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": " Network embeddings, which learn low-dimensional representations for each\nvertex in a large-scale network, have received considerable attention in recent\nyears. For a wide range of applications, vertices in a network are typically\naccompanied by rich textual information such as user profiles, paper abstracts,\netc. We propose to incorporate semantic features into network embeddings by\nmatching important words between text sequences for all pairs of vertices. We\nintroduce a word-by-word alignment framework that measures the compatibility of\nembeddings between word pairs, and then adaptively accumulates these alignment\nfeatures with a simple yet effective aggregation function. In experiments, we\nevaluate the proposed framework on three real-world benchmarks for downstream\ntasks, including link prediction and multi-label vertex classification. Results\ndemonstrate that our model outperforms state-of-the-art network embedding\nmethods by a large margin.\n",
"title": "Improved Semantic-Aware Network Embedding with Fine-Grained Word Alignment"
} | null | null | [
"Computer Science"
]
| null | true | null | 18801 | null | Validated | null | null |
null | {
"abstract": " Axion-like particles are promising candidates to make up the dark matter of\nthe universe, but it is challenging to design experiments that can detect them\nover their entire allowed mass range. Dark matter in general, and in particular\naxion-like particles and hidden photons, can be as light as roughly $10^{-22}\n\\;\\rm{eV}$ ($\\sim 10^{-8} \\;\\rm{Hz}$), with astrophysical anomalies providing\nmotivation for the lightest masses (\"fuzzy dark matter\"). We propose\nexperimental techniques for direct detection of axion-like dark matter in the\nmass range from roughly $10^{-13} \\;\\rm{eV}$ ($\\sim 10^2 \\;\\rm{Hz}$) down to\nthe lowest possible masses. In this range, these axion-like particles act as a\ntime-oscillating magnetic field coupling only to spin, inducing effects such as\na time-oscillating torque and periodic variations in the spin-precession\nfrequency with the frequency and direction set by fundamental physics. We show\nhow these signals can be measured using existing experimental technology,\nincluding torsion pendulums, atomic magnetometers, and atom interferometry.\nThese experiments demonstrate a strong discovery capability, with future\niterations of these experiments capable of pushing several orders of magnitude\npast current astrophysical bounds.\n",
"title": "Spin Precession Experiments for Light Axionic Dark Matter"
} | null | null | null | null | true | null | 18802 | null | Default | null | null |
null | {
"abstract": " We obtain restrictions on the persistence barcodes of Laplace-Beltrami\neigenfunctions and their linear combinations on compact surfaces with\nRiemannian metrics. Some applications to uniform approximation by linear\ncombinations of Laplace eigenfunctions are also discussed.\n",
"title": "Persistence barcodes and Laplace eigenfunctions on surfaces"
} | null | null | null | null | true | null | 18803 | null | Default | null | null |
null | {
"abstract": " Carbon materials have a range of properties such as high electrical\nconductivity, high specific surface area, and mechanical flexibility are\nrelevant for electrochemical applications. Carbon materials are utilised in\nenergy conversion-and-storage devices along with electrolytes of complementary\nproperties. In this work, we study the interaction of highly concentrated\nelectrolytes (ionic liquids) at a model carbon surface (circumcoronene) using\ndensity functional theory methods. Our results indicate the decisive role of\nthe dispersion interactions that noticeably strengthen the circumcoronene-ion\ninteraction. Also, we focus on the adsorption of halide anions as the\nelectrolytes containing these ions are promising for practical use in\nsupercapacitors and solar cells.\n",
"title": "DFT study of ionic liquids adsorption on circumcoronene shaped graphene"
} | null | null | null | null | true | null | 18804 | null | Default | null | null |
null | {
"abstract": " In this paper, we use a new approach to prove that the largest eigenvalue of\nthe sample covariance matrix of a normally distributed vector is bigger than\nthe true largest eigenvalue with probability 1 when the dimension is infinite.\nWe prove a similar result for the smallest eigenvalue.\n",
"title": "On the overestimation of the largest eigenvalue of a covariance matrix"
} | null | null | null | null | true | null | 18805 | null | Default | null | null |
null | {
"abstract": " Life is a complex biological phenomenon represented by numerous chemical,\nphysical and biological processes performed by a biothermodynamic\nsystem/cell/organism. Both living organisms and inanimate objects are subject\nto aging, a biological and physicochemical process characterized by changes in\nbiological and thermodynamic state. Thus, the same physical laws govern\nprocesses in both animate and inanimate matter. All life processes lead to\nchange of an organism's state. The change of biological and thermodynamic state\nof an organism in time underlies all of three kinds of aging (chronological,\nbiological and thermodynamic). Life and aging of an organism both start at the\nmoment of fertilization and continue through entire lifespan. Fertilization\nrepresents formation of a new organism. The new organism represents a new\nthermodynamic system. From the very beginning, it changes its state by changing\nthermodynamic parameters. The change of thermodynamic parameters is observed as\naging and can be related to change in entropy. Entropy is thus the parameter\nthat is related to all others and describes aging in the best manner. In the\nbeginning, entropy change appears as a consequence of accumulation of matter\n(growth). Later, decomposition and configurational changes dominate, as a\nconsequence of various chemical reactions (free radical, decomposition,\nfragmentation, accumulation of lipofuscin-like substances...).\n",
"title": "Thermodynamic Mechanism of Life and Aging"
} | null | null | null | null | true | null | 18806 | null | Default | null | null |
null | {
"abstract": " We study truncated point schemes of connected graded algebras as families\nover the parameter space of varying relations for the algebras, proving that\nthe families are flat over the open dense locus where the point schemes achieve\nthe expected (i.e. minimal) dimension.\nWhen the truncated point scheme is zero-dimensional we obtain its number of\npoints counted with multiplicity via a Chow ring computation. This latter\napplication in particular confirms a conjecture of Brazfield to the effect that\na generic two-generator, two-relator 4-dimensional Artin-Schelter regular\nalgebra has seventeen truncated point modules of length six.\n",
"title": "Flat families of point schemes for connected graded algebras"
} | null | null | null | null | true | null | 18807 | null | Default | null | null |
null | {
"abstract": " Hand-crafted features extracted from dynamic contrast-enhanced magnetic\nresonance images (DCE-MRIs) have shown strong predictive abilities in\ncharacterization of breast lesions. However, heterogeneity across medical image\ndatasets hinders the generalizability of these features. One of the sources of\nthe heterogeneity is the variation of MR scanner magnet strength, which has a\nstrong influence on image quality, leading to variations in the extracted image\nfeatures. Thus, statistical decision algorithms need to account for such data\nheterogeneity. Despite the variations, we hypothesize that there exist\nunderlying relationships between the features extracted from the datasets\nacquired with different magnet strength MR scanners. We compared the use of a\nmulti-task learning (MTL) method that incorporates those relationships during\nthe classifier training to support vector machines run on a merged dataset that\nincludes cases with various MRI strength images. As a result, higher predictive\npower is achieved with the MTL method.\n",
"title": "Multi-task Learning in the Computerized Diagnosis of Breast Cancer on DCE-MRIs"
} | null | null | null | null | true | null | 18808 | null | Default | null | null |
null | {
"abstract": " In this paper, we present a new dataset for \"distracted driver\" posture\nestimation. In addition, we propose a novel system that achieves 95.98% driving\nposture estimation classification accuracy. The system consists of a\ngenetically-weighted ensemble of Convolutional Neural Networks (CNNs). We show\nthat a weighted ensemble of classifiers using a genetic algorithm yields in\nbetter classification confidence. We also study the effect of different visual\nelements (i.e. hands and face) in distraction detection and classification by\nmeans of face and hand localizations. Finally, we present a thinned version of\nour ensemble that could achieve a 94.29% classification accuracy and operate in\na realtime environment.\n",
"title": "Real-time Distracted Driver Posture Classification"
} | null | null | null | null | true | null | 18809 | null | Default | null | null |
null | {
"abstract": " The purpose of this document is to create a data model and its serialization\nfor expressing generic time series data. Already existing IVOA data models are\nreused as much as possible. The model is also made as generic as possible to be\nopen to new extensions but at the same time closed for modifications. This\nenables maintaining interoperability throughout different versions of the data\nmodel. We define the necessary building blocks for metadata discovery,\nserialization of time series data and understanding it by clients. We present\nseveral categories of time series science cases with examples of\nimplementation. We also take into account the most pressing topics for time\nseries providers like tracking original images for every individual point of a\nlight curve or time-derived axes like frequency for gravitational wave\nanalysis. The main motivation for the creation of a new model is to provide a\nunified time series data publishing standard - not only for light curves but\nalso more generic time series data, e.g., radial velocity curves, power\nspectra, hardness ratio, provenance linkage, etc. The flexibility is the most\ncrucial part of our model - we are not dependent on any physical domain or\nframe models. While images or spectra are already stable and standardized\nproducts, the time series related domains are still not completely evolved and\nnew ones will likely emerge in near future. That is why we need to keep models\nlike Time Series Cube DM independent of any underlying physical models. In our\nopinion, this is the only correct and sustainable way for future development of\nIVOA standards.\n",
"title": "Time Series Cube Data Model"
} | null | null | null | null | true | null | 18810 | null | Default | null | null |
null | {
"abstract": " We use Markov state models (MSMs) to analyze the dynamics of a\n$\\beta$-hairpin-forming peptide in Monte Carlo (MC) simulations with\ninteracting protein crowders, for two different types of crowder proteins\n[bovine pancreatic trypsin inhibitor (BPTI) and GB1]. In these systems, at the\ntemperature used, the peptide can be folded or unfolded and bound or unbound to\ncrowder molecules. Four or five major free-energy minima can be identified. To\nestimate the dominant MC relaxation times of the peptide, we build MSMs using a\nrange of different time resolutions or lag times. We show that stable\nrelaxation-time estimates can be obtained from the MSM eigenfunctions through\nfits to autocorrelation data. The eigenfunctions remain sufficiently accurate\nto permit stable relaxation-time estimation down to small lag times, at which\npoint simple estimates based on the corresponding eigenvalues have large\nsystematic uncertainties. The presence of the crowders have a stabilizing\neffect on the peptide, especially with BPTI crowders, which can be attributed\nto a reduced unfolding rate $k_\\text{u}$, while the folding rate $k_\\text{f}$\nis left largely unchanged.\n",
"title": "Markov modeling of peptide folding in the presence of protein crowders"
} | null | null | null | null | true | null | 18811 | null | Default | null | null |
null | {
"abstract": " Current induced magnetization manipulation is a key issue for spintronic\napplication. Therefore, deterministic switching of the magnetization at the\npicoseconds timescale with a single electronic pulse represents a major step\ntowards the future developments of ultrafast spintronic. Here, we have studied\nthe ultrafast magnetization dynamics in engineered Gdx[FeCo]1-x based structure\nto compare the effect of femtosecond laser and hot-electrons pulses. We\ndemonstrate that a single femtosecond hot-electrons pulse allows a\ndeterministic magnetization reversal in either Gd-rich and FeCo-rich alloys\nsimilarly to a femtosecond laser pulse. In addition, we show that the limiting\nfactor of such manipulation for perpendicular magnetized films arises from the\nmulti-domain formation due to dipolar interaction. By performing time resolved\nmeasurements under various field, we demonstrate that the same magnetization\ndynamics is observed for both light and hot-electrons excitation and that the\nfull magnetization reversal take place within 5 ps. The energy efficiency of\nthe ultra-fast current induced magnetization manipulation is optimized thanks\nto the ballistic transport of hot-electrons before reaching the GdFeCo magnetic\nlayer.\n",
"title": "Ultra-fast magnetization manipulation using single femtosecond light and hot-electrons pulse"
} | null | null | null | null | true | null | 18812 | null | Default | null | null |
null | {
"abstract": " Orientation effects on the resistivity of copper grain boundaries are studied\nsystematically with two different atomistic tight binding methods. A\nmethodology is developed to model the resistivity of grain boundaries using the\nEmbedded Atom Model, tight binding methods and non-equilibrum Green's functions\n(NEGF). The methodology is validated against first principles calculations for\nsmall, ultra-thin body grain boundaries (<5nm) with 6.4% deviation in the\nresistivity. A statistical ensemble of 600 large, random structures with grains\nis studied. For structures with three grains, it is found that the distribution\nof resistivities is close to normal. Finally, a compact model for grain\nboundary resistivity is constructed based on a neural network.\n",
"title": "Grain Boundary Resistance in Copper Interconnects from an Atomistic Model to a Neural Network"
} | null | null | null | null | true | null | 18813 | null | Default | null | null |
null | {
