbifold groupoids is a weighted\nblowup invariant.\n",
"title": "Weighted blowup correspondence of orbifold Gromov--Witten invariants and applications"
}
| null | null | null | null | true | null |
14587
| null |
Default
| null | null |
null |
{
"abstract": " We use $>$9400 $\\log(m/M_{\\odot})>10$ quiescent and star-forming galaxies at\n$z\\lesssim2$ in COSMOS/UltraVISTA to study the average size evolution of these\nsystems, with focus on the rare, ultra-massive population at\n$\\log(m/M_{\\odot})>11.4$. The large 2-square degree survey area delivers a\nsample of $\\sim400$ such ultra-massive systems. Accurate sizes are derived\nusing a calibration based on high-resolution images from the Hubble Space\nTelescope. We find that, at these very high masses, the size evolution of\nstar-forming and quiescent galaxies is almost indistinguishable in terms of\nnormalization and power-law slope. We use this result to investigate possible\npathways of quenching massive $m>M^*$ galaxies at $z<2$. We consistently model\nthe size evolution of quiescent galaxies from the star-forming population by\nassuming different simple models for the suppression of star-formation. These\nmodels include an instantaneous and delayed quenching without altering the\nstructure of galaxies and a central starburst followed by compaction. We find\nthat instantaneous quenching reproduces well the observed mass-size relation of\nmassive galaxies at $z>1$. Our starburst$+$compaction model followed by\nindividual growth of the galaxies by minor mergers is preferred over other\nmodels without structural change for $\\log(m/M_{\\odot})>11.0$ galaxies at\n$z>0.5$. None of our models is able to meet the observations at $m>M^*$ and\n$z<1$ with out significant contribution of post-quenching growth of individual\ngalaxies via mergers. We conclude that quenching is a fast process in galaxies\nwith $ m \\ge 10^{11} M_\\odot$, and that major mergers likely play a major role\nin the final steps of their evolution.\n",
"title": "Constraints on Quenching of $z\\lesssim2$ Massive Galaxies from the Evolution of the average Sizes of Star-Forming and Quenched Populations in COSMOS"
}
| null | null | null | null | true | null |
14588
| null |
Default
| null | null |
null |
{
"abstract": " The main purpose of this paper is to propose a new preprocessing step in\norder to improve local feature descriptors and texture classification.\nPreprocessing is implemented by using transformations which help highlight\nsalient features that play a significant role in texture recognition. We\nevaluate and compare four different competing methods: three different\nanisotropic diffusion methods including the classical anisotropic Perona-Malik\ndiffusion and two subsequent regularizations of it and the application of a\nGaussian kernel, which is the classical multiscale approach in texture\nanalysis. The combination of the transformed images and the original ones are\nanalyzed. The results show that the use of the preprocessing step does lead to\nimproved texture recognition.\n",
"title": "Improving LBP and its variants using anisotropic diffusion"
}
| null | null | null | null | true | null |
14589
| null |
Default
| null | null |
null |
{
"abstract": " This paper is focused on dimension-free PAC-Bayesian bounds, under weak\npolynomial moment assumptions, allowing for heavy tailed sample distributions.\nIt covers the estimation of the mean of a vector or a matrix, with applications\nto least squares linear regression. Special efforts are devoted to the\nestimation of Gram matrices, due to their prominent role in high-dimension data\nanalysis.\n",
"title": "Dimension-free PAC-Bayesian bounds for matrices, vectors, and linear least squares regression"
}
| null | null | null | null | true | null |
14590
| null |
Default
| null | null |
null |
{
"abstract": " The paper surveys topological problems relevant to the motion planning\nproblem of robotics and includes some new results and constructions. First we\nanalyse the notion of topological complexity of configuration spaces which is\nresponsible for discontinuities in algorithms for robot navigation. Then we\npresent explicit motion planning algorithms for coordinated collision free\ncontrol of many particles moving in Euclidean spaces or on graphs. These\nalgorithms are optimal in the sense that they have minimal number of regions of\ncontinuity. Moreover, we describe in full detail the topology of configuration\nspaces of two particles on a tree and use it to construct some top-dimensional\ncohomology classes in configuration spaces of n particles on a tree.\n",
"title": "Configuration Spaces and Robot Motion Planning Algorithms"
}
| null | null | null | null | true | null |
14591
| null |
Default
| null | null |
null |
{
