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": " This paper describes a new algorithm for solar energy forecasting from a\nsequence of Cloud Optical Depth (COD) images. The algorithm is based on the\nfollowing simple observation: the dynamics of clouds represented by COD images\nresembles the motion (transport) of a density in a fluid flow. This suggests\nthat, to forecast the motion of COD images, it is sufficient to forecast the\nflow. The latter, in turn, can be accomplished by fitting a parametric model of\nthe fluid flow to the COD images observed in the past. Namely, the learning\nphase of the algorithm is composed of the following steps: (i) given a sequence\nof COD images, the snapshots of the optical flow are estimated from two\nconsecutive COD images; (ii) these snapshots are then assimilated into a\nNavier-Stokes Equation (NSE), i.e. an initial velocity field for NSE is\nselected so that the corresponding NSE' solution is as close as possible to the\noptical flow snapshots. The prediction phase consists of utilizing a linear\ntransport equation, which describes the propagation of COD images in the fluid\nflow predicted by NSE, to estimate the future motion of the COD images. The\nalgorithm has been tested on COD images provided by two geostationary\noperational environmental satellites from NOAA serving the west-hemisphere.\n",
"title": "Where computer vision can aid physics: dynamic cloud motion forecasting from satellite images"
}
| null | null | null | null | true | null |
11101
| null |
Default
| null | null |
null |
{
"abstract": " We determine the group strucure of the $23$-rd homotopy group $\\pi_{23}(G_2 :\n2)$, where $G_2$ is the Lie group of exceptional type, which hasn't been\ndetermined for $50$ years.\n",
"title": "On Mimura's extension problem"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11102
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we investigate existence and non-existence of a nontrivial\nsolution to the pseudo-relativistic nonlinear Schrödinger equation $$\\left(\n\\sqrt{-c^2\\Delta + m^2 c^4}-mc^2\\right) u + \\mu u = |u|^{p-1}u\\quad\n\\textrm{in}~\\mathbb{R}^n~(n \\geq 2)$$ involving an\n$H^{1/2}$-critical/supercritical power-type nonlinearity, i.e., $p \\geq\n\\frac{n+1}{n-1}$. We prove that in the non-relativistic regime, there exists a\nnontrivial solution provided that the nonlinearity is\n$H^{1/2}$-critical/supercritical but it is $H^1$-subcritical. On the other\nhand, we also show that there is no nontrivial bounded solution either $(i)$ if\nthe nonlinearity is $H^{1/2}$-critical/supercritical in the ultra-relativistic\nregime or $(ii)$ if the nonlinearity is $H^1$-critical/supercritical in all\ncases.\n",
"title": "On critical and supercritical pseudo-relativistic nonlinear Schrödinger equations"
}
| null | null | null | null | true | null |
11103
| null |
Default
| null | null |
null |
{
"abstract": " Purpose. We present a new method to evaluate the accuracy of an eye tracker\nbased eye localization system. Measuring the accuracy of an eye tracker's\nprimary intention, the estimated point of gaze, is usually done with volunteers\nand a set of fixation points used as ground truth. However, verifying the\naccuracy of the location estimate of a volunteer's eye center in 3D space is\nnot easily possible. This is because the eye center is an intangible point\nhidden by the iris. Methods. We evaluate the eye location accuracy by using an\neye phantom instead of eyes of volunteers. For this, we developed a testing\nstage with a realistic artificial eye and a corresponding kinematic model,\nwhich we trained with {\\mu}CT data. This enables us to precisely evaluate the\neye location estimate of an eye tracker. Results. We show that the proposed\ntesting stage with the corresponding kinematic model is suitable for such a\nvalidation. Further, we evaluate a particular eye tracker based navigation\nsystem and show that this system is able to successfully determine the eye\ncenter with sub-millimeter accuracy. Conclusions. We show the suitability of\nthe evaluated eye tracker for eye interventions, using the proposed testing\nstage and the corresponding kinematic model. The results further enable\nspecific enhancement of the navigation system to potentially get even better\nresults.\n",
"title": "Eye Tracker Accuracy: Quantitative Evaluation of the Invisible Eye Center Location"
}
| null | null | null | null | true | null |
11104
| null |
Default
| null | null |
null |
{
"abstract": " Feature selection problems arise in a variety of applications, such as\nmicroarray analysis, clinical prediction, text categorization, image\nclassification and face recognition, multi-label learning, and classification\nof internet traffic. Among the various classes of methods, forward feature\nselection methods based on mutual information have become very popular and are\nwidely used in practice. However, comparative evaluations of these methods have\nbeen limited by being based on specific datasets and classifiers. In this\npaper, we develop a theoretical framework that allows evaluating the methods\nbased on their theoretical properties. Our framework is grounded on the\nproperties of the target objective function that the methods try to\napproximate, and on a novel categorization of features, according to their\ncontribution to the explanation of the class; we derive upper and lower bounds\nfor the target objective function and relate these bounds with the feature\ntypes. Then, we characterize the types of approximations taken by the methods,\nand analyze how these approximations cope with the good properties of the\ntarget objective function. Additionally, we develop a distributional setting\ndesigned to illustrate the various deficiencies of the methods, and provide\nseveral examples of wrong feature selections. Based on our work, we identify\nclearly the methods that should be avoided, and the methods that currently have\nthe best performance.\n",
"title": "Theoretical Foundations of Forward Feature Selection Methods based on Mutual Information"
}
| null | null | null | null | true | null |
11105
| null |
Default
| null | null |
null |
{
"abstract": " We develop an approach to realize a quantum switch for Rydberg excitation in\natoms with $Y$-typed level configuration. We find that the steady population on\ntwo different Rydberg states can be reversibly exchanged in a controllable way\nby properly tuning the Rydberg-Rydberg interaction. Moreover, our numerical\nsimulations verify that the switching scheme is robust against spontaneous\ndecay, environmental disturbance, as well as the duration of operation on the\ninteraction, and also a high switching efficiency is quite attainable, which\nmakes it have potential applications in quantum information processing and\nother Rydberg-based quantum technologies.\n",
"title": "Robust quantum switch with Rydberg excitations"
}
| null | null | null | null | true | null |
11106
| null |
Default
| null | null |
null |
{
"abstract": " Pemantle and Steif provided a sharp threshold for the existence of a RPT\n(robust phase transition) for the continuous rotator model and the Potts model\nin terms of the branching number and the second eigenvalue of the transfer\noperator, where a robust phase transition is said to occur if an arbitrarily\nweak coupling with symmetry-breaking boundary conditions suffices to induce\nsymmetry breaking in the bulk. They further showed that for the Potts model RPT\noccurs at a different threshold than PT (phase transition in the sense of\nmultiple Gibbs measures), and conjectured that RPT and PT should occur at the\nsame threshold in the continuous rotator model. We consider the class of 4- and\n5-state rotation-invariant spin models with reflection symmetry on general\ntrees which contains the Potts model and the clock model with\nscalarproduct-interaction as limiting cases. The clock model can be viewed as a\nparticular discretization which is obtained from the classical rotator model on\nthe continuous one-dimensional sphere. We analyze the transition between PT=RPT\nand PT is unequal to RPT, in terms of the eigenvalues of the transfer matrix of\nthe model at the critical threshold value for the existence of RPT. The\ntransition between the two regimes depends sensitively on the third largest\neigenvalue.\n",
"title": "Non-robust phase transitions in the generalized clock model on trees"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11107
| null |
Validated
| null | null |
null |
{
"abstract": " The digital traces we leave behind when engaging with the modern world offer\nan interesting lens through which we study behavioral patterns as expression of\ngender. Although gender differentiation has been observed in a number of\nsettings, the majority of studies focus on a single data stream in isolation.\nHere we use a dataset of high resolution data collected using mobile phones, as\nwell as detailed questionnaires, to study gender differences in a large cohort.\nWe consider mobility behavior and individual personality traits among a group\nof more than $800$ university students. We also investigate interactions among\nthem expressed via person-to-person contacts, interactions on online social\nnetworks, and telecommunication. Thus, we are able to study the differences\nbetween male and female behavior captured through a multitude of channels for a\nsingle cohort. We find that while the two genders are similar in a number of\naspects, there are robust deviations that include multiple facets of social\ninteractions, suggesting the existence of inherent behavioral differences.\nFinally, we quantify how aspects of an individual's characteristics and social\nbehavior reveals their gender by posing it as a classification problem. We ask:\nHow well can we distinguish between male and female study participants based on\nbehavior alone? Which behavioral features are most predictive?\n",
"title": "The Role of Gender in Social Network Organization"
}
| null | null | null | null | true | null |
11108
| null |
Default
| null | null |
null |
{
"abstract": " For a free presentation $0 \\to R \\to F \\to G \\to 0$ of a Leibniz algebra $G$,\nthe Baer invariant ${\\cal M}^{\\sf Lie}(G) = \\frac{R \\cap [F, F]_{Lie}}{[F,\nR]_{Lie}}$ is called the Schur multiplier of $G$ relative to the Liezation\nfunctor or Schur Lie-multiplier. For a two-sided ideal $N$ of a Leibniz algebra\n$G$, we construct a four-term exact sequence relating the Schur Lie-multiplier\nof $G$ and $G/N$, which is applied to study and characterize Lie-nilpotency,\nLie-stem covers and Lie-capability of Leibniz algebras.\n",
"title": "The Schur Lie-Multiplier of Leibinz Algebras"
}
| null | null | null | null | true | null |
11109
| null |
Default
| null | null |
null |
{
"abstract": " This work presents a new tool to verify the correctness of cryptographic\nimplementations with respect to cache attacks. Our methodology discovers\nvulnerabilities that are hard to find with other techniques, observed as\nexploitable leakage. The methodology works by identifying secret dependent\nmemory and introducing forced evictions inside potentially vulnerable code to\nobtain cache traces that are analyzed using Mutual Information. If dependence\nis observed, the cryptographic implementation is classified as to leak\ninformation.\nWe demonstrate the viability of our technique in the design of the three main\ncryptographic primitives, i.e., AES, RSA and ECC, in eight popular up to date\ncryptographic libraries, including OpenSSL, Libgcrypt, Intel IPP and NSS. Our\nresults show that cryptographic code designers are far away from incorporating\nthe appropriate countermeasures to avoid cache leakages, as we found that 50%\nof the default implementations analyzed leaked information that lead to key\nextraction. We responsibly notified the designers of all the leakages found and\nsuggested patches to solve these vulnerabilities.\n",
"title": "Did we learn from LLC Side Channel Attacks? A Cache Leakage Detection Tool for Crypto Libraries"
}
| null | null | null | null | true | null |
11110
| null |
Default
| null | null |
null |
{