"abstract": " By applying the classic telescoping summation formula and its variants to\nidentities involving inverse hyperbolic tangent functions having inverse powers\nof the golden ratio as arguments and employing subtle properties of the\nFibonacci and Lucas numbers, we derive interesting general infinite product\nidentities involving these numbers.\n",
"title": "Some remarkable infinite product identities involving Fibonacci and Lucas numbers"
} | null | null | [
"Mathematics"
]
| null | true | null | 18814 | null | Validated | null | null |
null | {
"abstract": " In this paper, we explore the connection between convergence in distribution\nand Mallows distance in the context of positively associated random variables.\nOur results extend some known invariance principles for sequences with FKG\nproperty. Applications for processes with Gibbssian dependence structures are\nincluded.\n",
"title": "Limit Theorems in Mallows Distance for Processes with Gibssian Dependence"
} | null | null | null | null | true | null | 18815 | null | Default | null | null |
null | {
"abstract": " It is a longstanding debate concerning the absence of threshold for the\nsusceptible-infected-susceptible spreading model on networks with localized\nstate. The key to resolve this controversy is the dynamical interaction\npattern, which has not been uncovered. Here we show that the interaction\ndriving the localized-endemic state transition is not the global interaction\nbetween a node and all the other nodes on the network, but exists at the level\nof super node composed of highly connected node and its neighbors. The internal\ninteractions within a super node induce localized state with limited lifetime,\nwhile the interactions between neighboring super nodes via a path of two hops\nenable them to avoid trapping in the absorbing state, marking the onset of\nendemic state. The hybrid interactions render highly connected nodes\nexponentially increasing infection density, which truly account for the null\nthreshold. These results are crucial for correctly understanding diverse\nrecurrent contagion phenomena\n",
"title": "Localized-endemic state transition in the susceptible-infected-susceptible model on networks"
} | null | null | null | null | true | null | 18816 | null | Default | null | null |
null | {
"abstract": " A complete family of solutions for the one-dimensional reaction-diffusion\nequation \\[ u_{xx}(x,t)-q(x)u(x,t) = u_t(x,t) \\] with a coefficient $q$\ndepending on $x$ is constructed. The solutions represent the images of the heat\npolynomials under the action of a transmutation operator. Their use allows one\nto obtain an explicit solution of the noncharacteristic Cauchy problem for the\nconsidered equation with sufficiently regular Cauchy data as well as to solve\nnumerically initial boundary value problems. In the paper the Dirichlet\nboundary conditions are considered however the proposed method can be easily\nextended onto other standard boundary conditions. The proposed numerical method\nis shown to reveal good accuracy.\n",
"title": "Analytic approximation of solutions of parabolic partial differential equations with variable coefficients"
} | null | null | null | null | true | null | 18817 | null | Default | null | null |
null | {
"abstract": " In this paper we develop a numerical method to solve nonlinear optimal\ncontrol problems with final-state constraints. Specifically, we extend the\nPRojection Operator based Netwon's method for Trajectory Optimization (PRONTO),\nwhich was proposed by Hauser for unconstrained optimal control problems. While\nin the standard method final-state constraints can be only approximately\nhandled by means of a terminal penalty, in this work we propose a methodology\nto meet the constraints exactly. Moreover, our method guarantees recursive\nfeasibility of the final-state constraint. This is an appealing property\nespecially in realtime applications in which one would like to be able to stop\nthe computation even if the desired tolerance has not been reached, but still\nsatisfy the constraints. Following the same conceptual idea of PRONTO, the\nproposed strategy is based on two main steps which (differently from the\nstandard scheme) preserve the feasibility of the final-state constraints: (i)\nsolve a quadratic approximation of the nonlinear problem to find a descent\ndirection, and (ii) get a (feasible) trajectory by means of a feedback law\n(which turns out to be a nonlinear projection operator). To find the (feasible)\ndescent direction we take advantage of final-state constrained Linear Quadratic\noptimal control methods, while the second step is performed by suitably\ndesigning a constrained version of the trajectory tracking projection operator.\nThe effectiveness of the proposed strategy is tested on the optimal state\ntransfer of an inverted pendulum.\n",
"title": "Final-State Constrained Optimal Control via a Projection Operator Approach"
} | null | null | null | null | true | null | 18818 | null | Default | null | null |
null | {
"abstract": " In the pursuit of real-time motion planning, a commonly adopted practice is\nto compute a trajectory by running a planning algorithm on a simplified,\nlow-dimensional dynamical model, and then employ a feedback tracking controller\nthat tracks such a trajectory by accounting for the full, high-dimensional\nsystem dynamics. While this strategy of planning with model mismatch generally\nyields fast computation times, there are no guarantees of dynamic feasibility,\nwhich hampers application to safety-critical systems. Building upon recent work\nthat addressed this problem through the lens of Hamilton-Jacobi (HJ)\nreachability, we devise an algorithmic framework whereby one computes, offline,\nfor a pair of \"planner\" (i.e., low-dimensional) and \"tracking\" (i.e.,\nhigh-dimensional) models, a feedback tracking controller and associated\ntracking bound. This bound is then used as a safety margin when generating\nmotion plans via the low-dimensional model. Specifically, we harness the\ncomputational tool of sum-of-squares (SOS) programming to design a bilinear\noptimization algorithm for the computation of the feedback tracking controller\nand associated tracking bound. The algorithm is demonstrated via numerical\nexperiments, with an emphasis on investigating the trade-off between the\nincreased computational scalability afforded by SOS and its intrinsic\nconservativeness. Collectively, our results enable scaling the appealing\nstrategy of planning with model mismatch to systems that are beyond the reach\nof HJ analysis, while maintaining safety guarantees.\n",
"title": "Robust Tracking with Model Mismatch for Fast and Safe Planning: an SOS Optimization Approach"
} | null | null | null | null | true | null | 18819 | null | Default | null | null |
null | {
"abstract": " We consider estimation of the parameters of a Gaussian Stochastic Process\n(GaSP), in the context of emulation (approximation) of computer models for\nwhich the outcomes are real-valued scalars. The main focus is on estimation of\nthe GaSP parameters through various generalized maximum likelihood methods,\nmostly involving finding posterior modes; this is because full Bayesian\nanalysis in computer model emulation is typically prohibitively expensive. The\nposterior modes that are studied arise from objective priors, such as the\nreference prior. These priors have been studied in the literature for the\nsituation of an isotropic covariance function or under the assumption of\nseparability in the design of inputs for model runs used in the GaSP\nconstruction. In this paper, we consider more general designs (e.g., a Latin\nHypercube Design) with a class of commonly used anisotropic correlation\nfunctions, which can be written as a product of isotropic correlation\nfunctions, each having an unknown range parameter and a fixed roughness\nparameter. We discuss properties of the objective priors and marginal\nlikelihoods for the parameters of the GaSP and establish the posterior\npropriety of the GaSP parameters, but our main focus is to demonstrate that\ncertain parameterizations result in more robust estimation of the GaSP\nparameters than others, and that some parameterizations that are in common use\nshould clearly be avoided. These results are applicable to many frequently used\ncovariance functions, e.g., power exponential, Mat{é}rn, rational quadratic\nand spherical covariance. We also generalize the results to the GaSP model with\na nugget parameter. Both theoretical and numerical evidence is presented\nconcerning the performance of the studied procedures.\n",
"title": "Robust Gaussian Stochastic Process Emulation"
} | null | null | null | null | true | null | 18820 | null | Default | null | null |
null | {
"abstract": " Nine transiting Earth-sized planets have recently been discovered around\nnearby late M dwarfs, including the TRAPPIST-1 planets and two planets\ndiscovered by the MEarth survey, GJ 1132b and LHS 1140b. These planets are the\nsmallest known planets that may have atmospheres amenable to detection with\nJWST. We present model thermal emission and transmission spectra for each\nplanet, varying composition and surface pressure of the atmosphere. We base\nelemental compositions on those of Earth, Titan, and Venus and calculate the\nmolecular compositions assuming chemical equilibrium, which can strongly depend\non temperature. Both thermal emission and transmission spectra are sensitive to\nthe atmospheric composition; thermal emission spectra are sensitive to surface\npressure and temperature. We predict the observability of each planet's\natmosphere with JWST. GJ 1132b and TRAPPIST-1b are excellent targets for\nemission spectroscopy with JWST/MIRI, requiring fewer than 10 eclipse\nobservations. Emission photometry for TRAPPIST-1c requires 5-15 eclipses; LHS\n1140b and TRAPPIST-1d, TRAPPIST-1e, and TRAPPIST-1f, which could possibly have\nsurface liquid water, may be accessible with photometry. Seven of the nine\nplanets are strong candidates for transmission spectroscopy measurements with\nJWST, though the number of transits required depends strongly on the planets'\nactual masses. Using the measured masses, fewer than 20 transits are required\nfor a 5 sigma detection of spectral features for GJ 1132b and six of the\nTRAPPIST-1 planets. Dedicated campaigns to measure the atmospheres of these\nnine planets will allow us, for the first time, to probe formation and\nevolution processes of terrestrial planetary atmospheres beyond our solar\nsystem.\n",
"title": "Observing the Atmospheres of Known Temperate Earth-sized Planets with JWST"
} | null | null | null | null | true | null | 18821 | null | Default | null | null |
null | {
"abstract": " We study front propagation phenomena for a large class of nonlocal KPP-type\nreaction-diffusion equations in oscillatory environments, which model various\nforms of population growth with periodic dependence. The nonlocal diffusion is\nan anisotropic integro-differential operator of order $\\alpha \\in (0,2)$.\n",
"title": "Front Propagation for Nonlocal KPP Reaction-Diffusion Equations in Periodic Media"
} | null | null | [
"Mathematics"
]
| null | true | null | 18822 | null | Validated | null | null |
null | {
"abstract": " We study the problem of designing distributed functional observers for LTI\nsystems. Specifically, we consider a setting consisting of a state vector that\nevolves over time according to a dynamical process. A set of nodes distributed\nover a communication network wish to collaboratively estimate certain functions\nof the state. We first show that classical existence conditions for the design\nof centralized functional observers do not directly translate to the\ndistributed setting, due to the coupling that exists between the dynamics of\nthe functions of interest and the diverse measurements at the various nodes.\nAccordingly, we design transformations that reveal such couplings and identify\nportions of the corresponding dynamics that are locally detectable at each\nsensor node. We provide sufficient conditions on the network, along with state\nestimate update and exchange rules for each node, that guarantee asymptotic\nreconstruction of the functions at each sensor node.\n",
"title": "Distributed Functional Observers for LTI Systems"
} | null | null | null | null | true | null | 18823 | null | Default | null | null |
null | {
"abstract": " In this contribution, we extend the methodology proposed in Abry and Didier\n(2017) to obtain the first joint estimator of the real parts of the Hurst\neigenvalues of $n$-variate OFBM. The procedure consists of a wavelet regression\non the log-eigenvalues of the sample wavelet spectrum. The estimator is shown\nto be consistent for any time reversible OFBM and, under stronger assumptions,\nalso asymptotically normal starting from either continuous or discrete time\nmeasurements. Simulation studies establish the finite sample effectiveness of\nthe methodology and illustrate its benefits compared to univariate-like\n(entrywise) analysis. As an application, we revisit the well-known self-similar\ncharacter of Internet traffic by applying the proposed methodology to 4-variate\ntime series of modern, high quality Internet traffic data. The analysis reveals\nthe presence of a rich multivariate self-similarity structure.\n",
"title": "Wavelet eigenvalue regression for $n$-variate operator fractional Brownian motion"
} | null | null | null | null | true | null | 18824 | null | Default | null | null |
null | {
"abstract": " Many researches demonstrated that the DNA methylation, which occurs in the\ncontext of a CpG, has strong correlation with diseases, including cancer. There\nis a strong interest in analyzing the DNA methylation data to find how to\ndistinguish different subtypes of the tumor. However, the conventional\nstatistical methods are not suitable for analyzing the highly dimensional DNA\nmethylation data with bounded support. In order to explicitly capture the\nproperties of the data, we design a deep neural network, which composes of\nseveral stacked binary restricted Boltzmann machines, to learn the low\ndimensional deep features of the DNA methylation data. Experiments show these\nfeatures perform best in breast cancer DNA methylation data cluster analysis,\ncomparing with some state-of-the-art methods.\n",
"title": "Deep Neural Network for Analysis of DNA Methylation Data"
} | null | null | null | null | true | null | 18825 | null | Default | null | null |
null | {
"abstract": " A toy-model of publications and citations processes is proposed. The model\nshows that the role of randomness in the processes is essential and cannot be\nignored. Some other aspects of scientific publications rating are discussed.\n",
"title": "One look at the rating of scientific publications and corresponding toy-model"