"abstract": " We propose a novel guess-and-check principle to increase the efficiency of\nthread-modular verification of lock-free data structures. We build on a\nheuristic that guesses candidates for stateless effect summaries of programs by\nsearching the code for instances of a copy-and-check programming idiom common\nin lock-free data structures. These candidate summaries are used to compute the\ninterference among threads in linear time. Since a candidate summary need not\nbe a sound effect summary, we show how to fully automatically check whether the\nprecision of candidate summaries is sufficient. We can thus perform sound\nverification despite relying on an unsound heuristic. We have implemented our\napproach and found it up to two orders of magnitude faster than existing ones.\n",
"title": "Effect Summaries for Thread-Modular Analysis"
}
| null | null | null | null | true | null |
14592
| null |
Default
| null | null |
null |
{
"abstract": " The formation of Correlated Electron Pairs Oscillating around the Fermi level\nin Resonant Quantum States (CEPO-RQS), when a metal is cooled to its critical\ntemperature T=Tc, is studied. The necessary conditions for the existence of\nCEPO-RQS are analyzed. The participation of electron-electron interaction\nscreened by an electron dielectric constant of the form proposed by Thomas\nFermi is considered and a physical meaning for the electron-phonon-electron\ninteraction in the formation of the CEPO-RQS is given. The internal state of\nthe CEPO-RQS is characterized by taking into account the internal Hamiltonian,\nobtaining a general equation that represents its binding energy and depends\nonly on temperature and critical temperature. A parameter is also defined that\ncontains the properties that qualitatively characterizes the nature of a\nmaterial to form the CEPO-RQS.\n",
"title": "Theory of Correlated Pairs of Electrons Oscillating in Resonant Quantum States to Reach the Critical Temperature in a Metal"
}
| null | null | null | null | true | null |
14593
| null |
Default
| null | null |
null |
{
"abstract": " The GIKN construction was introduced by Gorodetski, Ilyashenko, Kleptsyn, and\nNalsky in [Functional Analysis and its Applications, 39 (2005), 21--30]. It\ngives a nonhyperbolic ergodic measure which is a weak$^*$ limit of a special\nsequence of measures supported on periodic orbits. This method was later\nadapted by numerous authors and provided examples of nonhyperbolic invariant\nmeasures in various settings. We prove that the result of the GIKN construction\nis always a loosely Kronecker measure in the sense of Ornstein, Rudolph, and\nWeiss (equivalently, standard measure in the sense of Katok, another name is\nloosely Bernoulli measure with zero entropy). For a proof we introduce and\nstudy the Feldman-Katok pseudometric $\\bar{F_{K}}$. The pseudodistance\n$\\bar{F_{K}}$ is a topological counterpart of the $\\bar f$ metric for\nfinite-state stationary stochastic processes introduced by Feldman and,\nindependently, by Katok, later developed by Ornstein, Rudolph, and Weiss. We\nshow that every measure given by the GIKN construction is the\n$\\bar{F_{K}}$-limit of a sequence of periodic measures. On the other hand we\nprove that a measure which is the $\\bar{F_{K}}$-limit of a sequence of ergodic\nmeasures is ergodic and its entropy is smaller or equal than the lower limit of\nentropies of measures in the sequence. Furthermore we demonstrate that\n$\\bar{F_{K}}$-Cauchy sequence of periodic measures tends in the weak$^*$\ntopology either to a periodic measure or to a loosely Kronecker measure.\n",
"title": "Feldman-Katok pseudometric and the GIKN construction of nonhyperbolic ergodic measures"
}
| null | null | null | null | true | null |
14594
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of sampling from posterior distributions for Bayesian\nmodels where some parameters are restricted to be orthogonal matrices. Such\nmatrices are sometimes used in neural networks models for reasons of\nregularization and stabilization of training procedures, and also can\nparameterize matrices of bounded rank, positive-definite matrices and others.\nIn \\citet{byrne2013geodesic} authors have already considered sampling from\ndistributions over manifolds using exact geodesic flows in a scheme similar to\nHamiltonian Monte Carlo (HMC). We propose new sampling scheme for a set of\northogonal matrices that is based on the same approach, uses ideas of\nRiemannian optimization and does not require exact computation of geodesic\nflows. The method is theoretically justified by proof of symplecticity for the\nproposed iteration. In experiments we show that the new scheme is comparable or\nfaster in time per iteration and more sample-efficient comparing to\nconventional HMC with explicit orthogonal parameterization and Geodesic\nMonte-Carlo. We also provide promising results of Bayesian ensembling for\northogonal neural networks and low-rank matrix factorization.\n",
"title": "Hamiltonian Monte-Carlo for Orthogonal Matrices"
}
| null | null | null | null | true | null |
14595
| null |
Default
| null | null |
null |
{