"abstract": " We consider layered decorated honeycomb lattices at two-thirds filling, as\nrealized in some trinuclear organometallic complexes. Localized $S=1$ moments\nwith a single-spin anisotropy emerge from the interplay of Coulomb repulsion\nand spin molecular-orbit coupling (SMOC). Magnetic anisotropies with bond\ndependent exchange couplings occur in the honeycomb layers when the direct\nintracluster exchange and the spin molecular-orbital coupling are both present.\nWe find that the effective spin exchange model within the layers is an XXZ +\n120$^\\circ$ honeycomb quantum compass model. The intrinsic non-spherical\nsymmetry of the multinuclear complexes leads to very different transverse and\nlongitudinal spin molecular-orbital couplings, which greatly enhances the\nsingle-spin and exchange coupling anisotropies. The interlayer coupling is\ndescribed by a XXZ model with anisotropic biquadratic terms. As the correlation\nstrength increases the systems becomes increasingly one-dimensional. Thus, if\nthe ratio of SMOC to the interlayer hopping is small this stabilizes the\nHaldane phase. However, as the ratio increases there is a quantum phase\ntransition to the topologically trivial `$D$-phase'. We also predict a quantum\nphase transition from a Haldane phase to a magnetically ordered phase at\nsufficiently strong external magnetic fields.\n",
"title": "Effects of anisotropy in spin molecular-orbital coupling on effective spin models of trinuclear organometallic complexes"
}
| null | null | null | null | true | null |
11111
| null |
Default
| null | null |
null |
{
"abstract": " The optical vortex coronagraph (OVC) is one of the promising ways for direct\nimaging exoplanets because of its small inner working angle and high\nthroughput. This paper presents the design and laboratory demonstration\nperformance at 633nm and 1520nm of the OVC based on liquid crystal polymers\n(LCP). Two LCPs has been manufactured in partnership with a commercial vendor.\nThe OVC can deliver a good performance in laboratory test and achieve the\ncontrast of the order 10^-6 at angular distance 3{\\lambda}/D, which is able to\nimage the giant exoplanets at a young stage in combination with extreme\nadaptive optics.\n",
"title": "Design and experimental test of an optical vortex coronagraph"
}
| null | null | null | null | true | null |
11112
| null |
Default
| null | null |
null |
{
"abstract": " Symbolic computation is an important approach in automated program analysis.\nMost state-of-the-art tools perform symbolic computation as interpreters and\ndirectly maintain symbolic data. In this paper, we show that it is feasible,\nand in fact practical, to use a compiler-based strategy instead. Using compiler\ntooling, we propose and implement a transformation which takes a standard\nprogram and outputs a program that performs semantically equivalent, but\npartially symbolic, computation. The transformed program maintains symbolic\nvalues internally and operates directly on them hence the program can be\nprocessed by a tool without support for symbolic manipulation.\nThe main motivation for the transformation is in symbolic verification, but\nthere are many other possible use-cases, including test generation and concolic\ntesting. Moreover using the transformation simplifies tools, since the symbolic\ncomputation is handled by the program directly. We have implemented the\ntransformation at the level of LLVM bitcode. The paper includes an experimental\nevaluation, based on an explicit-state software model checker as a verification\nbackend.\n",
"title": "Symbolic Computation via Program Transformation"
}
| null | null | null | null | true | null |
11113
| null |
Default
| null | null |
null |
{
"abstract": " We present a general model allowing \"quantum simulation\" of one-dimensional\nDirac models with 2- and 4-component spinors using ultracold atoms in driven 1D\ntilted optical latices. The resulting Dirac physics is illustrated by one of\nits well-known manifestations, Zitterbewegung. This general model can be\nextended and applied with great flexibility to more complex situations.\n",
"title": "Simulating Dirac models with ultracold atoms in optical lattices"
}
| null | null | null | null | true | null |
11114
| null |
Default
| null | null |
null |
{
"abstract": " Network analysis techniques remain rarely used for understanding\ninternational management strategies. Our paper highlights their value as\nresearch tool in this field of social science using a large set of micro-data\n(20,000) to investigate the presence of networks of subsidiaries overseas. The\nresearch question is the following: to what extent did/do global Japanese\nbusiness networks mirror organizational models existing in Japan? In\nparticular, we would like to assess how much the links building such business\nnetworks are shaped by the structure of big-size industrial conglomerates of\nfirms headquartered in Japan, also described as HK. The major part of the\nacademic community in the fields of management and industrial organization\nconsiders that formal links can be identified among firms belonging to HK. Miwa\nand Ramseyer (Miwa and Ramseyer 2002; Ramseyer 2006) challenge this claim and\nargue that the evidence supporting the existence of HK is weak. So far,\nquantitative empirical investigation has been conducted exclusively using data\nfor firms incorporated in Japan. Our study tests the Miwa-Ramseyer hypothesis\n(MRH) at the global level using information on the network of Japanese\nsubsidiaries overseas. The results obtained lead us to reject the MRH for the\nglobal dataset, as well as for subsets restricted to the two main\nregions/countries of destination of Japanese foreign investment. The results\nare robust to the weighting of the links, with different specifications, and\nare observed in most industrial sectors. The global Japanese network became\nincreasingly complex during the late 20th century as a consequence of increase\nin the number of Japanese subsidiaries overseas but the key features of the\nstructure remained rather stable. We draw implications of these findings for\nacademic research in international business and for professionals involved in\ncorporate strategy.\n",
"title": "Network analysis of Japanese global business using quasi-exhaustive micro-data for Japanese overseas subsidiaries"
}
| null | null | null | null | true | null |
11115
| null |
Default
| null | null |
null |
{
"abstract": " Online Social Networks (OSNs) have become one of the most important\nactivities on the Internet, such as Facebook and Google+. However, security and\nprivacy have become major concerns in existing C/S based OSNs. In this paper,\nwe propose a novel scheme called a Privacy-preserving Community-based P2P OSNs\nUsing Broadcast Encryption Supporting Recommendation Mechanism (PCBE) that\nsupports cross-platform availability in stringent privacy requirements. For the\nfirst time, we introduce recommendation mechanism into a privacy-preserving P2P\nbased OSNs, in which we firstly employ the Open Directory Project to generate\nuser interest model. We firstly introduce broadcast encryption into P2P\ncommunity-based social networks together with reputation mechanism to decrease\nthe system overhead. We formulate the security requirements and design goals\nfor privacy- preserving P2P based OSNs supporting recommendation mechanism. The\nRESTful web-services help to ensure cross-platform availability and\ntransmission security. As a result, thorough security analysis and performance\nevaluation on experiments demonstrate that the PCBE scheme indeed accords with\nour proposed design goals.\n",
"title": "A Privacy-preserving Community-based P2P OSNs Using Broadcast Encryption Supporting Recommendation Mechanism"
}
| null | null | null | null | true | null |
11116
| null |
Default
| null | null |
null |
{
"abstract": " The stochastic variance-reduced gradient method (SVRG) and its accelerated\nvariant (Katyusha) have attracted enormous attention in the machine learning\ncommunity in the last few years due to their superior theoretical properties\nand empirical behaviour on training supervised machine learning models via the\nempirical risk minimization paradigm. A key structural element in both of these\nmethods is the inclusion of an outer loop at the beginning of which a full pass\nover the training data is made in order to compute the exact gradient, which is\nthen used to construct a variance-reduced estimator of the gradient. In this\nwork we design {\\em loopless variants} of both of these methods. In particular,\nwe remove the outer loop and replace its function by a coin flip performed in\neach iteration designed to trigger, with a small probability, the computation\nof the gradient. We prove that the new methods enjoy the same superior\ntheoretical convergence properties as the original methods. However, we\ndemonstrate through numerical experiments that our methods have substantially\nsuperior practical behavior.\n",
"title": "Don't Jump Through Hoops and Remove Those Loops: SVRG and Katyusha are Better Without the Outer Loop"
}
| null | null | null | null | true | null |
11117
| null |
Default
| null | null |
null |
{
"abstract": " Recently, studies on deep Reservoir Computing (RC) highlighted the role of\nlayering in deep recurrent neural networks (RNNs). In this paper, the use of\nlinear recurrent units allows us to bring more evidence on the intrinsic\nhierarchical temporal representation in deep RNNs through frequency analysis\napplied to the state signals. The potentiality of our approach is assessed on\nthe class of Multiple Superimposed Oscillator tasks. Furthermore, our\ninvestigation provides useful insights to open a discussion on the main aspects\nthat characterize the deep learning framework in the temporal domain.\n",
"title": "Hierarchical Temporal Representation in Linear Reservoir Computing"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
11118
| null |
Validated
| null | null |
null |
{
"abstract": " Part-and-parcel of the study of \"multiplicative number theory\" is the study\nof the distribution of multiplicative functions in arithmetic progressions.\nAlthough appropriate analogies to the Bombieri-Vingradov Theorem have been\nproved for particular examples of multiplicative functions, there has not\npreviously been headway on a general theory; seemingly none of the different\nproofs of the Bombieri-Vingradov Theorem for primes adapt well to this\nsituation. In this article we find out why such a result has been so elusive,\nand discover what can be proved along these lines and develop some limitations.\nFor a fixed residue class $a$ we extend such averages out to moduli $\\leq\nx^{\\frac {20}{39}-\\delta}$.\n",
"title": "Bombieri-Vinogradov for multiplicative functions, and beyond the $x^{1/2}$-barrier"
}
| null | null | null | null | true | null |
11119
| null |
Default
| null | null |
null |
{
"abstract": " Topological Dirac and Weyl semimetals not only host quasiparticles analogous\nto the elementary fermionic particles in high-energy physics, but also have\nnontrivial band topology manifested by exotic Fermi arcs on the surface. Recent\nadvances suggest new types of topological semimetals, in which spatial\nsymmetries protect gapless electronic excitations without high-energy analogy.\nHere we observe triply-degenerate nodal points (TPs) near the Fermi level of\nWC, in which the low-energy quasiparticles are described as three-component\nfermions distinct from Dirac and Weyl fermions. We further observe the surface\nstates whose constant energy contours are pairs of Fermi arcs connecting the\nsurface projection of the TPs, proving the nontrivial topology of the newly\nidentified semimetal state.\n",
"title": "Three-component fermions with surface Fermi arcs in topological semimetal tungsten carbide"
}
| null | null | null | null | true | null |
11120
| null |
Default
| null | null |
null |
{
"abstract": " This work considers the inclusion detection problem of electrical impedance\ntomography with stochastic conductivities. It is shown that a conductivity\nanomaly with a random conductivity can be identified by applying the\nFactorization Method or the Monotonicity Method to the mean value of the\ncorresponding Neumann-to-Dirichlet map provided that the anomaly has high\nenough contrast in the sense of expectation. The theoretical results are\ncomplemented by numerical examples in two spatial dimensions.\n",