} | null | null | null | null | true | null | 18826 | null | Default | null | null |
null | {
"abstract": " Laser-induced adiabatic alignment and mixed-field orientation of\n2,6-difluoroiodobenzene (C6H3F2I) molecules are probed by Coulomb explosion\nimaging following either near-infrared strong-field ionization or\nextreme-ultraviolet multi-photon inner-shell ionization using free-electron\nlaser pulses. The resulting photoelectrons and fragment ions are captured by a\ndouble-sided velocity map imaging spectrometer and projected onto two\nposition-sensitive detectors. The ion side of the spectrometer is equipped with\nthe Pixel Imaging Mass Spectrometry (PImMS) camera, a time-stamping pixelated\ndetector that can record the hit positions and arrival times of up to four ions\nper pixel per acquisition cycle. Thus, the time-of-flight trace and ion\nmomentum distributions for all fragments can be recorded simultaneously. We\nshow that we can obtain a high degree of one- and three-dimensional alignment\nand mixed- field orientation, and compare the Coulomb explosion process induced\nat both wavelengths.\n",
"title": "Alignment, Orientation, and Coulomb Explosion of Difluoroiodobenzene Studied with the Pixel Imaging Mass Spectrometry (PImMS) Camera"
} | null | null | null | null | true | null | 18827 | null | Default | null | null |
null | {
"abstract": " We developed and used a collection of statistical methods (unsupervised\nmachine learning) to extract relevant information from a Twitter supplied data\nset consisting of alleged Russian trolls who (allegedly) attempted to influence\nthe 2016 US Presidential election. These unsupervised statistical methods allow\nfast identification of (i) emergent language communities within the troll\npopulation, (ii) the transnational scope of the operation and (iii) operational\ncharacteristics of trolls that can be used for future identification. Using\nnatural language processing, manifold learning and Fourier analysis, we\nidentify an operation that includes not only the 2016 US election, but also the\nFrench National and both local and national German elections. We show the\nresulting troll population is composed of users with common, but clearly\ncustomized, behavioral characteristics.\n",
"title": "Unsupervised Machine Learning of Open Source Russian Twitter Data Reveals Global Scope and Operational Characteristics"
} | null | null | null | null | true | null | 18828 | null | Default | null | null |
null | {
"abstract": " Distributed algorithms for solving additive or consensus optimization\nproblems commonly rely on first-order or proximal splitting methods. These\nalgorithms generally come with restrictive assumptions and at best enjoy a\nlinear convergence rate. Hence, they can require many iterations or\ncommunications among agents to converge. In many cases, however, we do not seek\na highly accurate solution for consensus problems. Based on this we propose a\ncontrolled relaxation of the coupling in the problem which allows us to compute\nan approximate solution, where the accuracy of the approximation can be\ncontrolled by the level of relaxation. The relaxed problem can be efficiently\nsolved in a distributed way using a combination of primal-dual interior-point\nmethods (PDIPMs) and message-passing. This algorithm purely relies on\nsecond-order methods and thus requires far fewer iterations and communications\nto converge. This is illustrated in numerical experiments, showing its superior\nperformance compared to existing methods.\n",
"title": "Distributed, scalable and gossip-free consensus optimization with application to data analysis"
} | null | null | [
"Mathematics"
]
| null | true | null | 18829 | null | Validated | null | null |
null | {
"abstract": " Network analysis needs tools to infer distributions over graphs of arbitrary\nsize from a single graph. Assuming the distribution is generated by a\ncontinuous latent space model which obeys certain natural symmetry and\nsmoothness properties, we establish three levels of consistency for\nnon-parametric maximum likelihood inference as the number of nodes grows: (i)\nthe estimated locations of all nodes converge in probability on their true\nlocations; (ii) the distribution over locations in the latent space converges\non the true distribution; and (iii) the distribution over graphs of arbitrary\nsize converges.\n",
"title": "Consistency of Maximum Likelihood for Continuous-Space Network Models"
} | null | null | null | null | true | null | 18830 | null | Default | null | null |
null | {
"abstract": " We consider stationary autoregressive processes with coefficients restricted\nto an ellipsoid, which includes autoregressive processes with absolutely\nsummable coefficients. We provide consistency results under different norms for\nthe estimation of such processes using constrained and penalized estimators. As\nan application we show some weak form of universal consistency. Simulations\nshow that directly including the constraint in the estimation can lead to more\nrobust results.\n",
"title": "Consistency Results for Stationary Autoregressive Processes with Constrained Coefficients"
} | null | null | null | null | true | null | 18831 | null | Default | null | null |
null | {
"abstract": " Most current results on coverage control using mobile sensors require that\none partitioned cell is associated with precisely one sensor. In this paper, we\nconsider a class of coverage control problems involving higher order Voronoi\npartitions, motivated by applications where more than one sensor is required to\nmonitor and cover one cell. Such applications are frequent in scenarios\nrequiring the sensors to localize targets. We introduce a framework depending\non a coverage performance function incorporating higher order Voronoi cells and\nthen design a gradient-based controller which allows the multi-sensor system to\nachieve a local equilibrium in a distributed manner. The convergence properties\nare studied and related to Lloyd algorithm. We study also the extension to\ncoverage of a discrete set of points. In addition, we provide a number of real\nworld scenarios where our framework can be applied. Simulation results are also\nprovided to show the controller performance.\n",
"title": "Higher order mobile coverage control with application to localization"
} | null | null | null | null | true | null | 18832 | null | Default | null | null |
null | {
"abstract": " A self-adjoint first order system with Hermitian $\\pi$-periodic potential\n$Q(z)$, integrable on compact sets, is considered. It is shown that all zeros\nof $\\Delta + 2e^{-i\\int_0^\\pi \\Im q dt}$ are double zeros if and only if this\nself-adjoint system is unitarily equivalent to one in which $Q(z)$ is\n$\\frac{\\pi}{2}$-periodic. Furthermore, the zeros of $\\Delta - 2e^{-i\\int_0^\\pi\n\\Im q dt}$ are all double zeros if and only if the associated self-adjoint\nsystem is unitarily equivalent to one in which $Q(z) = \\sigma_2 Q(z) \\sigma_2$.\nHere $\\Delta$ denotes the discriminant of the system and $\\sigma_0$, $\\sigma_2$\nare Pauli matrices. Finally, it is shown that all instability intervals vanish\nif and only if $Q = r\\sigma_0 + q\\sigma_2$, for some real valued $\\pi$-periodic\nfunctions $r$ and $q$ integrable on compact sets.\n",
"title": "Borg's Periodicity Theorems for first order self-adjoint systems with complex potentials"
} | null | null | null | null | true | null | 18833 | null | Default | null | null |
null | {
"abstract": " Informally, \"Information Inconsistency\" is the property that has been\nobserved in many Bayesian hypothesis testing and model selection procedures\nwhereby the Bayesian conclusion does not become definitive when the data seems\nto become definitive. An example is that, when performing a t-test using\nstandard conjugate priors, the Bayes factor of the alternative hypothesis to\nthe null hypothesis remains bounded as the t statistic grows to infinity. This\npaper shows that information inconsistency is ubiquitous in Bayesian hypothesis\ntesting under conjugate priors. Yet the title does not fully describe the\npaper, since we also show that theoretically recommended priors, including\nscale mixtures of conjugate priors and adaptive priors, are information\nconsistent. Hence the paper is simply a forceful warning that use of conjugate\npriors in testing and model selection is highly problematical, and should be\nreplaced by the information consistent alternatives.\n",
"title": "On the Ubiquity of Information Inconsistency for Conjugate Priors"
} | null | null | [
"Mathematics",
"Statistics"
]
| null | true | null | 18834 | null | Validated | null | null |
null | {
"abstract": " Material mixing induced by a Rayleigh-Taylor instability occurs ubiquitously\nin either nature or engineering when a light fluid pushes against a heavy\nfluid, accompanying with the formation and evolution of chaotic bubbles. Its\ngeneral evolution involves two mechanisms: bubble-merge and bubble-competition.\nThe former obeys a universa1 evolution law and has been well-studied, while the\nlatter depends on many factors and has not been well-recognized. In this paper,\nwe establish a theory for the latter to clarify and quantify the longstanding\nopen question: the dependence of bubbles evolution on the dominant factors of\narbitrary density ratio, broadband initial perturbations and various material\nproperties (e.g., viscosity, miscibility, surface tensor). Evolution of the\nmost important characteristic quantities, i.e., the diameter of dominant bubble\n$D$ and the height of bubble zone $h$, is derived: (i) the $D$ expands\nself-similarly with steady aspect ratio $\\beta \\equiv D/h \\thickapprox (1{\\rm{\n+ }}A)/4$, depending only on dimensionless density ratio $A$, and (ii) the $h$\ngrows quadratically with constant growth coefficient $\\alpha \\equiv h/(Ag{t^2})\n\\thickapprox [2\\phi/{\\ln}(2{\\eta _{\\rm{0}}})]^2$, depending on both\ndimensionless initial perturbation amplitude ${\\eta _{\\rm{0}}}$ and\nmaterial-property-associated linear growth rate ratio\n$\\phi\\equiv\\Gamma_{actual}/\\Gamma_{ideal}\\leqslant1$. The theory successfully\nexplains the continued puzzle about the widely varying $\\alpha\\in (0.02,0.12)$\nin experiments and simulations, conducted at all value of $A \\in (0,1)$ and\nwidely varying value of ${\\eta _{\\rm{0}}} \\in [{10^{ - 7}},{10^{ - 2}}]$ with\ndifferent materials. The good agreement between theory and experiments implies\nthat majority of actual mixing depends on initial perturbations and material\nproperties, to which more attention should be paid in either natural or\nengineering problems.\n",
"title": "Competition evolution of Rayleigh-Taylor bubbles"
} | null | null | null | null | true | null | 18835 | null | Default | null | null |
null | {
"abstract": " We construct nonlinear oblique projections along subalgebras of nilpotent Lie\nalgebras in terms of the Baker-Campbell-Hausdorff multiplication. We prove that\nthese nonlinear projections are real analytic on every Schubert cell of the\nGrassmann manifold whose points are the subalgebras of the nilpotent Lie\nalgebra under consideration.\n",
"title": "Nonlinear oblique projections"
} | null | null | null | null | true | null | 18836 | null | Default | null | null |
null | {
"abstract": " It becomes increasingly popular to perform mediation analysis for complex\ndata from sophisticated experimental studies. In this paper, we present Granger\nMediation Analysis (GMA), a new framework for causal mediation analysis of\nmultiple time series. This framework is motivated by a functional magnetic\nresonance imaging (fMRI) experiment where we are interested in estimating the\nmediation effects between a randomized stimulus time series and brain activity\ntime series from two brain regions. The stable unit treatment assumption for\ncausal mediation analysis is thus unrealistic for this type of time series\ndata. To address this challenge, our framework integrates two types of models:\ncausal mediation analysis across the variables and vector autoregressive models\nacross the temporal observations. We further extend this framework to handle\nmultilevel data to address individual variability and correlated errors between\nthe mediator and the outcome variables. These models not only provide valid\ncausal mediation for time series data but also model the causal dynamics across\ntime. We show that the modeling parameters in our models are identifiable, and\nwe develop computationally efficient methods to maximize the likelihood-based\noptimization criteria. Simulation studies show that our method reduces the\nestimation bias and improve statistical power, compared to existing approaches.\nOn a real fMRI data set, our approach not only infers the causal effects of\nbrain pathways but accurately captures the feedback effect of the outcome\nregion on the mediator region.\n",
"title": "Granger Mediation Analysis of Multiple Time Series with an Application to fMRI"
} | null | null | null | null | true | null | 18837 | null | Default | null | null |
null | {
"abstract": " In this article a DNN-based system for detection of three common voice\ndisorders (vocal nodules, polyps and cysts; laryngeal neoplasm; unilateral\nvocal paralysis) is presented. The input to the algorithm is (at least 3-second\nlong) audio recording of sustained vowel sound /a:/. The algorithm was\ndeveloped as part of the \"2018 FEMH Voice Data Challenge\" organized by Far\nEastern Memorial Hospital and obtained score value (defined in the challenge\nspecification) of 77.44. This was the second best result before final\nsubmission. Final challenge results are not yet known during writing of this\ndocument. The document also reports changes that were made for the final\nsubmission which improved the score value in cross-validation by 0.6% points.\n",
"title": "Parameterization of Sequence of MFCCs for DNN-based voice disorder detection"
} | null | null | null | null | true | null | 18838 | null | Default | null | null |
null | {
"abstract": " A key challenge in multi-robot and multi-agent systems is generating\nsolutions that are robust to other self-interested or even adversarial parties\nwho actively try to prevent the agents from achieving their goals. The\npracticality of existing works addressing this challenge is limited to only\nsmall-scale synchronous decision-making scenarios or a single agent planning\nits best response against a single adversary with fixed, procedurally\ncharacterized strategies. In contrast this paper considers a more realistic\nclass of problems where a team of asynchronous agents with limited observation\nand communication capabilities need to compete against multiple strategic\nadversaries with changing strategies. This problem necessitates agents that can\ncoordinate to detect changes in adversary strategies and plan the best response\naccordingly. Our approach first optimizes a set of stratagems that represent\nthese best responses. These optimized stratagems are then integrated into a\nunified policy that can detect and respond when the adversaries change their\nstrategies. The near-optimality of the proposed framework is established\ntheoretically as well as demonstrated empirically in simulation and hardware.\n",