"abstract": " The Gibbs sampler is a particularly popular Markov chain used for learning\nand inference problems in Graphical Models (GMs). These tasks are\ncomputationally intractable in general, and the Gibbs sampler often suffers\nfrom slow mixing. In this paper, we study the Swendsen-Wang dynamics which is a\nmore sophisticated Markov chain designed to overcome bottlenecks that impede\nthe Gibbs sampler. We prove O(\\log n) mixing time for attractive binary\npairwise GMs (i.e., ferromagnetic Ising models) on stochastic partitioned\ngraphs having n vertices, under some mild conditions, including low temperature\nregions where the Gibbs sampler provably mixes exponentially slow. Our\nexperiments also confirm that the Swendsen-Wang sampler significantly\noutperforms the Gibbs sampler when they are used for learning parameters of\nattractive GMs.\n",
"title": "Rapid Mixing Swendsen-Wang Sampler for Stochastic Partitioned Attractive Models"
}
| null | null | null | null | true | null |
14596
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we argue why and how the integration of recommender systems\nfor research can enhance the functionality and user experience in repositories.\nWe present the latest technical innovations in the CORE Recommender, which\nprovides research article recommendations across the global network of\nrepositories and journals. The CORE Recommender has been recently redeveloped\nand released into production in the CORE system and has also been deployed in\nseveral third-party repositories. We explain the design choices of this unique\nsystem and the evaluation processes we have in place to continue raising the\nquality of the provided recommendations. By drawing on our experience, we\ndiscuss the main challenges in offering a state-of-the-art recommender solution\nfor repositories. We highlight two of the key limitations of the current\nrepository infrastructure with respect to developing research recommender\nsystems: 1) the lack of a standardised protocol and capabilities for exposing\nanonymised user-interaction logs, which represent critically important input\ndata for recommender systems based on collaborative filtering and 2) the lack\nof a voluntary global sign-on capability in repositories, which would enable\nthe creation of personalised recommendation and notification solutions based on\npast user interactions.\n",
"title": "Towards effective research recommender systems for repositories"
}
| null | null | null | null | true | null |
14597
| null |
Default
| null | null |
null |
{
"abstract": " This article sets forth results on the existence, a priori estimates and\nboundedness of positive solutions of a singular quasilinear systems of elliptic\nequations involving variable exponents. The approach is based on Schauder's\nfixed point Theorem. A Moser iteration procedure is also obtained for singular\ncooperative systems involving variable exponents establishing a priori\nestimates and boundedness of solutions.\n",
"title": "Existence and a priori estimates of solutions for quasilinear singular elliptic systems with variable exponents"
}
| null | null | null | null | true | null |
14598
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we characterize planar central configurations in terms of a\nsectional curvature value of the Jacobi-Maupertuis metric. This\ncharacterization works for the $N$-body problem with general masses and any\n$1/r^{\\alpha}$ potential with $\\alpha> 0$. We also observe dynamical\nconsequences of these curvature values for relative equilibrium solutions.\nThese curvature methods work well for strong forces ($\\alpha \\ge 2$).\n",
"title": "On the sectional curvature along central configurations"
}
| null | null | null | null | true | null |
14599
| null |
Default
| null | null |
null |
{
"abstract": " Water fountain stars (WFs) are evolved objects with water masers tracing\nhigh-velocity jets (up to several hundreds of km s$^{-1}$). They could\nrepresent one of the first manifestations of collimated mass-loss in evolved\nobjects and thus, be a key to understanding the shaping mechanisms of planetary\nnebulae. Only 13 objects had been confirmed so far as WFs with interferometer\nobservations. We present new observations with the Australia Telescope Compact\nArray and archival observations with the Very Large Array of four objects that\nare considered to be WF candidates, mainly based on single-dish observations.\nWe confirm IRAS 17291-2147 and IRAS 18596+0315 (OH 37.1-0.8) as bona fide\nmembers of the WF class, with high-velocity water maser emission consistent\nwith tracing bipolar jets. We argue that IRAS 15544-5332 has been wrongly\nconsidered as a WF in previous works, since we see no evidence in our data nor\nin the literature that this object harbours high-velocity water maser emission.\nIn the case of IRAS 19067+0811, we did not detect any water maser emission, so\nits confirmation as a WF is still pending. With the result of this work, there\nare 15 objects that can be considered confirmed WFs. We speculate that there is\nno significant physical difference between WFs and obscured post-AGB stars in\ngeneral. The absence of high-velocity water maser emission in some obscured\npost-AGB stars could be attributed to a variability or orientation effect.\n",
"title": "Interferometric confirmation of \"water fountain\" candidates"
}
| null | null | null | null | true | null |
14600
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.