"title": "Detecting stochastic inclusions in electrical impedance tomography"
}
| null | null | null | null | true | null |
11121
| null |
Default
| null | null |
null |
{
"abstract": " Given a Hermitian manifold $(M^n,g)$, the Gauduchon connections are the one\nparameter family of Hermitian connections joining the Chern connection and the\nBismut connection. We will call $\\nabla^s = (1-\\frac{s}{2})\\nabla^c +\n\\frac{s}{2}\\nabla^b$ the $s$-Gauduchon connection of $M$, where $\\nabla^c$ and\n$\\nabla^b$ are respectively the Chern and Bismut connections. It is natural to\nask when a compact Hermitian manifold could admit a flat $s$-Gauduchon\nconnection. This is related to a question asked by Yau \\cite{Yau}. The cases\nwith $s=0$ (a flat Chern connection) or $s=2$ (a flat Bismut connection) are\nclassified respectively by Boothby \\cite{Boothby} in the 1950s or by Q. Wang\nand the authors recently \\cite{WYZ}. In this article, we observe that if either\n$s\\geq 4+2\\sqrt{3} \\approx 7.46$ or $s\\leq 4-2\\sqrt{3}\\approx 0.54$ and $s\\neq\n0$, then $g$ is Kähler. We also show that, when $n=2$, $g$ is always Kähler\nunless $s=2$. Note that non-Kähler compact Bismut flat surfaces are exactly\nthose isosceles Hopf surfaces by \\cite{WYZ}.\n",
"title": "On compact Hermitian manifolds with flat Gauduchon connections"
}
| null | null | null | null | true | null |
11122
| null |
Default
| null | null |
null |
{
"abstract": " There has been growing interest in extending traditional vector-based machine\nlearning techniques to their tensor forms. An example is the support tensor\nmachine (STM) that utilizes a rank-one tensor to capture the data structure,\nthereby alleviating the overfitting and curse of dimensionality problems in the\nconventional support vector machine (SVM). However, the expressive power of a\nrank-one tensor is restrictive for many real-world data. To overcome this\nlimitation, we introduce a support tensor train machine (STTM) by replacing the\nrank-one tensor in an STM with a tensor train. Experiments validate and confirm\nthe superiority of an STTM over the SVM and STM.\n",
"title": "A Support Tensor Train Machine"
}
| null | null | null | null | true | null |
11123
| null |
Default
| null | null |
null |
{
"abstract": " We investigate finite-size effects on diffusion in confined fluids using\nmolecular dynamics simulations and hydrodynamic calculations. Specifically, we\nconsider a Lennard-Jones fluid in slit pores without slip at the interface and\nshow that the use of periodic boundary conditions in the directions along the\nsurfaces results in dramatic finite-size effects, in addition to that of the\nphysically relevant confining length. As in the simulation of bulk fluids,\nthese effects arise from spurious hydrodynamic interactions between periodic\nimages and from the constraint of total momentum conservation. We derive\nanalytical expressions for the correction to the diffusion coefficient in the\nlimits of both elongated and flat systems, which are in excellent agreement\nwith the molecular simulation results except for the narrowest pores, where the\ndiscreteness of the fluid particles starts to play a role. The present work\nimplies that the diffusion coefficients for wide nanopores computed using\nelongated boxes suffer from finite-size artifacts which had not been previously\nappreciated. In addition, our analytical expression provides the correction to\nbe applied to the simulation results for finite (possibly small) systems. It\napplies not only to molecular but also to all mesoscopic hydrodynamic\nsimulations, including Lattice-Boltzmann, Multi-Particle Collision Dynamics or\nDissipative Particle Dynamics, which are often used to investigate confined\nsoft matter involving colloidal particles and polymers.\n",
"title": "Diffusion under confinement: hydrodynamic finite-size effects in simulation"
}
| null | null | null | null | true | null |
11124
| null |
Default
| null | null |
null |
{
"abstract": " It is known that Boosting can be interpreted as a gradient descent technique\nto minimize an underlying loss function. Specifically, the underlying loss\nbeing minimized by the traditional AdaBoost is the exponential loss, which is\nproved to be very sensitive to random noise/outliers. Therefore, several\nBoosting algorithms, e.g., LogitBoost and SavageBoost, have been proposed to\nimprove the robustness of AdaBoost by replacing the exponential loss with some\ndesigned robust loss functions. In this work, we present a new way to robustify\nAdaBoost, i.e., incorporating the robust learning idea of Self-paced Learning\n(SPL) into Boosting framework. Specifically, we design a new robust Boosting\nalgorithm based on SPL regime, i.e., SPLBoost, which can be easily implemented\nby slightly modifying off-the-shelf Boosting packages. Extensive experiments\nand a theoretical characterization are also carried out to illustrate the\nmerits of the proposed SPLBoost.\n",
"title": "SPLBoost: An Improved Robust Boosting Algorithm Based on Self-paced Learning"
}
| null | null | null | null | true | null |
11125
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the resistive switching behaviour of $\\mathrm{VO_2}$\nmicrobridges under current bias as a function of temperature and thermal\ncoupling with the heat bath. Upon increasing the electrical current bias, the\nformation of the metallic phase can progress smoothly or through sharp jumps.\nThe magnitude and threshold current values of these sharp resistance drops show\nrandom behaviour and are dramatically influenced by thermal dissipation\nconditions. Our results also evidence how the propagation of the metallic phase\ninduced by electrical current in $\\mathrm{VO_2}$, and thus the shape of the\nresulting high-conductivity path, are not predictable. We discuss the origin of\nthe switching events through a simple electro-thermal model based on the domain\nstructure of $\\mathrm{VO_2}$ films that can be useful to improve the stability\nand controllability of future $\\mathrm{VO_2}$-based devices.\n",
"title": "Influence of thermal boundary conditions on the current-driven resistive transition in $\\mathbf{VO_2}$ microbridges"
}
| null | null | null | null | true | null |
11126
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a convolutional neural network (CNN)-based method that\nlearns traffic as images and predicts large-scale, network-wide traffic speed\nwith a high accuracy. Spatiotemporal traffic dynamics are converted to images\ndescribing the time and space relations of traffic flow via a two-dimensional\ntime-space matrix. A CNN is applied to the image following two consecutive\nsteps: abstract traffic feature extraction and network-wide traffic speed\nprediction. The effectiveness of the proposed method is evaluated by taking two\nreal-world transportation networks, the second ring road and north-east\ntransportation network in Beijing, as examples, and comparing the method with\nfour prevailing algorithms, namely, ordinary least squares, k-nearest\nneighbors, artificial neural network, and random forest, and three deep\nlearning architectures, namely, stacked autoencoder, recurrent neural network,\nand long-short-term memory network. The results show that the proposed method\noutperforms other algorithms by an average accuracy improvement of 42.91%\nwithin an acceptable execution time. The CNN can train the model in a\nreasonable time and, thus, is suitable for large-scale transportation networks.\n",
"title": "Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction"
}
| null | null | null | null | true | null |
11127
| null |
Default
| null | null |
null |
{
"abstract": " Quantum computing is moving rapidly to the point of deployment of technology.\nFunctional quantum devices will require the ability to correct error in order\nto be scalable and effective. A leading choice of error correction, in\nparticular for modular or distributed architectures, is the surface code with\nlogical two-qubit operations realised via \"lattice surgery\". These operations\nconsist of \"merges\" and \"splits\" acting non-unitarily on the logical states and\nare not easily captured by standard circuit notation. This raises the question\nof how best to reason about lattice surgery in order efficiently to use quantum\nstates and operations in architectures with complex resource management issues.\nIn this paper we demonstrate that the operations of the ZX calculus, a form of\nquantum diagrammatic reasoning designed using category theory, match exactly\nthe operations of lattice surgery. Red and green \"spider\" nodes match rough and\nsmooth merges and splits, and follow the axioms of a dagger special associative\nFrobenius algebra. Some lattice surgery operations can require non-trivial\ncorrection operations, which are captured natively in the use of the ZX\ncalculus in the form of ensembles of diagrams. We give a first taste of the\npower of the calculus as a language for surgery by considering two operations\n(magic state use and producing a CNOT) and show how ZX diagram re-write rules\ngive lattice surgery procedures for these operations that are novel, efficient,\nand highly configurable.\n",
"title": "The ZX calculus is a language for surface code lattice surgery"
}
| null | null | null | null | true | null |
11128
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of learning a one-hidden-layer neural network: we\nassume the input $x\\in \\mathbb{R}^d$ is from Gaussian distribution and the\nlabel $y = a^\\top \\sigma(Bx) + \\xi$, where $a$ is a nonnegative vector in\n$\\mathbb{R}^m$ with $m\\le d$, $B\\in \\mathbb{R}^{m\\times d}$ is a full-rank\nweight matrix, and $\\xi$ is a noise vector. We first give an analytic formula\nfor the population risk of the standard squared loss and demonstrate that it\nimplicitly attempts to decompose a sequence of low-rank tensors simultaneously.\nInspired by the formula, we design a non-convex objective function $G(\\cdot)$\nwhose landscape is guaranteed to have the following properties: 1. All local\nminima of $G$ are also global minima.\n2. All global minima of $G$ correspond to the ground truth parameters.\n3. The value and gradient of $G$ can be estimated using samples.\nWith these properties, stochastic gradient descent on $G$ provably converges\nto the global minimum and learn the ground-truth parameters. We also prove\nfinite sample complexity result and validate the results by simulations.\n",
"title": "Learning One-hidden-layer Neural Networks with Landscape Design"
}
| null | null | null | null | true | null |
11129
| null |
Default
| null | null |
null |
{
"abstract": " The direct growth of graphene on semiconducting or insulating substrates\nmight help to overcome main drawbacks of metal-based synthesis, like metal-atom\ncontaminations of graphene, transfer issues, etc. Here we present the growth of\ngraphene on n-doped semiconducting Ge(110) by using an atomic carbon source and\nthe study of the structural and electronic properties of the obtained\ninterface. We found that graphene interacts weakly with the underlying Ge(110)\nsubstrate that keeps graphene's electronic structure almost intact promoting\nthis interface for future graphene-semiconductor applications. The effect of\ndopants in Ge on the electronic properties of graphene is also discussed.\n",
"title": "Growth and electronic structure of graphene on semiconducting Ge(110)"
}
| null | null | null | null | true | null |
11130
| null |
Default
| null | null |
null |
{
"abstract": " We study non-conservative like SODEs admitting explicit Lagrangian\ndescriptions. Such systems are equivalent to the system of Lagrange equations\nof some Lagrangian $L$, including a covariant force field which represents\nnon-conservative forces. We find necessary and sufficient conditions for the\nexistence of a differentiable function $\\Phi:\\mathbb{R}\\rightarrow\\mathbb{R}$\nsuch that the initial system is equivalent to the system of Euler-Lagrange\nequations of the deformed Lagrangian $\\Phi(L)$. We give various examples of\nsuch deformations.\n",