"title": "Near-Optimal Adversarial Policy Switching for Decentralized Asynchronous Multi-Agent Systems"
} | null | null | null | null | true | null | 18839 | null | Default | null | null |
null | {
"abstract": " In this work, we study the extent to which structural connectomes and\ntopological derivative measures are unique to individual changes within human\nbrains. To do so, we classify structural connectome pairs from two large\nlongitudinal datasets as either belonging to the same individual or not. Our\ndata is comprised of 227 individuals from the Alzheimer's Disease Neuroimaging\nInitiative (ADNI) and 226 from the Parkinson's Progression Markers Initiative\n(PPMI). We achieve 0.99 area under the ROC curve score for features which\nrepresent either weights or network structure of the connectomes (node degrees,\nPageRank and local efficiency). Our approach may be useful for eliminating\nnoisy features as a preprocessing step in brain aging studies and early\ndiagnosis classification problems.\n",
"title": "Structural Connectome Validation Using Pairwise Classification"
} | null | null | null | null | true | null | 18840 | null | Default | null | null |
null | {
"abstract": " A Robust Markov Decision Process (RMDP) is a sequential decision making model\nthat accounts for uncertainty in the parameters of dynamic systems. This\nuncertainty introduces difficulties in learning an optimal policy, especially\nfor environments with large state spaces. We propose two algorithms, RTD-DQN\nand Deep-RoK, for solving large-scale RMDPs using nonlinear approximation\nschemes such as deep neural networks. The RTD-DQN algorithm incorporates the\nrobust Bellman temporal difference error into a robust loss function, yielding\nrobust policies for the agent. The Deep-RoK algorithm is a robust Bayesian\nmethod, based on the Extended Kalman Filter (EKF), that accounts for both the\nuncertainty in the weights of the approximated value function and the\nuncertainty in the transition probabilities, improving the robustness of the\nagent. We provide theoretical results for our approach and test the proposed\nalgorithms on a continuous state domain.\n",
"title": "Deep Robust Kalman Filter"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 18841 | null | Validated | null | null |
null | {
"abstract": " The Fisher information matrix (FIM) is a fundamental quantity to represent\nthe characteristics of a stochastic model, including deep neural networks\n(DNNs). The present study reveals novel statistics of FIM that are universal\namong a wide class of DNNs. To this end, we use random weights and large width\nlimits, which enables us to utilize mean field theories. We investigate the\nasymptotic statistics of the FIM's eigenvalues and reveal that most of them are\nclose to zero while the maximum takes a huge value. This implies that the\neigenvalue distribution has a long tail. Because the landscape of the parameter\nspace is defined by the FIM, it is locally flat in most dimensions, but\nstrongly distorted in others. We also demonstrate the potential usage of the\nderived statistics through two exercises. First, small eigenvalues that induce\nflatness can be connected to a norm-based capacity measure of generalization\nability. Second, the maximum eigenvalue that induces the distortion enables us\nto quantitatively estimate an appropriately sized learning rate for gradient\nmethods to converge.\n",
"title": "Universal Statistics of Fisher Information in Deep Neural Networks: Mean Field Approach"
} | null | null | [
"Statistics"
]
| null | true | null | 18842 | null | Validated | null | null |
null | {
"abstract": " In this paper we study the extension problem for the sublaplacian on a\n$H$-type group and use the solutions to prove trace Hardy and Hardy\ninequalities for fractional powers of the sublaplacian.\n",
"title": "An extension problem and trace Hardy inequality for the sublaplacian on $H$-type groups"
} | null | null | null | null | true | null | 18843 | null | Default | null | null |
null | {
"abstract": " Univariate isotonic regression (IR) has been used for nonparametric\nestimation in dose-response and dose-finding studies. One undesirable property\nof IR is the prevalence of piecewise-constant stretches in its estimates,\nwhereas the dose-response function is usually assumed to be strictly\nincreasing. We propose a simple modification to IR, called centered isotonic\nregression (CIR). CIR's estimates are strictly increasing in the interior of\nthe dose range. In the absence of monotonicity violations, CIR and IR both\nreturn the original observations. Numerical examination indicates that for\nsample sizes typical of dose-response studies and with realistic dose-response\ncurves, CIR provides a substantial reduction in estimation error compared with\nIR when monotonicity violations occur. We also develop analytical interval\nestimates for IR and CIR, with good coverage behavior. An R package implements\nthese point and interval estimates.\n",
"title": "Centered Isotonic Regression: Point and Interval Estimation for Dose-Response Studies"
} | null | null | [
"Statistics"
]
| null | true | null | 18844 | null | Validated | null | null |
null | {
"abstract": " Measuring domain relevance of data and identifying or selecting well-fit\ndomain data for machine translation (MT) is a well-studied topic, but denoising\nis not yet. Denoising is concerned with a different type of data quality and\ntries to reduce the negative impact of data noise on MT training, in\nparticular, neural MT (NMT) training. This paper generalizes methods for\nmeasuring and selecting data for domain MT and applies them to denoising NMT\ntraining. The proposed approach uses trusted data and a denoising curriculum\nrealized by online data selection. Intrinsic and extrinsic evaluations of the\napproach show its significant effectiveness for NMT to train on data with\nsevere noise.\n",
"title": "Denoising Neural Machine Translation Training with Trusted Data and Online Data Selection"
} | null | null | null | null | true | null | 18845 | null | Default | null | null |
null | {
"abstract": " Cyber attacks are growing in frequency and severity. Over the past year alone\nwe have witnessed massive data breaches that stole personal information of\nmillions of people and wide-scale ransomware attacks that paralyzed critical\ninfrastructure of several countries. Combating the rising cyber threat calls\nfor a multi-pronged strategy, which includes predicting when these attacks will\noccur. The intuition driving our approach is this: during the planning and\npreparation stages, hackers leave digital traces of their activities on both\nthe surface web and dark web in the form of discussions on platforms like\nhacker forums, social media, blogs and the like. These data provide predictive\nsignals that allow anticipating cyber attacks. In this paper, we describe\nmachine learning techniques based on deep neural networks and autoregressive\ntime series models that leverage external signals from publicly available Web\nsources to forecast cyber attacks. Performance of our framework across ground\ntruth data over real-world forecasting tasks shows that our methods yield a\nsignificant lift or increase of F1 for the top signals on predicted cyber\nattacks. Our results suggest that, when deployed, our system will be able to\nprovide an effective line of defense against various types of targeted cyber\nattacks.\n",
"title": "Discovering Signals from Web Sources to Predict Cyber Attacks"
} | null | null | null | null | true | null | 18846 | null | Default | null | null |
null | {
"abstract": " In this paper and its sequels, we give an unified treatment of the\nhigher-degree smoothness of admissible perturbations and related results used\nin the global perturbation method for GW and Floer theories.\n",
"title": "Higher-degree Smoothness of Perturbations I"
} | null | null | null | null | true | null | 18847 | null | Default | null | null |
null | {
"abstract": " We present the results of an optical spectroscopic monitoring program\ntargeting NGC 5548 as part of a larger multi-wavelength reverberation mapping\ncampaign. The campaign spanned six months and achieved an almost daily cadence\nwith observations from five ground-based telescopes. The H$\\beta$ and He II\n$\\lambda$4686 broad emission-line light curves lag that of the 5100 $\\AA$\noptical continuum by $4.17^{+0.36}_{-0.36}$ days and $0.79^{+0.35}_{-0.34}$\ndays, respectively. The H$\\beta$ lag relative to the 1158 $\\AA$ ultraviolet\ncontinuum light curve measured by the Hubble Space Telescope is roughly\n$\\sim$50% longer than that measured against the optical continuum, and the lag\ndifference is consistent with the observed lag between the optical and\nultraviolet continua. This suggests that the characteristic radius of the\nbroad-line region is $\\sim$50% larger than the value inferred from optical data\nalone. We also measured velocity-resolved emission-line lags for H$\\beta$ and\nfound a complex velocity-lag structure with shorter lags in the line wings,\nindicative of a broad-line region dominated by Keplerian motion. The responses\nof both the H$\\beta$ and He II $\\lambda$4686 emission lines to the driving\ncontinuum changed significantly halfway through the campaign, a phenomenon also\nobserved for C IV, Ly $\\alpha$, He II(+O III]), and Si IV(+O IV]) during the\nsame monitoring period. Finally, given the optical luminosity of NGC 5548\nduring our campaign, the measured H$\\beta$ lag is a factor of five shorter than\nthe expected value implied by the $R_\\mathrm{BLR} - L_\\mathrm{AGN}$ relation\nbased on the past behavior of NGC 5548.\n",
"title": "Space Telescope and Optical Reverberation Mapping Project. V. Optical Spectroscopic Campaign and Emission-Line Analysis for NGC 5548"
} | null | null | null | null | true | null | 18848 | null | Default | null | null |
null | {
"abstract": " An algorithm for solving smooth nonconvex optimization problems is proposed\nthat, in the worst-case, takes $\\mathcal{O}(\\epsilon^{-3/2})$ iterations to\ndrive the norm of the gradient of the objective function below a prescribed\npositive real number $\\epsilon$ and can take $\\mathcal{O}(\\epsilon^{-3})$\niterations to drive the leftmost eigenvalue of the Hessian of the objective\nabove $-\\epsilon$. The proposed algorithm is a general framework that covers a\nwide range of techniques including quadratically and cubically regularized\nNewton methods, such as the Adaptive Regularisation using Cubics (ARC) method\nand the recently proposed Trust-Region Algorithm with Contractions and\nExpansions (TRACE). The generality of our method is achieved through the\nintroduction of generic conditions that each trial step is required to satisfy,\nwhich in particular allow for inexact regularized Newton steps to be used.\nThese conditions center around a new subproblem that can be approximately\nsolved to obtain trial steps that satisfy the conditions. A new instance of the\nframework, distinct from ARC and TRACE, is described that may be viewed as a\nhybrid between quadratically and cubically regularized Newton methods.\nNumerical results demonstrate that our hybrid algorithm outperforms a cublicly\nregularized Newton method.\n",
"title": "An Inexact Regularized Newton Framework with a Worst-Case Iteration Complexity of $\\mathcal{O}(ε^{-3/2})$ for Nonconvex Optimization"
} | null | null | null | null | true | null | 18849 | null | Default | null | null |
null | {
"abstract": " We consider an inverse boundary value problem for Maxwell's equations, which\naims to recover the electromagnetic material properties of a body from\nmeasurements on the boundary. We show that a Lipschitz continuous conductivity,\nelectric permittivity, and magnetic permeability are uniquely determined by\nknowledge of all tangential electric and magnetic fields on the boundary of the\nbody at a fixed frequency.\n",
"title": "An inverse problem for Maxwell's equations with Lipschitz parameters"
} | null | null | null | null | true | null | 18850 | null | Default | null | null |
null | {
"abstract": " We discuss the process of building semantic maps, how to interactively label\nentities in them, and how to use them to enable context-aware navigation\nbehaviors in human environments. We utilize planar surfaces, such as walls and\ntables, and static objects, such as door signs, as features for our semantic\nmapping approach. Users can interactively annotate these features by having the\nrobot follow him/her, entering the label through a mobile app, and performing a\npointing gesture toward the landmark of interest. Our gesture based approach\ncan reliably estimate which object is being pointed at and detect ambiguous\ngestures with probabilistic modeling. Our person following method attempts to\nmaximize future utility by a search for future actions assuming constant\nvelocity model for the human. We describe a method to extract metric goals from\na semantic map landmark and to plan a human aware path that takes into account\nthe personal spaces of people. Finally, we demonstrate context-awareness for\nperson following in two scenarios: interactive labeling and door passing. We\nbelieve that future navigation approaches and service robotics applications can\nbe made more effective by further exploiting the structure of human\nenvironments.\n",
"title": "Context Aware Robot Navigation using Interactively Built Semantic Maps"
} | null | null | null | null | true | null | 18851 | null | Default | null | null |
null | {
"abstract": " Multimodal sensory data resembles the form of information perceived by humans\nfor learning, and are easy to obtain in large quantities. Compared to unimodal\ndata, synchronization of concepts between modalities in such data provides\nsupervision for disentangling the underlying explanatory factors of each\nmodality. Previous work leveraging multimodal data has mainly focused on\nretaining only the modality-invariant factors while discarding the rest. In\nthis paper, we present a partitioned variational autoencoder (PVAE) and several\ntraining objectives to learn disentangled representations, which encode not\nonly the shared factors, but also modality-dependent ones, into separate latent\nvariables. Specifically, PVAE integrates a variational inference framework and\na multimodal generative model that partitions the explanatory factors and\nconditions only on the relevant subset of them for generation. We evaluate our\nmodel on two parallel speech/image datasets, and demonstrate its ability to\nlearn disentangled representations by qualitatively exploring within-modality\nand cross-modality conditional generation with semantics and styles specified\nby examples. For quantitative analysis, we evaluate the classification accuracy\nof automatically discovered semantic units. Our PVAE can achieve over 99%\naccuracy on both modalities.\n",