"title": "Alternative Lagrangians obtained by scalar deformations"
}
| null | null | null | null | true | null |
11131
| null |
Default
| null | null |
null |
{
"abstract": " Constrained Markov Decision Process (CMDP) is a natural framework for\nreinforcement learning tasks with safety constraints, where agents learn a\npolicy that maximizes the long-term reward while satisfying the constraints on\nthe long-term cost. A canonical approach for solving CMDPs is the primal-dual\nmethod which updates parameters in primal and dual spaces in turn. Existing\nmethods for CMDPs only use on-policy data for dual updates, which results in\nsample inefficiency and slow convergence. In this paper, we propose a policy\nsearch method for CMDPs called Accelerated Primal-Dual Optimization (APDO),\nwhich incorporates an off-policy trained dual variable in the dual update\nprocedure while updating the policy in primal space with on-policy likelihood\nratio gradient. Experimental results on a simulated robot locomotion task show\nthat APDO achieves better sample efficiency and faster convergence than\nstate-of-the-art approaches for CMDPs.\n",
"title": "Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning"
}
| null | null |
[
"Statistics"
] | null | true | null |
11132
| null |
Validated
| null | null |
null |
{
"abstract": " The role of scalable high-performance workflows and flexible workflow\nmanagement systems that can support multiple simulations will continue to\nincrease in importance. For example, with the end of Dennard scaling, there is\na need to substitute a single long running simulation with multiple repeats of\nshorter simulations, or concurrent replicas. Further, many scientific problems\ninvolve ensembles of simulations in order to solve a higher-level problem or\nproduce statistically meaningful results. However most supercomputing software\ndevelopment and performance enhancements have focused on optimizing single-\nsimulation performance. On the other hand, there is a strong inconsistency in\nthe definition and practice of workflows and workflow management systems. This\ninconsistency often centers around the difference between several different\ntypes of workflows, including modeling and simulation, grid, uncertainty\nquantification, and purely conceptual workflows. This work explores this\nphenomenon by examining the different types of workflows and workflow\nmanagement systems, reviewing the perspective of a large supercomputing\nfacility, examining the common features and problems of workflow management\nsystems, and finally presenting a proposed solution based on the concept of\ncommon building blocks. The implications of the continuing proliferation of\nworkflow management systems and the lack of interoperability between these\nsystems are discussed from a practical perspective. In doing so, we have begun\nan investigation of the design and implementation of open workflow systems for\nsupercomputers based upon common components.\n",
"title": "Toward Common Components for Open Workflow Systems"
}
| null | null | null | null | true | null |
11133
| null |
Default
| null | null |
null |
{
"abstract": " The standard Kernel Quadrature method for numerical integration with random\npoint sets (also called Bayesian Monte Carlo) is known to converge in root mean\nsquare error at a rate determined by the ratio $s/d$, where $s$ and $d$ encode\nthe smoothness and dimension of the integrand. However, an empirical\ninvestigation reveals that the rate constant $C$ is highly sensitive to the\ndistribution of the random points. In contrast to standard Monte Carlo\nintegration, for which optimal importance sampling is well-understood, the\nsampling distribution that minimises $C$ for Kernel Quadrature does not admit a\nclosed form. This paper argues that the practical choice of sampling\ndistribution is an important open problem. One solution is considered; a novel\nautomatic approach based on adaptive tempering and sequential Monte Carlo.\nEmpirical results demonstrate a dramatic reduction in integration error of up\nto 4 orders of magnitude can be achieved with the proposed method.\n",
"title": "On the Sampling Problem for Kernel Quadrature"
}
| null | null | null | null | true | null |
11134
| null |
Default
| null | null |
null |
{
"abstract": " A quasi-relativistic two-component approach for an efficient calculation of\n$\\mathcal{P,T}$-odd interactions caused by a permanent electric dipole moment\nof the electron (eEDM) is presented. The approach uses a (two-component)\ncomplex generalized Hartree-Fock (cGHF) and a complex generalized Kohn-Sham\n(cGKS) scheme within the zeroth order regular approximation (ZORA). In\napplications to select heavy-elemental polar diatomic molecular radicals, which\nare promising candidates for an eEDM experiment, the method is compared to\nrelativistic four-component electron-correlation calculations and confirms\nvalues for the effective electrical field acting on the unpaired electron for\nRaF, BaF, YbF and HgF. The calculations show that purely relativistic effects,\ninvolving only the lower component of the Dirac bi-spinor, are well described\nby treating only the upper component explicitly.\n",
"title": "Zeroth order regular approximation approach to electric dipole moment interactions of the electron"
}
| null | null |
[
"Physics"
] | null | true | null |
11135
| null |
Validated
| null | null |
null |
{
"abstract": " An important preprocessing step in most data analysis pipelines aims to\nextract a small set of sources that explain most of the data. Currently used\nalgorithms for blind source separation (BSS), however, often fail to extract\nthe desired sources and need extensive cross-validation. In contrast, their\nrarely used probabilistic counterparts can get away with little\ncross-validation and are more accurate and reliable but no simple and scalable\nimplementations are available. Here we present a novel probabilistic BSS\nframework (DECOMPOSE) that can be flexibly adjusted to the data, is extensible\nand easy to use, adapts to individual sources and handles large-scale data\nthrough algorithmic efficiency. DECOMPOSE encompasses and generalises many\ntraditional BSS algorithms such as PCA, ICA and NMF and we demonstrate\nsubstantial improvements in accuracy and robustness on artificial and real\ndata.\n",
"title": "Trace your sources in large-scale data: one ring to find them all"
}
| null | null |
[
"Statistics"
] | null | true | null |
11136
| null |
Validated
| null | null |
null |
{
"abstract": " We consider a system of N particles interacting via a short-range smooth\npotential, in a intermediate regime between the weak-coupling and the\nlow-density. We provide a rigorous derivation of the Linear Landau equation\nfrom this particle system. The strategy of the proof consists in showing the\nasymptotic equivalence between the one-particle marginal and the solution of\nthe linear Boltzmann equation with vanishing mean free path.Then, following the\nideas of Landau, we prove the asympotic equivalence between the solutions of\nthe Boltzmann and Landau linear equation in the grazing collision limit.\n",
"title": "The rigorous derivation of the linear Landau equation from a particle system in a weak-coupling limit"
}
| null | null | null | null | true | null |
11137
| null |
Default
| null | null |
null |
{
"abstract": " We study the complexity of geometric problems on spaces of low fractal\ndimension. It was recently shown by [Sidiropoulos & Sridhar, SoCG 2017] that\nseveral problems admit improved solutions when the input is a pointset in\nEuclidean space with fractal dimension smaller than the ambient dimension. In\nthis paper we prove nearly-matching lower bounds, thus establishing\nnearly-optimal bounds for various problems as a function of the fractal\ndimension.\nMore specifically, we show that for any set of $n$ points in $d$-dimensional\nEuclidean space, of fractal dimension $\\delta\\in (1,d)$, for any $\\epsilon >0$\nand $c\\geq 1$, any $c$-spanner must have treewidth at least $\\Omega \\left(\n\\frac{n^{1-1/(\\delta - \\epsilon)}}{c^{d-1}} \\right)$, matching the previous\nupper bound. The construction used to prove this lower bound on the treewidth\nof spanners can also be used to derive lower bounds on the running time of\nalgorithms for various problems, assuming the Exponential Time Hypothesis. We\nprovide two prototypical results of this type. For any $\\delta \\in (1,d)$ and\nany $\\epsilon >0$ we show that:\n1) $d$-dimensional Euclidean TSP on $n$ points with fractal dimension at most\n$\\delta$ cannot be solved in time $2^{O\\left(n^{1-1/(\\delta - \\epsilon)}\n\\right)}$. The best-known upper bound is $2^{O(n^{1-1/\\delta} \\log n)}$.\n2) The problem of finding $k$-pairwise non-intersecting $d$-dimensional unit\nballs/axis parallel unit cubes with centers having fractal dimension at most\n$\\delta$ cannot be solved in time $f(k)n^{O \\left(k^{1-1/(\\delta -\n\\epsilon)}\\right)}$ for any computable function $f$. The best-known upper bound\nis $n^{O(k^{1-1/\\delta} \\log n)}$.\nThe above results nearly match previously known upper bounds from\n[Sidiropoulos & Sridhar, SoCG 2017], and generalize analogous lower bounds for\nthe case of ambient dimension due to [Marx & Sidiropoulos, SoCG 2014].\n",
"title": "Fractal dimension and lower bounds for geometric problems"
}
| null | null | null | null | true | null |
11138
| null |
Default
| null | null |
null |
{
"abstract": " We present a unified framework to analyze the global convergence of Langevin\ndynamics based algorithms for nonconvex finite-sum optimization with $n$\ncomponent functions. At the core of our analysis is a direct analysis of the\nergodicity of the numerical approximations to Langevin dynamics, which leads to\nfaster convergence rates. Specifically, we show that gradient Langevin dynamics\n(GLD) and stochastic gradient Langevin dynamics (SGLD) converge to the almost\nminimizer within $\\tilde O\\big(nd/(\\lambda\\epsilon) \\big)$ and $\\tilde\nO\\big(d^7/(\\lambda^5\\epsilon^5) \\big)$ stochastic gradient evaluations\nrespectively, where $d$ is the problem dimension, and $\\lambda$ is the spectral\ngap of the Markov chain generated by GLD. Both of the results improve upon the\nbest known gradient complexity results. Furthermore, for the first time we\nprove the global convergence guarantee for variance reduced stochastic gradient\nLangevin dynamics (VR-SGLD) to the almost minimizer after $\\tilde\nO\\big(\\sqrt{n}d^5/(\\lambda^4\\epsilon^{5/2})\\big)$ stochastic gradient\nevaluations, which outperforms the gradient complexities of GLD and SGLD in a\nwide regime. Our theoretical analyses shed some light on using Langevin\ndynamics based algorithms for nonconvex optimization with provable guarantees.\n",
"title": "Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization"
}
| null | null | null | null | true | null |
11139
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we propose an eigenvalue analysis -- of system dynamics models\n-- based on the Mutual Information measure, which in turn will be estimated via\nthe Kernel Density Estimation method. We postulate that the proposed approach\nrepresents a novel and efficient multivariate eigenvalue sensitivity analysis.\n",
"title": "Eigenvalue Analysis via Kernel Density Estimation"
}
| null | null | null | null | true | null |
11140
| null |
Default
| null | null |
null |
{
"abstract": " In this article, we present a brief narration of the origin and the overview\nof the recent developments done on the Kolkata Paise Restaurant (KPR) problem,\nwhich can serve as a prototype for a broader class of resource allocation\nproblems in the presence of a large number of competing agents, typically\nstudied using coordination and anti-coordination games. We discuss the KPR and\nits several extensions, as well as its applications in many economic and social\nphenomena. We end the article with some discussions on our ongoing experimental\nanalysis of the same problem. We demonstrate that this provides an interesting\npicture of how people analyze complex situations, and design their strategies\nor react to them.\n",