"title": "Disentangling by Partitioning: A Representation Learning Framework for Multimodal Sensory Data"
} | null | null | null | null | true | null | 18852 | null | Default | null | null |
null | {
"abstract": " Blind gain and phase calibration (BGPC) is a bilinear inverse problem\ninvolving the determination of unknown gains and phases of the sensing system,\nand the unknown signal, jointly. BGPC arises in numerous applications, e.g.,\nblind albedo estimation in inverse rendering, synthetic aperture radar\nautofocus, and sensor array auto-calibration. In some cases, sparse structure\nin the unknown signal alleviates the ill-posedness of BGPC. Recently there has\nbeen renewed interest in solutions to BGPC with careful analysis of error\nbounds. In this paper, we formulate BGPC as an eigenvalue/eigenvector problem,\nand propose to solve it via power iteration, or in the sparsity or joint\nsparsity case, via truncated power iteration. Under certain assumptions, the\nunknown gains, phases, and the unknown signal can be recovered simultaneously.\nNumerical experiments show that power iteration algorithms work not only in the\nregime predicted by our main results, but also in regimes where theoretical\nanalysis is limited. We also show that our power iteration algorithms for BGPC\ncompare favorably with competing algorithms in adversarial conditions, e.g.,\nwith noisy measurement or with a bad initial estimate.\n",
"title": "Blind Gain and Phase Calibration via Sparse Spectral Methods"
} | null | null | null | null | true | null | 18853 | null | Default | null | null |
null | {
"abstract": " Empirical observations show that ecological communities can have a huge\nnumber of coexisting species, also with few or limited number of resources.\nThese ecosystems are characterized by multiple type of interactions, in\nparticular displaying cooperative behaviors. However, standard modeling of\npopulation dynamics based on Lotka-Volterra type of equations predicts that\necosystem stability should decrease as the number of species in the community\nincreases and that cooperative systems are less stable than communities with\nonly competitive and/or exploitative interactions. Here we propose a stochastic\nmodel of population dynamics, which includes exploitative interactions as well\nas cooperative interactions induced by cross-feeding. The model is exactly\nsolved and we obtain results for relevant macro-ecological patterns, such as\nspecies abundance distributions and correlation functions. In the large system\nsize limit, any number of species can coexist for a very general class of\ninteraction networks and stability increases as the number of species grows.\nFor pure mutualistic/commensalistic interactions we determine the topological\nproperties of the network that guarantee species coexistence. We also show that\nthe stationary state is globally stable and that inferring species interactions\nthrough species abundance correlation analysis may be misleading. Our\ntheoretical approach thus show that appropriate models of cooperation naturally\nleads to a solution of the long-standing question about complexity-stability\nparadox and on how highly biodiverse communities can coexist.\n",
"title": "Reconciling cooperation, biodiversity and stability in complex ecological communities"
} | null | null | [
"Quantitative Biology"
]
| null | true | null | 18854 | null | Validated | null | null |
null | {
"abstract": " We have discovered that the extremely red, low-gravity L7 dwarf 2MASS\nJ11193254-1137466 is a 0.14\" (3.6 AU) binary using Keck laser guide star\nadaptive optics imaging. 2MASS J11193254-1137466 has previously been identified\nas a likely member of the TW Hydrae Association (TWA). Using our updated\nphotometric distance and proper motion, a kinematic analysis based on the\nBANYAN II model gives an 82% probability of TWA membership. At TWA's 10$\\pm$3\nMyr age and using hot-start evolutionary models, 2MASS J11193254-1137466AB is a\npair of $3.7^{+1.2}_{-0.9}$ $M_{\\rm Jup}$ brown dwarfs, making it the\nlowest-mass binary discovered to date. We estimate an orbital period of\n$90^{+80}_{-50}$ years. One component is marginally brighter in $K$ band but\nfainter in $J$ band, making this a probable flux-reversal binary, the first\ndiscovered with such a young age. We also imaged the spectrally similar TWA L7\ndwarf WISEA J114724.10-204021.3 with Keck and found no sign of binarity. Our\nevolutionary model-derived $T_{\\rm eff}$ estimate for WISEA J114724.10-204021.3\nis $\\approx$230 K higher than for 2MASS J11193254-1137466AB, at odds with their\nspectral similarity. This discrepancy suggests that WISEA J114724.10-204021.3\nmay actually be a tight binary with masses and temperatures very similar to\n2MASS J11193254-1137466AB, or further supporting the idea that near-infrared\nspectra of young ultracool dwarfs are shaped by factors other than temperature\nand gravity. 2MASS J11193254-1137466AB will be an essential benchmark for\ntesting evolutionary and atmospheric models in the young planetary-mass regime.\n",
"title": "The Young L Dwarf 2MASS J11193254-1137466 is a Planetary-Mass Binary"
} | null | null | null | null | true | null | 18855 | null | Default | null | null |
null | {
"abstract": " We determine the abundances of neutron-capture elements from Sr to Eu for\nfive very-metal-poor stars (-3<[Fe/H]<-2) in the Milky Way halo to reveal the\norigin of light neutron-capture elements. Previous spectroscopic studies have\nshown evidence of at least two components in the r-process; one referred to as\nthe \"main r-process\" and the other as the \"weak r-process,\" which is mainly\nresponsible for producing heavy and light neutron-capture elements,\nrespectively. Observational studies of metal-poor stars suggest that there is a\nuniversal pattern in the main r-process, similar to the abundance pattern of\nthe r-process component of solar-system material. Still, it is uncertain\nwhether the abundance pattern of the weak r-process shows universality or\ndiversity, due to the sparseness of measured light neutron-capture elements. We\nhave detected the key elements, Mo, Ru, and Pd, in five target stars to give an\nanswer to this question. The abundance patterns of light neutron-capture\nelements from Sr to Pd suggest a diversity in the weak r-process. In\nparticular, scatter in the abundance ratio between Ru and Pd is significant\nwhen the abundance patterns are normalized at Zr. Our results are compared with\nthe elemental abundances predicted by nucleosynthesis models of supernovae with\nparameters such as electron fraction or proto-neutron-star mass, to investigate\nsources of such diversity in the abundance patterns of light neutron-capture\nelements. This paper presents that the variation in the abundances of observed\nstars can be explained with a small range of parameters, which can serve as\nconstraints on future modeling of supernova models.\n",
"title": "Diversity of Abundance Patterns of Light Neutron-capture Elements in Very-metal-poor Stars"
} | null | null | [
"Physics"
]
| null | true | null | 18856 | null | Validated | null | null |
null | {
"abstract": " The identification of the Stuxnet worm in 2010 provided a highly publicized\nexample of a cyber attack used to damage an industrial control system\nphysically. This raised public awareness about the possibility of similar\nattacks against other industrial targets -- including critical infrastructure.\nIn this paper, we use hypergames to analyze how adversarial perturbations can\nbe used to manipulate a system using optimal control. Hypergames form an\nextension of game theory that enables us to model strategic interactions where\nthe players may have significantly different perceptions of the game(s) they\nare playing. Past work with hypergames has been limited to relatively simple\ninteractions consisting of a small set of discrete choices for each player, but\nhere, we apply hypergames to larger systems with continuous variables. We find\nthat manipulating constraints can be a more effective attacker strategy than\ndirectly manipulating objective function parameters. Moreover, the attacker\nneed not change the underlying system to carry out a successful attack -- it\nmay be sufficient to deceive the defender controlling the system. It is\npossible to scale our approach up to even larger systems, but the ability to do\nso will depend on the characteristics of the system in question, and we\nidentify several characteristics that will make those systems amenable to\nhypergame analysis.\n",
"title": "Hypergames and Cyber-Physical Security for Control Systems"
} | null | null | null | null | true | null | 18857 | null | Default | null | null |
null | {
"abstract": " The objective assessment of the prestige of an academic institution is a\ndifficult and hotly debated task. In the last few years, different types of\nUniversity Rankings have been proposed to quantify the excellence of different\nresearch institutions in the world. Albeit met with criticism in some cases,\nthe relevance of university rankings is being increasingly acknowledged:\nindeed, rankings are having a major impact on the design of research policies,\nboth at the institutional and governmental level. Yet, the debate on what\nrankings are {\\em exactly} measuring is enduring. Here, we address the issue by\nmeasuring a quantitive and reliable proxy of the academic reputation of a given\ninstitution and by evaluating its correlation with different university\nrankings. Specifically, we study citation patterns among universities in five\ndifferent Web of Science Subject Categories and use the \\pr~algorithm on the\nfive resulting citation networks. The rationale behind our work is that\nscientific citations are driven by the reputation of the reference so that the\nPageRank algorithm is expected to yield a rank which reflects the reputation of\nan academic institution in a specific field. Our results allow to quantifying\nthe prestige of a set of institutions in a certain research field based only on\nhard bibliometric data. Given the volume of the data analysed, our findings are\nstatistically robust and less prone to bias, at odds with ad--hoc surveys often\nemployed by ranking bodies in order to attain similar results. Because our\nfindings are found to correlate extremely well with the ARWU Subject rankings,\nthe approach we propose in our paper may open the door to new, Academic Ranking\nmethodologies that go beyond current methods by reconciling the qualitative\nevaluation of Academic Prestige with its quantitative measurements via\npublication impact.\n",
"title": "Measuring the academic reputation through citation networks via PageRank"
} | null | null | null | null | true | null | 18858 | null | Default | null | null |
null | {
"abstract": " The ab initio description of the spectral interior of the absorption spectrum\nposes both a theoretical and computational challenge for modern electronic\nstructure theory. Due to the often spectrally dense character of this domain in\nthe quantum propagator's eigenspectrum for medium-to-large sized systems,\ntraditional approaches based on the partial diagonalization of the propagator\noften encounter oscillatory and stagnating convergence. Electronic structure\nmethods which solve the molecular response problem through the solution of\nspectrally shifted linear systems, such as the complex polarization propagator,\noffer an alternative approach which is agnostic to the underlying spectral\ndensity or domain location. This generality comes at a seemingly high\ncomputational cost associated with solving a large linear system for each\nspectral shift in some discretization of the spectral domain of interest. We\npresent a novel, adaptive solution based on model order reduction techniques\nvia interpolation. Model order reduction reduces the computational complexity\nof mathematical models and is ubiquitous in the simulation of dynamical\nsystems. The efficiency and effectiveness of the proposed algorithm in the ab\ninitio prediction of X-Ray absorption spectra is demonstrated using a test set\nof challenging water clusters which are spectrally dense in the neighborhood of\nthe oxygen K-edge. Based on a single, user defined tolerance we automatically\ndetermine the order of the reduced models and approximate the absorption\nspectrum up to the given tolerance. We also illustrate that the automatically\ndetermined model order increases logarithmically with the problem dimension,\ncompared to a linear increase of the number of eigenvalues within the energy\nwindow. Furthermore, we observed that the computational cost of the proposed\nalgorithm only scales quadratically with respect to the problem dimension.\n",
"title": "A Model Order Reduction Algorithm for Estimating the Absorption Spectrum"
} | null | null | [
"Computer Science",
"Physics"
]
| null | true | null | 18859 | null | Validated | null | null |
null | {
"abstract": " We analyze the rank gradient of finitely generated groups with respect to\nsequences of subgroups of finite index that do not necessarily form a chain, by\nconnecting it to the cost of p.m.p. actions. We generalize several results that\nwere only known for chains before. The connection is made by the notion of\nlocal-global convergence.\nIn particular, we show that for a finitely generated group $\\Gamma$ with\nfixed price $c$, every Farber sequence has rank gradient $c-1$. By adapting\nLackenby's trichotomy theorem to this setting, we also show that in a finitely\npresented amenable group, every sequence of subgroups with index tending to\ninfinity has vanishing rank gradient.\n",
"title": "Uniform rank gradient, cost and local-global convergence"
} | null | null | null | null | true | null | 18860 | null | Default | null | null |
null | {