"title": "The Saga of KPR: Theoretical and Experimental developments"
}
| null | null | null | null | true | null |
11141
| null |
Default
| null | null |
null |
{
"abstract": " We prove that integer programming with three quantifier alternations is\n$NP$-complete, even for a fixed number of variables. This complements earlier\nresults by Lenstra and Kannan, which together say that integer programming with\nat most two quantifier alternations can be done in polynomial time for a fixed\nnumber of variables. As a byproduct of the proof, we show that for two\npolytopes $P,Q \\subset \\mathbb{R}^4$ , counting the projection of integer\npoints in $Q \\backslash P$ is $\\#P$-complete. This contrasts the 2003 result by\nBarvinok and Woods, which allows counting in polynomial time the projection of\ninteger points in $P$ and $Q$ separately.\n",
"title": "The computational complexity of integer programming with alternations"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
11142
| null |
Validated
| null | null |
null |
{
"abstract": " Dictionary learning and component analysis are part of one of the most\nwell-studied and active research fields, at the intersection of signal and\nimage processing, computer vision, and statistical machine learning. In\ndictionary learning, the current methods of choice are arguably K-SVD and its\nvariants, which learn a dictionary (i.e., a decomposition) for sparse coding\nvia Singular Value Decomposition. In robust component analysis, leading methods\nderive from Principal Component Pursuit (PCP), which recovers a low-rank matrix\nfrom sparse corruptions of unknown magnitude and support. However, K-SVD is\nsensitive to the presence of noise and outliers in the training set.\nAdditionally, PCP does not provide a dictionary that respects the structure of\nthe data (e.g., images), and requires expensive SVD computations when solved by\nconvex relaxation. In this paper, we introduce a new robust decomposition of\nimages by combining ideas from sparse dictionary learning and PCP. We propose a\nnovel Kronecker-decomposable component analysis which is robust to gross\ncorruption, can be used for low-rank modeling, and leverages separability to\nsolve significantly smaller problems. We design an efficient learning algorithm\nby drawing links with a restricted form of tensor factorization. The\neffectiveness of the proposed approach is demonstrated on real-world\napplications, namely background subtraction and image denoising, by performing\na thorough comparison with the current state of the art.\n",
"title": "Robust Kronecker-Decomposable Component Analysis for Low-Rank Modeling"
}
| null | null | null | null | true | null |
11143
| null |
Default
| null | null |
null |
{
"abstract": " Despite the wide use of machine learning in adversarial settings including\ncomputer security, recent studies have demonstrated vulnerabilities to evasion\nattacks---carefully crafted adversarial samples that closely resemble\nlegitimate instances, but cause misclassification. In this paper, we examine\nthe adequacy of the leading approach to generating adversarial samples---the\ngradient descent approach. In particular (1) we perform extensive experiments\non three datasets, MNIST, USPS and Spambase, in order to analyse the\neffectiveness of the gradient-descent method against non-linear support vector\nmachines, and conclude that carefully reduced kernel smoothness can\nsignificantly increase robustness to the attack; (2) we demonstrate that\nseparated inter-class support vectors lead to more secure models, and propose a\nquantity similar to margin that can efficiently predict potential\nsusceptibility to gradient-descent attacks, before the attack is launched; and\n(3) we design a new adversarial sample construction algorithm based on\noptimising the multiplicative ratio of class decision functions.\n",
"title": "Adequacy of the Gradient-Descent Method for Classifier Evasion Attacks"
}
| null | null | null | null | true | null |
11144
| null |
Default
| null | null |
null |
{
"abstract": " Communities are ubiquitous in nature and society. Individuals that share\ncommon properties often self-organize to form communities. Avoiding the\nshortages of computation complexity, pre-given information and unstable results\nin different run, in this paper, we propose a simple and effcient method to\ndeepen our understanding of the emergence and diversity of communities in\ncomplex systems. By introducing the rational random selection, our method\nreveals the hidden deterministic and normal diverse community states of\ncommunity structure. To demonstrate this method, we test it with real-world\nsystems. The results show that our method could not only detect community\nstructure with high sensitivity and reliability, but also provide instructional\ninformation about the hidden deterministic community world and our normal\ndiverse community world by giving out the core-community, the real-community,\nthe tide and the diversity. This is of paramount importance in understanding,\npredicting, and controlling a variety of collective behaviors in complex\nsystems.\n",
"title": "The way to uncover community structure with core and diversity"
}
| null | null | null | null | true | null |
11145
| null |
Default
| null | null |
null |
{
"abstract": " Let $ X_{\\lambda_1},\\ldots,X_{\\lambda_n}$ be dependent non-negative random\nvariables and $Y_i=I_{p_i} X_{\\lambda_i}$, $i=1,\\ldots,n$, where\n$I_{p_1},\\ldots,I_{p_n}$ are independent Bernoulli random variables independent\nof $X_{\\lambda_i}$'s, with ${\\rm E}[I_{p_i}]=p_i$, $i=1,\\ldots,n$. In actuarial\nsciences, $Y_i$ corresponds to the claim amount in a portfolio of risks. In\nthis paper, we compare the largest claim amounts of two sets of interdependent\nportfolios, in the sense of usual stochastic order, when the variables in one\nset have the parameters $\\lambda_1,\\ldots,\\lambda_n$ and $p_1,\\ldots,p_n$ and\nthe variables in the other set have the parameters\n$\\lambda^{*}_1,\\ldots,\\lambda^{*}_n$ and $p^*_1,\\ldots,p^*_n$. For\nillustration, we apply the results to some important models in actuary.\n",
"title": "Stochastic comparisons of the largest claim amounts from two sets of interdependent heterogeneous portfolios"
}
| null | null |
[
"Quantitative Finance"
] | null | true | null |
11146
| null |
Validated
| null | null |
null |
{
"abstract": " Term-resolution provides an elegant mechanism to prove that a quantified\nBoolean formula (QBF) is true. It is a dual to Q-resolution (also referred to\nas clause-resolution) and is practically highly important as it enables\ncertifying answers of DPLL-based QBF solvers. While term-resolution and\nQ-resolution are very similar, they're not completely symmetric. In particular,\nQ-resolution operates on clauses and term-resolution operates on models of the\nmatrix. This paper investigates what impact this asymmetry has. We'll see that\nthere is a large class of formulas (formulas with \"big models\") whose\nterm-resolution proofs are exponential. As a possible remedy, the paper\nsuggests to prove true QBFs by refuting their negation ({\\em negate-refute}),\nrather than proving them by term-resolution. The paper shows that from the\ntheoretical perspective this is indeed a favorable approach. In particular,\nnegation-refutation can p-simulates term-resolution and there is an exponential\nseparation between the two calculi. These observations further our\nunderstanding of proof systems for QBFs and provide a strong theoretical\nunderpinning for the effort towards non-CNF QBF solvers.\n",
"title": "An Achilles' Heel of Term-Resolution"
}
| null | null | null | null | true | null |
11147
| null |
Default
| null | null |
null |
{
"abstract": " This article is concerned with quantitative unique continuation estimates for\nequations involving a \"sum of squares\" operator $\\mathcal{L}$ on a compact\nmanifold $\\mathcal{M}$ assuming: $(i)$ the Chow-Rashevski-Hörmander condition\nensuring the hypoellipticity of $\\mathcal{L}$, and $(ii)$ the analyticity of\n$\\mathcal{M}$ and the coefficients of $\\mathcal{L}$.\nThe first result is the tunneling estimate $\\|\\varphi\\|_{L^2(\\omega)} \\geq\nCe^{- \\lambda^{\\frac{k}{2}}}$ for normalized eigenfunctions $\\varphi$ of\n$\\mathcal{L}$ from a nonempty open set $\\omega\\subset \\mathcal{M}$, where $k$\nis the hypoellipticity index of $\\mathcal{L}$ and $\\lambda$ the eigenvalue.\nThe main result is a stability estimate for solutions to the hypoelliptic\nwave equation $(\\partial_t^2+\\mathcal{L})u=0$: for $T>2 \\sup_{x \\in\n\\mathcal{M}}(dist(x,\\omega))$ (here, $dist$ is the sub-Riemannian distance),\nthe observation of the solution on $(0,T)\\times \\omega$ determines the data.\nThe constant involved in the estimate is $Ce^{c\\Lambda^k}$ where $\\Lambda$ is\nthe typical frequency of the data.\nWe then prove the approximate controllability of the hypoelliptic heat\nequation $(\\partial_t+\\mathcal{L})v=1_\\omega f$ in any time, with appropriate\n(exponential) cost, depending on $k$. In case $k=2$ (Grushin, Heisenberg...),\nwe further show approximate controllability to trajectories with polynomial\ncost in large time.\nWe also explain how the analyticity assumption can be relaxed, and a boundary\n$\\partial \\mathcal{M}$ can be added in some situations.\nMost results turn out to be optimal on a family of Grushin-type operators.\nThe main proof relies on the general strategy developed by the authors in\narXiv:1506.04254.\n",
"title": "Tunneling estimates and approximate controllability for hypoelliptic equations"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11148
| null |
Validated
| null | null |
null |
{
"abstract": " To enable electric vehicles (EVs) to access to the internet of intelligent\nvehicles (IoIV), charging EVs wirelessly anytime and anywhere becomes an urgent\nneed. The resonant beam charging (RBC) technology can provide high-power and\nlong-range wireless energy for EVs. However, the RBC system is unefficient. To\nimprove the RBC power transmission efficiency, the adaptive resonant beam\ncharging (ARBC) technology was introduced. In this paper, after analyzing the\nmodular model of the ARBC system, we obtain the closed-form formula of the\nend-to-end power transmission efficiency. Then, we prove that the optimal power\ntransmission efficiency uniquely exists. Moreover, we analyze the relationships\namong the optimal power transmission efficiency, the source power, the output\npower, and the beam transmission efficiency, which provide the guidelines for\nthe optimal ARBC system design and implementation. Hence, perpetual energy can\nbe supplied to EVs in IoIV virtually.\n",
"title": "Optimal Resonant Beam Charging for Electronic Vehicles in Internet of Intelligent Vehicles"
}
| null | null | null | null | true | null |
11149
| null |
Default
| null | null |
null |
{
"abstract": " Betweenness centrality---measuring how many shortest paths pass through a\nvertex---is one of the most important network analysis concepts for assessing\nthe relative importance of a vertex. The well-known algorithm of Brandes [2001]\ncomputes, on an $n$-vertex and $m$-edge graph, the betweenness centrality of\nall vertices in $O(nm)$ worst-case time. In follow-up work, significant\nempirical speedups were achieved by preprocessing degree-one vertices and by\ngraph partitioning based on cut vertices. We further contribute an algorithmic\ntreatment of degree-two vertices, which turns out to be much richer in\nmathematical structure than the case of degree-one vertices. Based on these\nthree algorithmic ingredients, we provide a strengthened worst-case running\ntime analysis for betweenness centrality algorithms. More specifically, we\nprove an adaptive running time bound $O(kn)$, where $k < m$ is the size of a\nminimum feedback edge set of the input graph.\n",
"title": "An Adaptive Version of Brandes' Algorithm for Betweenness Centrality"
}
| null | null |
[
"Computer Science"
] | null | true | null |
11150
| null |
Validated
| null | null |
null |