"abstract": " Fuel cells, batteries, thermochemical and other energy conversion devices\ninvolve the transport of a number of (electro-)chemical species through\ndistinct materials so that they can meet and react at specified multi-material\ninterfaces. Therefore, morphology or arrangement of these different materials\ncan be critical in the performance of an energy conversion device. In this\npaper, we study a model problem motivated by a solar-driven thermochemical\nconversion device that splits water into hydrogen and oxygen. We formulate the\nproblem as a system of coupled multi-material reaction-diffusion equations\nwhere each species diffuses selectively through a given material and where the\nreaction occurs at multi-material interfaces. We express the problem of optimal\ndesign of the material arrangement as a saddle point problem and obtain an\neffective functional which shows that regions with very fine phase mixtures of\nthe material arise naturally. To explore this further, we introduce a\nphase-field formulation of the optimal design problem, and numerically study\nselected examples.\n",
"title": "Optimal design of a model energy conversion device"
} | null | null | null | null | true | null | 18861 | null | Default | null | null |
null | {
"abstract": " Comments play an important role within online creative communities because\nthey make it possible to foster the production and improvement of authors'\nartifacts. We investigate how comment-based communication help shape members'\nbehavior within online creative communities. In this paper, we report the\nresults of a preliminary study aimed at mining the communication network of a\nmusic community for collaborative songwriting, where users collaborate online\nby first uploading new songs and then by adding new tracks and providing\nfeedback in forms of comments.\n",
"title": "Mining Communication Data in a Music Community: A Preliminary Analysis"
} | null | null | [
"Computer Science"
]
| null | true | null | 18862 | null | Validated | null | null |
null | {
"abstract": " We develop new numerical schemes for Vlasov--Poisson equations with\nhigh-order accuracy. Our methods are based on a spatially\nmonotonicity-preserving (MP) scheme and are modified suitably so that\npositivity of the distribution function is also preserved. We adopt an\nefficient semi-Lagrangian time integration scheme that is more accurate and\ncomputationally less expensive than the three-stage TVD Runge-Kutta\nintegration. We apply our spatially fifth- and seventh-order schemes to a suite\nof simulations of collisionless self-gravitating systems and electrostatic\nplasma simulations, including linear and nonlinear Landau damping in one\ndimension and Vlasov--Poisson simulations in a six-dimensional phase space. The\nhigh-order schemes achieve a significantly improved accuracy in comparison with\nthe third-order positive-flux-conserved scheme adopted in our previous study.\nWith the semi-Lagrangian time integration, the computational cost of our\nhigh-order schemes does not significantly increase, but remains roughly the\nsame as that of the third-order scheme. Vlasov--Poisson simulations on $128^3\n\\times 128^3$ mesh grids have been successfully performed on a massively\nparallel computer.\n",
"title": "Multidimensional VlasovPoisson Simulations with High-order Monotonicity- and Positivity-preserving Schemes"
} | null | null | null | null | true | null | 18863 | null | Default | null | null |
null | {
"abstract": " Markov Chain Monte Carlo based Bayesian data analysis has now become the\nmethod of choice for analyzing and interpreting data in almost all disciplines\nof science. In astronomy, over the last decade, we have also seen a steady\nincrease in the number of papers that employ Monte Carlo based Bayesian\nanalysis. New, efficient Monte Carlo based methods are continuously being\ndeveloped and explored. In this review, we first explain the basics of Bayesian\ntheory and discuss how to set up data analysis problems within this framework.\nNext, we provide an overview of various Monte Carlo based methods for\nperforming Bayesian data analysis. Finally, we discuss advanced ideas that\nenable us to tackle complex problems and thus hold great promise for the\nfuture. We also distribute downloadable computer software (available at\nthis https URL ) that implements some of the algorithms and\nexamples discussed here.\n",
"title": "Markov Chain Monte Carlo Methods for Bayesian Data Analysis in Astronomy"
} | null | null | null | null | true | null | 18864 | null | Default | null | null |
null | {
"abstract": " We use nonabelian Poincaré duality to recover the stable splitting of\ncompactly supported mapping spaces, $\\rm{Map_c}$$(M,\\Sigma^nX)$, where $M$ is a\nparallelizable $n$-manifold. Our method for deriving this splitting is new, and\nnaturally extends to give a more general stable splitting of the space of\ncompactly supported sections of a certain bundle on $M$ with fibers\n$\\Sigma^nX$, twisted by the tangent bundle of $M$. This generalization\nincorporates possible $O(n)$-actions on $X$ as well as accommodating\nnon-parallelizable manifolds.\n",
"title": "Stable splitting of mapping spaces via nonabelian Poincaré duality"
} | null | null | null | null | true | null | 18865 | null | Default | null | null |
null | {
"abstract": " In this report, an automated bartender system was developed for making orders\nin a bar using hand gestures. The gesture recognition of the system was\ndeveloped using Machine Learning techniques, where the model was trained to\nclassify gestures using collected data. The final model used in the system\nreached an average accuracy of 95%. The system raised ethical concerns both in\nterms of user interaction and having such a system in a real world scenario,\nbut it could initially work as a complement to a real bartender.\n",
"title": "Static Gesture Recognition using Leap Motion"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 18866 | null | Validated | null | null |
null | {
"abstract": " For spaces of constant, linear, and quadratic splines of maximal smoothness\non the Powell-Sabin 12-split of a triangle, the so-called S-bases were recently\nintroduced. These are simplex spline bases with B-spline-like properties on the\n12-split of a single triangle, which are tied together across triangles in a\nBézier-like manner.\nIn this paper we give a formal definition of an S-basis in terms of certain\nbasic properties. We proceed to investigate the existence of S-bases for the\naforementioned spaces and additionally the cubic case, resulting in an\nexhaustive list. From their nature as simplex splines, we derive simple\ndifferentiation and recurrence formulas to other S-bases. We establish a\nMarsden identity that gives rise to various quasi-interpolants and domain\npoints forming an intuitive control net, in terms of which conditions for\n$C^0$-, $C^1$-, and $C^2$-smoothness are derived.\n",
"title": "B-spline-like bases for $C^2$ cubics on the Powell-Sabin 12-split"
} | null | null | null | null | true | null | 18867 | null | Default | null | null |
null | {
"abstract": " It is known that the primary source of dietary vitamin C is fruit and\nvegetables and the plasma level of vitamin C has been considered a good\nsurrogate biomarker of vitamin C intake by fruit and vegetable consumption. To\ncombine the information about association between vitamin C intake and the\nplasma level of vitamin C, numerical approximation methods for likelihood\nfunction of correlation coefficient are studied. The least squares approach is\nused to estimate a log-likelihood function by a function from a space of\nB-splines having desirable mathematical properties. The likelihood interval\nfrom the Highest Likelihood Regions (HLR) is used for further inference. This\napproach can be easily extended to the realm of meta-analysis involving sample\ncorrelations from different studies by use of an approximated combined\nlikelihood function. The sample correlations between vitamin C intake and serum\nlevel of vitamin C from many studies are used to illustrate application of this\napproach.\n",
"title": "Synthesizing Correlations with Computational Likelihood Approach: Vitamin C Data"
} | null | null | null | null | true | null | 18868 | null | Default | null | null |
null | {
"abstract": " We propose a new table-top experimental configuration for the direct\ndetection of dark matter axions with mass in the $(10^{-6} \\rm eV - 10^{-2} \\rm\neV)$ range using non-perturbative effects in a system with non-trivial spatial\ntopology. Different from most experimental setups found in literature on direct\ndark matter axion detection, which relies on $\\dot{\\theta}$ or\n$\\vec{\\nabla}\\theta$, we found that our system is in principle sensitive to a\nstatic $\\theta\\geq 10^{-14}$ and can also be used to set limit on the\nfundamental constant $\\theta_{\\rm QED}$ which becomes the fundamental\nobservable parameter of the Maxwell system if some conditions are met.\nConnection with Witten effect when the induced electric charge $e'$ is\nproportional to $\\theta$ and the magnetic monopole becomes the dyon with\nnon-vanishing $e'=-e \\frac{\\theta}{2\\pi}$ is also discussed.\n",
"title": "Axion detection via Topological Casimir Effect"
} | null | null | [
"Physics"
]
| null | true | null | 18869 | null | Validated | null | null |
null | {
"abstract": " Micro-panel data are collected and analysed in many research and industry\nareas. Cluster analysis of micro-panel data is an unsupervised learning\nexploratory method identifying subgroup clusters in a data set which include\nhomogeneous objects in terms of the development dynamics of monitored\nvariables. The supply of clustering methods tailored to micro-panel data is\nlimited. The present paper focuses on a feature-based clustering method,\nintroducing a novel two-step characteristic-based approach designed for this\ntype of data. The proposed CluMP method aims to identify clusters that are at\nleast as internally homogeneous and externally heterogeneous as those obtained\nby alternative methods already implemented in the statistical system R. We\ncompare the clustering performance of the devised algorithm with two extant\nmethods using simulated micro-panel data sets. Our approach has yielded similar\nor better outcomes than the other methods, the advantage of the proposed\nalgorithm being time efficiency which makes it applicable for large data sets.\n",
"title": "Novel Feature-Based Clustering of Micro-Panel Data (CluMP)"
} | null | null | null | null | true | null | 18870 | null | Default | null | null |
null | {
"abstract": " Ab initio low-energy effective Hamiltonians of two typical high-temperature\ncopper-oxide superconductors, whose mother compounds are La$_2$CuO$_4$ and\nHgBa$_2$CuO$_4$, are derived by utilizing the multi-scale ab initio scheme for\ncorrelated electrons (MACE). The effective Hamiltonians obtained in the present\nstudy serve as platforms of future studies to accurately solve the low-energy\neffective Hamiltonians beyond the density functional theory. It allows further\nstudy on the superconducting mechanism from the first principles and\nquantitative basis without adjustable parameters not only for the available\ncuprates but also for future design of higher Tc in general. More concretely,\nwe derive effective Hamiltonians for three variations, 1)one-band Hamiltonian\nfor the antibonding orbital generated from strongly hybridized Cu\n$3d_{x^2-y^2}$ and O $2p_\\sigma$ orbitals 2)two-band Hamiltonian constructed\nfrom the antibonding orbital and Cu $3d_{3z^2-r^2}$ orbital hybridized mainly\nwith the apex oxygen $p_z$ orbital 3)three-band Hamiltonian consisting mainly\nof Cu $3d_{x^2-y^2}$ orbitals and two O $2p_\\sigma$ orbitals. Differences\nbetween the Hamiltonians for La$_2$CuO$_4$ and HgBa$_2$CuO$_4$, which have\nrelatively low and high critical temperatures, respectively, at optimally doped\ncompounds, are elucidated. The main differences are summarized as i) the oxygen\n$2p_\\sigma$ orbitals are farther(~3.7eV) below from the Cu $d_{x^2-y^2}$\norbital for the La compound than the Hg compound(~2.4eV) in the three-band\nHamiltonian. This causes a substantial difference in the character of the\n$d_{x^2-y^2}-2p_\\sigma$ antibonding band at the Fermi level and makes the\neffective onsite Coulomb interaction U larger for the La compound than the Hg\ncompound for the two- and one-band Hamiltonians. ii)The ratio of the\nsecond-neighbor to the nearest transfer t'/t is also substantially different\n(~0.26) for the Hg and ~0.15 for the La compound in the one-band Hamiltonian.\n",
"title": "Ab initio effective Hamiltonians for cuprate superconductors"
} | null | null | null | null | true | null | 18871 | null | Default | null | null |
null | {
"abstract": " In clinical practice and biomedical research, measurements are often\ncollected sparsely and irregularly in time while the data acquisition is\nexpensive and inconvenient. Examples include measurements of spine bone mineral\ndensity, cancer growth through mammography or biopsy, a progression of defect\nof vision, or assessment of gait in patients with neurological disorders. Since\nthe data collection is often costly and inconvenient, estimation of progression\nfrom sparse observations is of great interest for practitioners.\nFrom the statistical standpoint, such data is often analyzed in the context\nof a mixed-effect model where time is treated as both random and fixed effect.\nAlternatively, researchers analyze Gaussian processes or functional data where\nobservations are assumed to be drawn from a certain distribution of processes.\nThese models are flexible but rely on probabilistic assumptions and require\nvery careful implementation.\nIn this study, we propose an alternative elementary framework for analyzing\nlongitudinal data, relying on matrix completion. Our method yields point\nestimates of progression curves by iterative application of the SVD. Our\nframework covers multivariate longitudinal data, regression and can be easily\nextended to other settings.\nWe apply our methods to understand trends of progression of motor impairment\nin children with Cerebral Palsy. Our model approximates individual progression\ncurves and explains 30% of the variability. Low-rank representation of\nprogression trends enables discovering that subtypes of Cerebral Palsy exhibit\ndifferent progression trends.\n",
"title": "Longitudinal data analysis using matrix completion"
} | null | null | [
"Statistics"
]
| null | true | null | 18872 | null | Validated | null | null |
null | {
"abstract": " We present a new model of the optical nebular emission from HII regions by\ncombin- ing the results of the Binary Population and Spectral Synthesis (bpass)\ncode with the photoion- ization code cloudy (Ferland et al. 1998). We explore a\nvariety of emission-line diagnostics of these star-forming HII regions and\nexamine the effects of metallicity and interacting binary evo- lution on the\nnebula emission-line production. We compare the line emission properties of HII\nregions with model stellar populations, and provide new constraints on their\nstellar populations and supernova progenitors. We find that models including\nmassive binary stars can successfully match all the observational constraints\nand provide reasonable age and mass estimation of the HII regions and supernova\nprogenitors.\n",