{
"abstract": " Understanding the nature of the turbulent fluctuations below the ion\ngyroradius in solar-wind turbulence is a great challenge. Recent studies have\nbeen mostly in favor of kinetic Alfvén wave (KAW) type of fluctuations, but\nother kinds of fluctuations with characteristics typical of magnetosonic,\nwhistler and ion Bernstein modes, could also play a role depending on the\nplasma parameters. Here we investigate the properties of the sub-proton-scale\ncascade with high-resolution hybrid-kinetic simulations of freely-decaying\nturbulence in 3D3V phase space, including electron inertia effects. Two proton\nplasma beta are explored: the \"intermediate\" $\\beta_p=1$ and \"low\"\n$\\beta_p=0.2$ regimes, both typically observed in solar wind and corona. The\nmagnetic energy spectum exhibits $k_\\perp^{-8/3}$ and $k_\\|^{-7/2}$ power laws\nat $\\beta_p=1$, while they are slightly steeper at $\\beta_p=0.2$. Nevertheless,\nboth regimes develop a spectral anisotropy consistent with $k_\\|\\sim\nk_\\perp^{2/3}$ at $k_\\perp\\rho_p>1$, and pronounced small-scale intermittency.\nIn this context, we find that the kinetic-scale cascade is dominated by\nKAW-like fluctuations at $\\beta_p=1$, whereas the low-$\\beta$ case presents a\nmore complex scenario suggesting the simultaneous presence of different types\nof fluctuations. In both regimes, however, a non-negligible role of ion\nBernstein type of fluctuations at the smallest scales seems to emerge.\n",
"title": "Kinetic cascade in solar-wind turbulence: 3D3V hybrid-kinetic simulations with electron inertia"
}
| null | null | null | null | true | null |
11151
| null |
Default
| null | null |
null |
{
"abstract": " Neural field theory is used to quantitatively analyze the two-dimensional\nspatiotemporal correlation properties of gamma-band (30 -- 70 Hz) oscillations\nevoked by stimuli arriving at the primary visual cortex (V1), and modulated by\npatchy connectivities that depend on orientation preference (OP). Correlation\nfunctions are derived analytically under different stimulus and measurement\nconditions. The predictions reproduce a range of published experimental\nresults, including the existence of two-point oscillatory temporal\ncross-correlations with zero time-lag between neurons with similar OP, the\ninfluence of spatial separation of neurons on the strength of the correlations,\nand the effects of differing stimulus orientations.\n",
"title": "Gamma-Band Correlations in Primary Visual Cortex"
}
| null | null |
[
"Quantitative Biology"
] | null | true | null |
11152
| null |
Validated
| null | null |
null |
{
"abstract": " We address the problem of camera-to-laser-scanner calibration using a\ncheckerboard and multiple image-laser scan pairs. Distinguishing which laser\npoints measure the checkerboard and which lie on the background is essential to\nany such system. We formulate the checkerboard extraction as a combinatorial\noptimization problem with a clear cut objective function. We propose a\nbranch-and-bound technique that deterministically and globally optimizes the\nobjective. Unlike what is available in the literature, the proposed method is\nnot heuristic and does not require assumptions such as constraints on the\nbackground or relying on discontinuity of the range measurements to partition\nthe data into line segments. The proposed approach is generic and can be\napplied to both 3D or 2D laser scanners as well as the cases where multiple\ncheckerboards are present. We demonstrate the effectiveness of the proposed\napproach by providing numerical simulations as well as experimental results.\n",
"title": "A Branch-and-Bound Algorithm for Checkerboard Extraction in Camera-Laser Calibration"
}
| null | null | null | null | true | null |
11153
| null |
Default
| null | null |
null |
{
"abstract": " The use of volunteers has emerged as low-cost alternative to generate\naccurate geographical information, an approach known as Volunteered Geographic\nInformation (VGI). However, VGI is limited by the number and availability of\nvolunteers in the area to be mapped, hindering scalability for large areas and\nmaking difficult to map within a time-frame. Fortunately, the availability of\nstreet-view imagery enables the virtual exploration of urban environments,\nmaking possible the recruitment of contributors not necessarily located in the\narea to be mapped. In this paper, we describe the design, implementation, and\nevaluation of the Virtual City Explorer (VCE), a system to collect the\ncoordinates of Points of Interest within a bounded area on top of a street view\nservice with the use of paid crowdworkers. Our evaluation suggests that paid\ncrowdworkers are effective for finding PoIs, and cover almost all the area.\nWith respect to completeness, our approach does not find all PoIs found by\nexperts or VGI communities, but is able to find PoIs that were not found by\nthem, suggesting complementarity. We also studied the impact of making PoIs\nalready discovered by a certain number of workers \\emph{taboo} for incoming\nworkers, finding that it encourages more exploration from workers , increase\nthe number of detected PoIs , and reduce costs.\n",
"title": "On the mapping of Points of Interest through StreetView imagery and paid crowdsourcing"
}
| null | null | null | null | true | null |
11154
| null |
Default
| null | null |
null |
{
"abstract": " Objective: The coupling between neuronal populations and its magnitude have\nbeen shown to be informative for various clinical applications. One method to\nestimate brain connectivity is with electroencephalography (EEG) from which the\ncross-spectrum between different sensor locations is derived. We wish to test\nthe efficacy of tensor factorisation in the estimation of brain connectivity.\nMethods: Complex tensor factorisation based on PARAFAC2 is used to decompose\nthe EEG into scalp components described by the spatial, spectral, and complex\ntrial profiles. An EEG model in the complex domain was derived that shows the\nsuitability of PARAFAC2. A connectivity metric was also derived on the complex\ntrial profiles of the extracted components. Results: Results on a benchmark EEG\ndataset confirmed that PARAFAC2 can estimate connectivity better than\ntraditional tensor analysis such as PARAFAC within a range of signal-to-noise\nratios. The analysis of EEG from patients with mild cognitive impairment or\nAlzheimer's disease showed that PARAFAC2 identifies loss of brain connectivity\nbetter than traditional approaches and agreeing with prior pathological\nknowledge. Conclusion: The complex PARAFAC2 algorithm is suitable for EEG\nconnectivity estimation since it allows to extract meaningful coupled sources\nand provides better estimates than complex PARAFAC. Significance: A new\nparadigm that employs complex tensor factorisation has demonstrated to be\nsuccessful in identifying brain connectivity and the location of couples\nsources for both a benchmark and a real-world EEG dataset. This can enable\nfuture applications and has the potential to solve some the issues that\ndeteriorate the performance of traditional connectivity metrics.\n",
"title": "Complex tensor factorisation with PARAFAC2 for the estimation of brain connectivity from the EEG"
}
| null | null | null | null | true | null |
11155
| null |
Default
| null | null |
null |
{
"abstract": " Synchronized measurements of a large power grid enable an unprecedented\nopportunity to study the spatialtemporal correlations. Statistical analytics\nfor those massive datasets start with high-dimensional data matrices.\nUncertainty is ubiquitous in a future's power grid. These data matrices are\nrecognized as random matrices. This new point of view is fundamental in our\ntheoretical analysis since true covariance matrices cannot be estimated\naccurately in a high-dimensional regime. As an alternative, we consider\nlarge-dimensional sample covariance matrices in the asymptotic regime to\nreplace the true covariance matrices. The self-adjoint polynomials of\nlarge-dimensional random matrices are studied as statistics for big data\nanalytics. The calculation of the asymptotic spectrum distribution (ASD) for\nsuch a matrix polynomial is understandably challenging. This task is made\npossible by a recent breakthrough in free probability, an active research\nbranch in random matrix theory. This is the very reason why the work of this\npaper is inspired initially. The new approach is interesting in many aspects.\nThe mathematical reason may be most critical. The real-world problems can be\nsolved using this approach, however.\n",
"title": "A New Approach of Exploiting Self-Adjoint Matrix Polynomials of Large Random Matrices for Anomaly Detection and Fault Location"
}
| null | null | null | null | true | null |
11156
| null |
Default
| null | null |
null |
{
"abstract": " We study pool-based active learning with abstention feedbacks, where a\nlabeler can abstain from labeling a queried example with some unknown\nabstention rate. This is an important problem with many useful applications. We\ntake a Bayesian approach to the problem and develop two new greedy algorithms\nthat learn both the classification problem and the unknown abstention rate at\nthe same time. These are achieved by simply incorporating the estimated\nabstention rate into the greedy criteria. We prove that both of our algorithms\nhave near-optimality guarantees: they respectively achieve a\n${(1-\\frac{1}{e})}$ constant factor approximation of the optimal expected or\nworst-case value of a useful utility function. Our experiments show the\nalgorithms perform well in various practical scenarios.\n",
"title": "Bayesian Pool-based Active Learning With Abstention Feedbacks"
}
| null | null | null | null | true | null |
11157
| null |
Default
| null | null |
null |
{
"abstract": " Many chemical systems cannot be described by quantum chemistry methods based\non a singlereference wave function. Accurate predictions of energetic and\nspectroscopic properties require a delicate balance between describing the most\nimportant configurations (static correlation) and obtaining dynamical\ncorrelation efficiently. The former is most naturally done through a\nmulticonfigurational (MC) wave function, whereas the latter can be done by,\ne.g., perturbation theory. We have employed a different strategy, namely, a\nhybrid between multiconfigurational wave functions and density-functional\ntheory (DFT) based on range separation. The method is denoted by MC short-range\n(sr) DFT and is more efficient than perturbative approaches as it capitalizes\non the efficient treatment of the (short-range) dynamical correlation by DFT\napproximations. In turn, the method also improves DFT with standard\napproximations through the ability of multiconfigurational wave functions to\nrecover large parts of the static correlation. Until now, our implementation\nwas restricted to closed-shell systems, and to lift this restriction, we\npresent here the generalization of MC-srDFT to open-shell cases. The additional\nterms required to treat open-shell systems are derived and implemented in the\nDALTON program. This new method for open-shell systems is illustrated on\ndioxygen and [Fe(H2O)6]3+.\n",
"title": "Multiconfigurational Short-Range Density-Functional Theory for Open-Shell Systems"
}
| null | null | null | null | true | null |
11158
| null |
Default
| null | null |
null |
{
"abstract": " Shape analyses of tephra grains result in understanding eruption mechanism of\nvolcanoes. However, we have to define and select parameter set such as\nconvexity for the precise discrimination of tephra grains. Selection of the\nbest parameter set for the recognition of tephra shapes is complicated.\nActually, many shape parameters have been suggested. Recently, neural network\nhas made a great success in the field of machine learning. Convolutional neural\nnetwork can recognize the shape of images without human bias and shape\nparameters. We applied the simple convolutional neural network developed for\nthe handwritten digits to the recognition of tephra shapes. The network was\ntrained by Morphologi tephra images, and it can recognize the tephra shapes\nwith approximately 90% of accuracy.\n",