"title": "Emission-line Diagnostics of Nearby HII Regions Including Supernova Hosts"
} | null | null | null | null | true | null | 18873 | null | Default | null | null |
null | {
"abstract": " Column closed pattern subgroups $U$ of the finite upper unitriangular groups\n$U_n(q)$ are defined as sets of matrices in $U_n(q)$ having zeros in a\nprescribed set of columns besides the diagonal ones. We explain Jedlitschky's\nconstruction of monomial linearisation and apply this to $C U$ yielding a\ngeneralisation of Yan's coadjoint cluster representations. Then we give a\ncomplete classification of the resulting supercharacters, by describing the\nresulting orbits and determining the Hom-spaces between orbit modules.\n",
"title": "On monomial linearisation and supercharacters of pattern subgroups"
} | null | null | null | null | true | null | 18874 | null | Default | null | null |
null | {
"abstract": " Multicomponent nanoparticles can be synthesized with either homogeneous or\nphase-segregated architectures depending on the synthesis conditions and\nelements incorporated. To understand the parameters that determine their\nstructural fate, multicomponent metal-oxide nanoparticles consisting of\ncombinations of Co, Ni, and Cu were synthesized via scanning probe block\ncopolymer lithography and characterized using correlated electron microscopy.\nThese studies revealed that the miscibility, ratio of the metallic components,\nand the synthesis temperature determine the crystal structure and architecture\nof the nanoparticles. A Co-Ni-O system forms a rock salt structure largely due\nto the miscibility of CoO and NiO, while Cu-Ni-O, which has large miscibility\ngaps, forms either homogeneous oxides, heterojunctions, or alloys depending on\nthe annealing temperature and composition. Moreover, a higher ordered\nstructure, Co-Ni-Cu-O, was found to follow the behavior of lower ordered\nsystems.\n",
"title": "The Structural Fate of Individual Multicomponent Metal-Oxide Nanoparticles in Polymer Nanoreactors"
} | null | null | null | null | true | null | 18875 | null | Default | null | null |
null | {
"abstract": " Unsupervised neural nets such as Restricted Boltzmann Machines(RBMs) and Deep\nBelif Networks(DBNs), are powerful in automatic feature extraction,unsupervised\nweight initialization and density estimation. In this paper,we demonstrate that\nthe parameters of these neural nets can be dramatically reduced without\naffecting their performance. We describe a method to reduce the parameters\nrequired by RBM which is the basic building block for deep architectures.\nFurther we propose an unsupervised sparse deep architectures selection\nalgorithm to form sparse deep neural networks.Experimental results show that\nthere is virtually no loss in either generative or discriminative performance.\n",
"title": "On Compression of Unsupervised Neural Nets by Pruning Weak Connections"
} | null | null | null | null | true | null | 18876 | null | Default | null | null |
null | {
"abstract": " In 1947 Nathan Fine gave a beautiful product for the number of binomial\ncoefficients $\\binom{n}{m}$, for $m$ in the range $0 \\leq m \\leq n$, that are\nnot divisible by $p$. We give a matrix product that generalizes Fine's formula,\nsimultaneously counting binomial coefficients with $p$-adic valuation $\\alpha$\nfor each $\\alpha \\geq 0$. For each $n$ this information is naturally encoded in\na polynomial generating function, and the sequence of these polynomials is\n$p$-regular in the sense of Allouche and Shallit. We also give a further\ngeneralization to multinomial coefficients.\n",
"title": "A matrix generalization of a theorem of Fine"
} | null | null | null | null | true | null | 18877 | null | Default | null | null |
null | {
"abstract": " Under noisy environments, to achieve the robust performance of speaker\nrecognition is still a challenging task. Motivated by the promising performance\nof multi-task training in a variety of image processing tasks, we explore the\npotential of multi-task adversarial training for learning a noise-robust\nspeaker embedding. In this paper we present a novel framework which consists of\nthree components: an encoder that extracts noise-robust speaker embedding; a\nclassifier that classifies the speakers; a discriminator that discriminates the\nnoise type of the speaker embedding. Besides, we propose a training strategy\nusing the training accuracy as an indicator to stabilize the multi-class\nadversarial optimization process. We conduct our experiments on the English and\nMandarin corpus and the experimental results demonstrate that our proposed\nmulti-task adversarial training method could greatly outperform the other\nmethods without adversarial training in noisy environments. Furthermore,\nexperiments indicate that our method is also able to improve the speaker\nverification performance the clean condition.\n",
"title": "Training Multi-Task Adversarial Network For Extracting Noise-Robust Speaker Embedding"
} | null | null | null | null | true | null | 18878 | null | Default | null | null |
null | {
"abstract": " In this paper, we consider zero-sum repeated games in which the maximizer is\nrestricted to strategies requiring no more than a limited amount of randomness.\nParticularly, we analyze the maxmin payoff of the maximizer in two models: the\nfirst model forces the maximizer to randomize her action in each stage just by\nconditioning her decision to outcomes of a given sequence of random source,\nwhereas, in the second model, the maximizer is a team of players who are free\nto privately randomize their corresponding actions but do not have access to\nany explicit source of shared randomness needed for cooperation. The works of\nGossner and Vieille, and Gossner and Tomala adopted the method of types to\nestablish their results; however, we utilize the idea of random hashing which\nis the core of randomness extractors in the information theory literature. In\naddition, we adopt the well-studied tool of simulation of a source from another\nsource. By utilizing these tools, we are able to simplify the prior results and\ngeneralize them as well. We characterize the maxmin payoff of the maximizer in\nthe repeated games under study. Particularly, the maxmin payoff of the first\nmodel is fully described by the function $J(h)$ which is the maximum payoff\nthat the maximizer can secure in a one-shot game by choosing mixed strategies\nof entropy at most $h$. In the second part of the paper, we study the\ncomputational aspects of $J(h)$. We offer three explicit lower bounds on the\nentropy-payoff trade-off curve. To do this, we provide and utilize new results\nfor the set of distributions that guarantee a certain payoff for Alice. In\nparticular, we study how this set of distributions shrinks as we increase the\nsecurity level. While the use of total variation distance is common in game\ntheory, our derivation indicates the suitability of utilizing the\nRenyi-divergence of order two.\n",
"title": "Playing Games with Bounded Entropy"
} | null | null | null | null | true | null | 18879 | null | Default | null | null |
null | {
"abstract": " Random code-trees with necks were introduced recently to generalise the\nnotion of $V$-variable and random homogeneous sets. While it is known that the\nHausdorff and packing dimensions coincide irrespective of overlaps, their exact\nHausdorff and packing measure has so far been largely ignored. In this article\nwe consider the general question of an appropriate gauge function for positive\nand finite Hausdorff and packing measure. We first survey the current state of\nknowledge and establish some bounds on these gauge functions. We then show that\nself-similar code-trees do not admit a gauge functions that simultaneously give\npositive and finite Hausdorff measure almost surely. This surprising result is\nin stark contrast to the random recursive model and sheds some light on the\nquestion of whether $V$-variable sets interpolate between random homogeneous\nand random recursive sets. We conclude by discussing implications of our\nresults.\n",
"title": "Exact Hausdorff and packing measures for random self-similar code-trees with necks"
} | null | null | null | null | true | null | 18880 | null | Default | null | null |
null | {
"abstract": " Let R be a commutative ring with unity, M a module over R and let S be a\nG-set for a finite group G. We define a set MS to be the set of elements\nexpressed as the formal finite sum of the form similar to the elements of group\nring RG. The set MS is a module over the group ring RG under the addition and\nthe scalar multiplication similar to the RG-module MG. With this notion, we not\nonly generalize but also unify the theories of both of the group algebra and\nthe group module, and we also establish some significant properties of MS. In\nparticular, we describe a method for decomposing a given RG-module MS as a\ndirect sum of RG-submodules. Furthermore, we prove the semisimplicity problem\nof MS with regard to the properties of M, S and G.\n",
"title": "On Modules over a G-set"
} | null | null | null | null | true | null | 18881 | null | Default | null | null |
null | {
"abstract": " We give the first polynomial-time algorithms on graphs of bounded maximum\ninduced matching width (mim-width) for problems that are not locally checkable.\nIn particular, we give $n^{\\mathcal{O}(w)}$-time algorithms on graphs of\nmim-width at most $w$, when given a decomposition, for the following problems:\nLongest Induced Path, Induced Disjoint Paths and $H$-Induced Topological Minor\nfor fixed $H$. Our results imply that the following graph classes have\npolynomial-time algorithms for these three problems: Interval and Bi-Interval\ngraphs, Circular Arc, Permutation and Circular Permutation graphs, Convex\ngraphs, $k$-Trapezoid, Circular $k$-Trapezoid, $k$-Polygon, Dilworth-$k$ and\nCo-$k$-Degenerate graphs for fixed $k$.\n",
"title": "Polynomial-time algorithms for the Longest Induced Path and Induced Disjoint Paths problems on graphs of bounded mim-width"
} | null | null | [
"Computer Science"
]
| null | true | null | 18882 | null | Validated | null | null |
null | {
"abstract": " Previous studies have found that a significant number of bug reports are\nmisclassified between bugs and non-bugs, and that manually classifying bug\nreports is a time-consuming task. To address this problem, we propose a bug\nreports classification model with N-gram IDF, a theoretical extension of\nInverse Document Frequency (IDF) for handling words and phrases of any length.\nN-gram IDF enables us to extract key terms of any length from texts, these key\nterms can be used as the features to classify bug reports. We build\nclassification models with logistic regression and random forest using features\nfrom N-gram IDF and topic modeling, which is widely used in various software\nengineering tasks. With a publicly available dataset, our results show that our\nN-gram IDF-based models have a superior performance than the topic-based models\non all of the evaluated cases. Our models show promising results and have a\npotential to be extended to other software engineering tasks.\n",
"title": "Bug or Not? Bug Report Classification Using N-Gram IDF"
} | null | null | [
"Computer Science"
]
| null | true | null | 18883 | null | Validated | null | null |
null | {
"abstract": " Cell nuclei detection is a challenging research topic because of limitations\nin cellular image quality and diversity of nuclear morphology, i.e. varying\nnuclei shapes, sizes, and overlaps between multiple cell nuclei. This has been\na topic of enduring interest with promising recent success shown by deep\nlearning methods. These methods train Convolutional Neural Networks (CNNs) with\na training set of input images and known, labeled nuclei locations. Many such\nmethods are supplemented by spatial or morphological processing. Using a set of\ncanonical cell nuclei shapes, prepared with the help of a domain expert, we\ndevelop a new approach that we call Shape Priors with Convolutional Neural\nNetworks (SP-CNN). We further extend the network to introduce a shape prior\n(SP) layer and then allowing it to become trainable (i.e. optimizable). We call\nthis network tunable SP-CNN (TSP-CNN). In summary, we present new network\nstructures that can incorporate 'expected behavior' of nucleus shapes via two\ncomponents: learnable layers that perform the nucleus detection and a fixed\nprocessing part that guides the learning with prior information. Analytically,\nwe formulate two new regularization terms that are targeted at: 1) learning the\nshapes, 2) reducing false positives while simultaneously encouraging detection\ninside the cell nucleus boundary. Experimental results on two challenging\ndatasets reveal that the proposed SP-CNN and TSP-CNN can outperform\nstate-of-the-art alternatives.\n",
"title": "Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection"
} | null | null | null | null | true | null | 18884 | null | Default | null | null |
null | {
"abstract": " Recently, the vertical shear instability (VSI) has become an attractive\npurely hydrodynamic candidate for the anomalous angular momentum transport\nrequired for weakly ionized accretion disks. In direct three-dimensional\nnumerical simulations of VSI turbulence in disks, a meridional circulation\npattern was observed that is opposite to the usual viscous flow behavior. Here,\nwe investigate whether this feature can possibly be explained by an anisotropy\nof the VSI turbulence. Using three-dimensional hydrodynamical simulations, we\ncalculate the turbulent Reynolds stresses relevant for angular momentum\ntransport for a representative section of a disk.\nWe find that the vertical stress is significantly stronger than the radial\nstress. Using our results in viscous disk simulations with different viscosity\ncoefficients for the radial and vertical direction, we find good agreement with\nthe VSI turbulence for the stresses and meridional flow; this provides\nadditional evidence for the anisotropy. The results are important with respect\nto the transport of small embedded particles in disks.\n",
"title": "Anisotropic hydrodynamic turbulence in accretion disks"
} | null | null | [
"Physics"
]
| null | true | null | 18885 | null | Validated | null | null |
null | {