"title": "Shape recognition of volcanic ash by simple convolutional neural network"
}
| null | null |
[
"Computer Science",
"Physics"
] | null | true | null |
11159
| null |
Validated
| null | null |
null |
{
"abstract": " The high-performance computing resources and the constant improvement of both\nnumerical simulation accuracy and the experimental measurements with which they\nare confronted, bring a new compulsory step to strengthen the credence given to\nthe simulation results: uncertainty quantification. This can have different\nmeanings, according to the requested goals (rank uncertainty sources, reduce\nthem, estimate precisely a critical threshold or an optimal working point) and\nit could request mathematical methods with greater or lesser complexity. This\npaper introduces the Uranie platform, an Open-source framework which is\ncurrently developed at the Alternative Energies and Atomic Energy Commission\n(CEA), in the nuclear energy division, in order to deal with uncertainty\npropagation, surrogate models, optimisation issues, code calibration... This\nplatform benefits from both its dependencies, but also from personal\ndevelopments, to offer an efficient data handling model, a C++ and Python\ninterpreter, advanced graphical tools, several parallelisation solutions...\nThese methods are very generic and can then be applied to many kinds of code\n(as Uranie considers them as black boxes) so to many fields of physics as well.\nIn this paper, the example of thermal exchange between a plate-sheet and a\nfluid is introduced to show how Uranie can be used to perform a large range of\nanalysis. The code used to produce the figures of this paper can be found in\nthis https URL along with the sources of the\nplatform.\n",
"title": "The Uranie platform: an Open-source software for optimisation, meta-modelling and uncertainty analysis"
}
| null | null | null | null | true | null |
11160
| null |
Default
| null | null |
null |
{
"abstract": " Visual question answering is a recently proposed artificial intelligence task\nthat requires a deep understanding of both images and texts. In deep learning,\nimages are typically modeled through convolutional neural networks, and texts\nare typically modeled through recurrent neural networks. While the requirement\nfor modeling images is similar to traditional computer vision tasks, such as\nobject recognition and image classification, visual question answering raises a\ndifferent need for textual representation as compared to other natural language\nprocessing tasks. In this work, we perform a detailed analysis on natural\nlanguage questions in visual question answering. Based on the analysis, we\npropose to rely on convolutional neural networks for learning textual\nrepresentations. By exploring the various properties of convolutional neural\nnetworks specialized for text data, such as width and depth, we present our\n\"CNN Inception + Gate\" model. We show that our model improves question\nrepresentations and thus the overall accuracy of visual question answering\nmodels. We also show that the text representation requirement in visual\nquestion answering is more complicated and comprehensive than that in\nconventional natural language processing tasks, making it a better task to\nevaluate textual representation methods. Shallow models like fastText, which\ncan obtain comparable results with deep learning models in tasks like text\nclassification, are not suitable in visual question answering.\n",
"title": "Learning Convolutional Text Representations for Visual Question Answering"
}
| null | null | null | null | true | null |
11161
| null |
Default
| null | null |
null |
{
"abstract": " For natural microswimmers, the interplay of swimming activity and external\nflow can promote robust motion, e.g. propulsion against (\"upstream rheotaxis\")\nor perpendicular to the direction of flow. These effects are generally\nattributed to their complex body shapes and flagellar beat patterns. Here,\nusing catalytic Janus particles as a model experimental system, we report on a\nstrong directional response that occurs for spherical active particles in a\nchannel flow. The particles align their propulsion axes to be nearly\nperpendicular to both the direction of flow and the normal vector of a nearby\nbounding surface. We develop a deterministic theoretical model of spherical\nmicroswimmers near a planar wall that captures the experimental observations.\nWe show how the directional response emerges from the interplay of shear flow\nand near-surface swimming activity. Finally, adding the effect of thermal\nnoise, we obtain probability distributions for the swimmer orientation that\nsemi-quantitatively agree with the experimental distributions.\n",
"title": "Cross-stream migration of active particles"
}
| null | null | null | null | true | null |
11162
| null |
Default
| null | null |
null |
{
"abstract": " In this article, we construct a two-block Gibbs sampler using Polson et al.\n(2013) data augmentation technique with Polya-Gamma latent variables for\nBayesian logistic linear mixed models under proper priors. Furthermore, we\nprove the uniform ergodicity of this Gibbs sampler, which guarantees the\nexistence of the central limit theorems for MCMC based estimators.\n",
"title": "Analysis of the Polya-Gamma block Gibbs sampler for Bayesian logistic linear mixed models"
}
| null | null | null | null | true | null |
11163
| null |
Default
| null | null |
null |
{
"abstract": " Understanding excited carrier dynamics in semiconductors is crucial for the\ndevelopment of photovoltaics and efficient photonic devices. However,\noverlapping spectral features in optical/NIR pump-probe spectroscopy often\nrender assignments of separate electron and hole carrier dynamics ambiguous.\nHere, ultrafast electron and hole dynamics in germanium nanocrystalline thin\nfilms are directly and simultaneously observed by attosecond transient\nabsorption spectroscopy (ATAS) in the extreme ultraviolet at the germanium\nM_{4,5}-edge (~30 eV). We decompose the ATAS spectra into contributions of\nelectronic state blocking and photo-induced band shifts at a carrier density of\n8*10^{20}cm^{-3}. Separate electron and hole relaxation times are observed as a\nfunction of hot carrier energies. A first order electron and hole decay of ~1\nps suggests a Shockley-Read-Hall recombination mechanism. The simultaneous\nobservation of electrons and holes with ATAS paves the way for investigating\nfew to sub-femtosecond dynamics of both holes and electrons in complex\nsemiconductor materials and across junctions.\n",
"title": "Direct and Simultaneous Observation of Ultrafast Electron and Hole Dynamics in Germanium"
}
| null | null | null | null | true | null |
11164
| null |
Default
| null | null |
null |
{
"abstract": " Robots have gained relevance in society, increasingly performing critical\ntasks. Nonetheless, robot security is being underestimated. Robotics security\nis a complex landscape, which often requires a cross-disciplinar perspective to\nwhich classical security lags behind. To address this issue, we present the\nRobot Security Framework (RSF), a methodology to perform systematic security\nassessments in robots. We propose, adapt and develop specific terminology and\nprovide guidelines to enable a holistic security assessment following four main\nlayers (Physical, Network, Firmware and Application). We argue that modern\nrobotics should regard as equally relevant internal and external communication\nsecurity. Finally, we advocate against \"security by obscurity\". We conclude\nthat the field of security in robotics deserves further research efforts.\n",
"title": "Introducing the Robot Security Framework (RSF), a standardized methodology to perform security assessments in robotics"
}
| null | null | null | null | true | null |
11165
| null |
Default
| null | null |
null |
{
"abstract": " Rapport plays an important role during communication because it can help\npeople understand each other's feelings or ideas and leads to a smooth\ncommunication. Computational rapport model has been proposed based on theory in\nprevious work. But there lacks solid verification. In this paper, we apply\nstructural equation model (SEM) to the theoretical model on both dyads of\nfriend and stranger. The results indicate some unfavorable paths. Based on the\nresults and more literature, we modify the original model to integrate more\nnonverbal behaviors, including gaze and smile. Fit indices and other\nexamination show the goodness of our new models, which can give us more insight\ninto rapport management during conversation.\n",
"title": "Statistical Verification of Computational Rapport Model"
}
| null | null | null | null | true | null |
11166
| null |
Default
| null | null |
null |
{
"abstract": " The main theorems of this paper are (1) there is no least transitive model of\nKelley--Morse set theory $\\mathsf{KM}$ and (2) there is a least\n$\\beta$-model---that is, a transitive model which is correct about which of its\nclasses are well-founded---of Gödel--Bernays set theory $\\mathsf{GBC}$ +\nElementary Transfinite Recursion. Along the way I characterize when a countable\nmodel of $\\mathsf{ZFC}$ has a least $\\mathsf{GBC}$-realization and show that no\ncountable model of $\\mathsf{ZFC}$ has a least $\\mathsf{KM}$-realization. I also\nshow that fragments of Elementary Transfinite Recursion have least\n$\\beta$-models and, for sufficiently weak fragments, least transitive models.\nThese fragments can be separated from each other and from the full principle of\nElementary Transfinite Recursion by consistency strength. The main question\nleft unanswered by this article is whether there is a least transitive model of\n$\\mathsf{GBC}$ + Elementary Transfinite Recursion.\n",
"title": "Least models of second-order set theories"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11167
| null |
Validated
| null | null |
null |
{
"abstract": " We present near-infrared spectra for 144 candidate planetary systems\nidentified during Campaigns 1-7 of the NASA K2 Mission. The goal of the survey\nwas to characterize planets orbiting low-mass stars, but our IRTF/SpeX and\nPalomar/TripleSpec spectroscopic observations revealed that 49% of our targets\nwere actually giant stars or hotter dwarfs reddened by interstellar extinction.\nFor the 72 stars with spectra consistent with classification as cool dwarfs\n(spectral types K3 - M4), we refined their stellar properties by applying\nempirical relations based on stars with interferometric radius measurements.\nAlthough our revised temperatures are generally consistent with those reported\nin the Ecliptic Plane Input Catalog (EPIC), our revised stellar radii are\ntypically 0.13 solar radii (39%) larger than the EPIC values, which were based\non model isochrones that have been shown to underestimate the radii of cool\ndwarfs. Our improved stellar characterizations will enable more efficient\nprioritization of K2 targets for follow-up studies.\n",
"title": "Characterizing K2 Candidate Planetary Systems Orbiting Low-Mass Stars I: Classifying Low-mass Host Stars Observed During Campaigns 1-7"
}
| null | null | null | null | true | null |
11168
| null |
Default
| null | null |
null |
{
"abstract": " An efficient Bayesian technique for estimation problems in fundamental\nstellar astronomy is tested on simulated data for a binary observed both\nastrometrically and spectroscopically. Posterior distributions are computed for\nthe components' masses and for the binary's parallax. One thousand independent\nrepetitions of the simulation demonstrate that the 1- and 2-$\\!\\sigma$\ncredibility intervals for these fundamental quantities have close to the\ncorrect coverage fractions. In addition, the simulations allow the\ninvestigation of the statistical properties of a Bayesian goodness-of-fit\ncriterion and of the corresponding p-value. The criterion has closely similar\nproperties to the traditional chi^{2} test for minimum-chi^{2} solutions.\n",
"title": "Binary orbits from combined astrometric and spectroscopic data"
}
| null | null | null | null | true | null |
11169