"abstract": " In this paper, we study the optimal convergence rate for distributed convex\noptimization problems in networks. We model the communication restrictions\nimposed by the network as a set of affine constraints and provide optimal\ncomplexity bounds for four different setups, namely: the function $F(\\xb)\n\\triangleq \\sum_{i=1}^{m}f_i(\\xb)$ is strongly convex and smooth, either\nstrongly convex or smooth or just convex. Our results show that Nesterov's\naccelerated gradient descent on the dual problem can be executed in a\ndistributed manner and obtains the same optimal rates as in the centralized\nversion of the problem (up to constant or logarithmic factors) with an\nadditional cost related to the spectral gap of the interaction matrix. Finally,\nwe discuss some extensions to the proposed setup such as proximal friendly\nfunctions, time-varying graphs, improvement of the condition numbers.\n",
"title": "Optimal Algorithms for Distributed Optimization"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 18886 | null | Validated | null | null |
null | {
"abstract": " Brain-Computer Interface (BCI) uses brain signals in order to provide a new\nmethod for communication between human and outside world. Feature extraction,\nselection and classification are among the main matters of concerns in signal\nprocessing stage of BCI. In this article, we present our findings about the\nmost effective features and classifiers in some brain tasks. Six different\ngroups of classical features and twelve classifiers have been examined in nine\ndatasets of brain signal. The results indicate that energy of brain signals in\n{\\alpha} and \\b{eta} frequency bands, together with some statistical parameters\nare more effective, comparing to the other types of extracted features. In\naddition, Bayesian classifier with Gaussian distribution assumption and also\nSupport Vector Machine (SVM) show to classify different BCI datasets more\naccurately than the other classifiers. We believe that the results can give an\ninsight about a strategy for blind classification of brain signals in\nbrain-computer interface.\n",
"title": "Evaluation of Classical Features and Classifiers in Brain-Computer Interface Tasks"
} | null | null | null | null | true | null | 18887 | null | Default | null | null |
null | {
"abstract": " We provide a uniform framework to study the exceptional homogeneous compact\ngeometries of type C3. This framework is then used to show that these are\nsimply connected, answering a question by Kramer and Lytchak, and to calculate\nthe full automorphism groups.\n",
"title": "On exceptional compact homogeneous geometries of type C3"
} | null | null | null | null | true | null | 18888 | null | Default | null | null |
null | {
"abstract": " In the following paper we present a simple intensity estimation method of\ntransaction arrivals on the intraday electricity market. Assuming the\ninterarrival times distribution, we utilize a maximum likelihood estimation.\nThe method's performance is briefly tested using German Intraday Continuous\ndata. Despite the simplicity of the method, the results are encouraging. The\nsupplementary materials containing the R-codes and the data are attached to\nthis paper.\n",
"title": "Intensity estimation of transaction arrivals on the intraday electricity market"
} | null | null | null | null | true | null | 18889 | null | Default | null | null |
null | {
"abstract": " Oxygen functional groups are one of the most important subjects in the study\nof electrochemical properties of carbon materials which can change the\nwettability, conductivity and pore size distributions of carbon materials, and\ncan occur redox reactions. In the electrode materials of carbon-based\nsupercapacitors, the oxygen functional groups have widely been used to improve\nthe capacitive performance. In this paper, we not only analyzed the reasons for\nthe increase of the capacity that promoted by oxygen functional groups in the\ncharge-discharge cycling tests, but also analyzed the mechanism how the\npseudocapacitance was provided by the oxygen functional groups in the\nacid/alkaline aqueous electrolyte. Moreover, we also discussed the effect of\nthe oxygen functional groups in electrochemical impedance spectroscopy.\n",
"title": "Capacitive Mechanism of Oxygen Functional Groups on Carbon Surface in Supercapacitors"
} | null | null | [
"Physics"
]
| null | true | null | 18890 | null | Validated | null | null |
null | {
"abstract": " We study asymptotic properties of conditional least squares estimators for\nthe drift parameters of two-factor affine diffusions based on continuous time\nobservations. We distinguish three cases: subcritical, critical and\nsupercritical. For all the drift parameters, in the subcritical and\nsupercritical cases, asymptotic normality and asymptotic mixed normality is\nproved, while in the critical case, non-standard asymptotic behavior is\ndescribed.\n",
"title": "On conditional least squares estimation for affine diffusions based on continuous time observations"
} | null | null | null | null | true | null | 18891 | null | Default | null | null |
null | {
"abstract": " This paper is concerned with minimization of a fourth-order linearized\nCanham-Helfrich energy subject to Dirichlet boundary conditions on curves\ninside the domain. Such problems arise in the modeling of the mechanical\ninteraction of biomembranes with embedded particles. There, the curve\nconditions result from the imposed particle--membrane coupling. We prove\nalmost-$H^{\\frac{5}{2}}$ regularity of the solution and then consider two\npossible penalty formulations. For the combination of these penalty\nformulations with a Bogner-Fox-Schmit finite element discretization we prove\ndiscretization error estimates which are optimal in view of the solution's\nreduced regularity. The error estimates are based on a general estimate for\nlinear penalty problems in Hilbert spaces. Finally, we illustrate the\ntheoretical results by numerical computations. An important feature of the\npresented discretization is that it does not require to resolve the particle\nboundary. This is crucial in order to avoid re-meshing if the presented problem\narises as subproblem in a model where particles are allowed to move or rotate.\n",
"title": "Discretization error estimates for penalty formulations of a linearized Canham-Helfrich type energy"
} | null | null | null | null | true | null | 18892 | null | Default | null | null |
null | {
"abstract": " Let $G$ be a simply-connected semisimple algebraic group over an\nalgebraically closed field of characteristic $p$, assumed to be larger than the\nCoxeter number. The \"support variety\" of a $G$-module $M$ is a certain closed\nsubvariety of the nilpotent cone of $G$, defined in terms of cohomology for the\nfirst Frobenius kernel $G_1$. In the 1990s, Humphreys proposed a conjectural\ndescription of the support varieties of tilting modules; this conjecture has\nbeen proved for $G = \\mathrm{SL}_n$ in earlier work of the second author.\nIn this paper, we show that for any $G$, the support variety of a tilting\nmodule always contains the variety predicted by Humphreys, and that they\ncoincide (i.e., the Humphreys conjecture is true) when $p$ is sufficiently\nlarge. We also prove variants of these statements involving \"relative support\nvarieties.\"\n",
"title": "On the Humphreys conjecture on support varieties of tilting modules"
} | null | null | null | null | true | null | 18893 | null | Default | null | null |
null | {
"abstract": " Machine learning models have been widely used in security applications such\nas intrusion detection, spam filtering, and virus or malware detection.\nHowever, it is well-known that adversaries are always trying to adapt their\nattacks to evade detection. For example, an email spammer may guess what\nfeatures spam detection models use and modify or remove those features to avoid\ndetection. There has been some work on making machine learning models more\nrobust to such attacks. However, one simple but promising approach called {\\em\nrandomization} is underexplored. This paper proposes a novel\nrandomization-based approach to improve robustness of machine learning models\nagainst evasion attacks. The proposed approach incorporates randomization into\nboth model training time and model application time (meaning when the model is\nused to detect attacks). We also apply this approach to random forest, an\nexisting ML method which already has some degree of randomness. Experiments on\nintrusion detection and spam filtering data show that our approach further\nimproves robustness of random-forest method. We also discuss how this approach\ncan be applied to other ML models.\n",
"title": "Using Randomness to Improve Robustness of Machine-Learning Models Against Evasion Attacks"
} | null | null | null | null | true | null | 18894 | null | Default | null | null |
null | {
"abstract": " The main aim of this paper is to give a new generalization of Hurwitz-Lerch\nZeta function of two variables.Also, we investigate several interesting\nproperties such as integral representations, summation formula and a connection\nwith generalized hypergeometric function. To strengthen the main results we\nalso consider many important special cases.\n",
"title": "Further extension of the generalized Hurwitz-Lerch Zeta function of two variables"
} | null | null | null | null | true | null | 18895 | null | Default | null | null |
null | {
"abstract": " Stochastic user equilibrium is an important issue in the traffic assignment\nproblems, tradition models for the stochastic user equilibrium problem are\ndesigned as mathematical programming problems. In this article, a\nPhysarum-inspired model for the probit-based stochastic user equilibrium\nproblem is proposed. There are two main contributions of our work. On the one\nhand, the origin Physarum model is modified to find the shortest path in\ntraffic direction networks with the properties of two-way traffic\ncharacteristic. On the other hand, the modified Physarum-inspired model could\nget the equilibrium flows when traveller's perceived transportation cost\ncomplies with normal distribution. The proposed method is constituted with a\ntwo-step procedure. First, the modified Physarum model is applied to get the\nauxiliary flows. Second, the auxiliary flows are averaged to obtain the\nequilibrium flows. Numerical examples are conducted to illustrate the\nperformance of the proposed method, which is compared with the Method of\nSuccessive Average method.\n",
"title": "A Physarum-inspired model for the probit-based stochastic user equilibrium problem"
} | null | null | null | null | true | null | 18896 | null | Default | null | null |
null | {
"abstract": " We investigate quantifier alternation hierarchies in first-order logic on\nfinite words. Levels in these hierarchies are defined by counting the number of\nquantifier alternations in formulas. We prove that one can decide membership of\na regular language in the levels $\\mathcal{B}{\\Sigma}_2$ (finite boolean\ncombinations of formulas having only one alternation) and ${\\Sigma}_3$\n(formulas having only two alternations and beginning with an existential\nblock). Our proofs work by considering a deeper problem, called separation,\nwhich, once solved for lower levels, allows us to solve membership for higher\nlevels.\n",
"title": "Going Higher in First-Order Quantifier Alternation Hierarchies on Words"
} | null | null | null | null | true | null | 18897 | null | Default | null | null |
null | {
"abstract": " Let $\\mathcal{I}$ be an analytic P-ideal [respectively, a summable ideal] on\nthe positive integers and let $(x_n)$ be a sequence taking values in a metric\nspace $X$. First, it is shown that the set of ideal limit points of $(x_n)$ is\nan $F_\\sigma$-set [resp., a closet set]. Let us assume that $X$ is also\nseparable and the ideal $\\mathcal{I}$ satisfies certain additional assumptions,\nwhich however includes several well-known examples, e.g., the collection of\nsets with zero asymptotic density, sets with zero logarithmic density, and some\nsummable ideals. Then, it is shown that the set of ideal limit points of\n$(x_n)$ is equal to the set of ideal limit points of almost all its\nsubsequences.\n",
"title": "Invariance of Ideal Limit Points"
} | null | null | [
"Mathematics"
]
| null | true | null | 18898 | null | Validated | null | null |
null | {
"abstract": " The unprecedented demand for large amount of data has catalyzed the trend of\ncombining human insights with machine learning techniques, which facilitate the\nuse of crowdsourcing to enlist label information both effectively and\nefficiently. The classic work on crowdsourcing mainly focuses on the label\ninference problem under the categorization setting. However, inferring the true\nlabel requires sophisticated aggregation models that usually can only perform\nwell under certain assumptions. Meanwhile, no matter how complicated the\naggregation model is, the true model that generated the crowd labels remains\nunknown. Therefore, the label inference problem can never infer the ground\ntruth perfectly. Based on the fact that the crowdsourcing labels are abundant\nand utilizing aggregation will lose such kind of rich annotation information\n(e.g., which worker provided which labels), we believe that it is critical to\ntake the diverse labeling abilities of the crowdsourcing workers as well as\ntheir correlations into consideration. To address the above challenge, we\npropose to tackle three research problems, namely inference, learning, and\nteaching.\n",
"title": "Optimizing the Wisdom of the Crowd: Inference, Learning, and Teaching"
} | null | null | [
"Statistics"
]
| null | true | null | 18899 | null | Validated | null | null |
null | {
"abstract": " The hardness of the learning with errors (LWE) problem is one of the most\nfruitful resources of modern cryptography. In particular, it is one of the most\nprominent candidates for secure post-quantum cryptography. Understanding its\nquantum complexity is therefore an important goal. We show that under quantum\npolynomial time reductions, LWE is equivalent to a relaxed version of the\ndihedral coset problem (DCP), which we call extrapolated DCP (eDCP). The extent\nof extrapolation varies with the LWE noise rate. By considering different\nextents of extrapolation, our result generalizes Regev's famous proof that if\nDCP is in BQP (quantum poly-time) then so is LWE (FOCS'02). We also discuss a\nconnection between eDCP and Childs and Van Dam's algorithm for generalized\nhidden shift problems (SODA'07). Our result implies that a BQP solution for LWE\nmight not require the full power of solving DCP, but rather only a solution for\nits relaxed version, eDCP, which could be easier.\n",
"title": "Learning With Errors and Extrapolated Dihedral Cosets"
} | null | null | [
"Computer Science"
]
| null | true | null | 18900 | null | Validated | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.