| null |
Default
| null | null |
null |
{
"abstract": " We have investigated tunneling current through a suspended graphene Corbino\ndisk in high magnetic fields at the Dirac point, i.e. at filling factor $\\nu$ =\n0. At the onset of the dielectric breakdown the current through the disk grows\nexponentially before ohmic behaviour, but in a manner distinct from thermal\nactivation. We find that Zener tunneling between Landau sublevels dominates,\nfacilitated by tilting of the source-drain bias potential. According to our\nanalytic modelling, the Zener tunneling is strongly affected by the gyrotropic\nforce (Lorentz force) due to the high magnetic field\n",
"title": "Gyrotropic Zener tunneling and nonlinear IV curves in the zero-energy Landau level of graphene in a strong magnetic field"
}
| null | null | null | null | true | null |
11170
| null |
Default
| null | null |
null |
{
"abstract": " The Graph Convolutional Network (GCN) model and its variants are powerful\ngraph embedding tools for facilitating classification and clustering on graphs.\nHowever, a major challenge is to reduce the complexity of layered GCNs and make\nthem parallelizable and scalable on very large graphs --- state-of the art\ntechniques are unable to achieve scalability without losing accuracy and\nefficiency. In this paper, we propose novel parallelization techniques for\ngraph sampling-based GCNs that achieve superior scalable performance on very\nlarge graphs without compromising accuracy. Specifically, our GCN guarantees\nwork-efficient training and produces order of magnitude savings in computation\nand communication. To scale GCN training on tightly-coupled shared memory\nsystems, we develop parallelization strategies for the key steps in training:\nFor the graph sampling step, we exploit parallelism within and across multiple\nsampling instances, and devise an efficient data structure for concurrent\naccesses that provides theoretical guarantee of near-linear speedup with number\nof processing units. For the feature propagation step within the sampled graph,\nwe improve cache utilization and reduce DRAM communication by data\npartitioning. We prove that our partitioning strategy is a 2-approximation for\nminimizing the communication time compared to the optimal strategy. We\ndemonstrate that our parallel graph embedding outperforms state-of-the-art\nmethods in scalability (with respect to number of processors, graph size and\nGCN model size), efficiency and accuracy on several large datasets. On a\n40-core Xeon platform, our parallel training achieves 64$\\times$ speedup (with\nAVX) in the sampling step and 25$\\times$ speedup in the feature propagation\nstep, compared to the serial implementation, resulting in a net speedup of\n21$\\times$.\n",
"title": "Accurate, Efficient and Scalable Graph Embedding"
}
| null | null | null | null | true | null |
11171
| null |
Default
| null | null |
null |
{
"abstract": " There are many different relatedness measures, based for instance on citation\nrelations or textual similarity, that can be used to cluster scientific\npublications. We propose a principled methodology for evaluating the accuracy\nof clustering solutions obtained using these relatedness measures. We formally\nshow that the proposed methodology has an important consistency property. The\nempirical analyses that we present are based on publications in the fields of\ncell biology, condensed matter physics, and economics. Using the BM25\ntext-based relatedness measure as evaluation criterion, we find that\nbibliographic coupling relations yield more accurate clustering solutions than\ndirect citation relations and co-citation relations. The so-called extended\ndirect citation approach performs similarly to or slightly better than\nbibliographic coupling in terms of the accuracy of the resulting clustering\nsolutions. The other way around, using a citation-based relatedness measure as\nevaluation criterion, BM25 turns out to yield more accurate clustering\nsolutions than other text-based relatedness measures.\n",
"title": "A principled methodology for comparing relatedness measures for clustering publications"
}
| null | null | null | null | true | null |
11172
| null |
Default
| null | null |
null |
{
"abstract": " Nowadays many companies have available large amounts of raw, unstructured\ndata. Among Big Data enabling technologies, a central place is held by the\nMapReduce framework and, in particular, by its open source implementation,\nApache Hadoop. For cost effectiveness considerations, a common approach entails\nsharing server clusters among multiple users. The underlying infrastructure\nshould provide every user with a fair share of computational resources,\nensuring that Service Level Agreements (SLAs) are met and avoiding wastes. In\nthis paper we consider two mathematical programming problems that model the\noptimal allocation of computational resources in a Hadoop 2.x cluster with the\naim to develop new capacity allocation techniques that guarantee better\nperformance in shared data centers. Our goal is to get a substantial reduction\nof power consumption while respecting the deadlines stated in the SLAs and\navoiding penalties associated with job rejections. The core of this approach is\na distributed algorithm for runtime capacity allocation, based on Game Theory\nmodels and techniques, that mimics the MapReduce dynamics by means of\ninteracting players, namely the central Resource Manager and Class Managers.\n",
"title": "A Game-Theoretic Approach for Runtime Capacity Allocation in MapReduce"
}
| null | null | null | null | true | null |
11173
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the butterfly effect and charge diffusion near the quantum\nphase transition in holographic approach. We argue that their criticality is\ncontrolled by the holographic scaling geometry with deformations induced by a\nrelevant operator at finite temperature. Specifically, in the quantum critical\nregion controlled by a single fixed point, the butterfly velocity decreases\nwhen deviating from the critical point. While, in the non-critical region, the\nbehavior of the butterfly velocity depends on the specific phase at low\ntemperature. Moreover, in the holographic Berezinskii-Kosterlitz-Thouless\ntransition, the universal behavior of the butterfly velocity is absent.\nFinally, the tendency of our holographic results matches with the numerical\nresults of Bose-Hubbard model. A comparison between our result and that in the\n$O(N)$ nonlinear sigma model is also given.\n",
"title": "Holographic Butterfly Effect and Diffusion in Quantum Critical Region"
}
| null | null | null | null | true | null |
11174
| null |
Default
| null | null |
null |
{
"abstract": " Although all superconducting cuprates display charge-ordering tendencies,\ntheir low-temperature properties are distinct, impeding efforts to understand\nthe phenomena within a single conceptual framework. While some systems exhibit\nstripes of charge and spin, with a locked periodicity, others host charge\ndensity waves (CDWs) without any obviously related spin order. Here we use\nresonant inelastic x-ray scattering (RIXS) to follow the evolution of charge\ncorrelations in the canonical stripe ordered cuprate\nLa$_{1.875}$Ba$_{0.125}$CuO$_{4}$ (LBCO~$1/8$) across its ordering transition.\nWe find that high-temperature charge correlations are unlocked from the\nwavevector of the spin correlations, signaling analogies to CDW phases in\nvarious other cuprates. This indicates that stripe order at low temperatures is\nstabilized by the coupling of otherwise independent charge and spin density\nwaves, with important implications for the relation between charge and spin\ncorrelations in the cuprates.\n",
"title": "High-temperature charge density wave correlations in La$_{1.875}$Ba$_{0.125}$CuO$_{4}$ without spin-charge locking"
}
| null | null | null | null | true | null |
11175
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we use deep feedforward artificial neural networks to\napproximate solutions to partial differential equations in complex geometries.\nWe show how to modify the backpropagation algorithm to compute the partial\nderivatives of the network output with respect to the space variables which is\nneeded to approximate the differential operator. The method is based on an\nansatz for the solution which requires nothing but feedforward neural networks\nand an unconstrained gradient based optimization method such as gradient\ndescent or a quasi-Newton method.\nWe show an example where classical mesh based methods cannot be used and\nneural networks can be seen as an attractive alternative. Finally, we highlight\nthe benefits of deep compared to shallow neural networks and device some other\nconvergence enhancing techniques.\n",
"title": "A unified deep artificial neural network approach to partial differential equations in complex geometries"
}
| null | null | null | null | true | null |
11176
| null |
Default
| null | null |
null |
{
"abstract": " The purpose of this work is mostly expository and aims to elucidate the\nJordan-Kinderlehrer-Otto (JKO) scheme for uncertainty propagation, and a\nvariant, the Laugesen-Mehta-Meyn-Raginsky (LMMR) scheme for filtering. We point\nout that these variational schemes can be understood as proximal operators in\nthe space of density functions, realizing gradient flows. These schemes hold\nthe promise of leading to efficient ways for solving the Fokker-Planck equation\nas well as the equations of non-linear filtering. Our aim in this paper is to\ndevelop in detail the underlying ideas in the setting of linear stochastic\nsystems with Gaussian noise and recover known results.\n",
"title": "Gradient Flows in Uncertainty Propagation and Filtering of Linear Gaussian Systems"
}
| null | null | null | null | true | null |
11177
| null |
Default
| null | null |
null |
{
"abstract": " Simulation-based image quality metrics are adapted and investigated for\ncharacterizing the parameter dependences of linear iterative image\nreconstruction for DBT. Three metrics based on 2D DBT simulation are\ninvestigated: (1) a root-mean-square-error (RMSE) between the test phantom and\nreconstructed image, (2) a gradient RMSE where the comparison is made after\ntaking a spatial gradient of both image and phantom, and (3) a\nregion-of-interest (ROI) Hotelling observer (HO) for\nsignal-known-exactly/background-known-exactly (SKE/BKE) and\nsignal-known-exactly/background-known-statistically (SKE/BKS) detection tasks.\nTwo simulation studies are performed using the aforementioned metrics, varying\nvoxel aspect ratio and regularization strength for two types of Tikhonov\nregularized least-squares optimization. The RMSE metrics are applied to a 2D\ntest phantom and the ROI-HO metric is applied to two tasks relevant to DBT:\nlarge, low contrast lesion detection and small, high contrast\nmicrocalcification detection. The RMSE metric trends are compared with visual\nassessment of the reconstructed test phantom. The ROI-HO metric trends are\ncompared with 3D reconstructed images from ACR phantom data acquired with a\nHologic Selenia Dimensions DBT system. Sensitivity of image RMSE to mean pixel\nvalue is found to limit its applicability to the assessment of DBT image\nreconstruction. Image gradient RMSE is insensitive to mean pixel value and\nappears to track better with subjective visualization of the reconstructed\nbar-pattern phantom. The ROI-HO metric shows an increasing trend with\nregularization strength for both forms of Tikhonov-regularized least-squares;\nhowever, this metric saturates at intermediate regularization strength\nindicating a point of diminishing returns for signal detection. Visualization\nwith reconstructed ACR phantom images appears to show a similar dependence with\nregularization strength.\n",
"title": "Investigating Simulation-Based Metrics for Characterizing Linear Iterative Reconstruction in Digital Breast Tomosynthesis"
}
| null | null | null | null | true | null |
11178
| null |
Default
| null | null |
null |
{
"abstract": " Temporal-Difference learning (TD) [Sutton, 1988] with function approximation\ncan converge to solutions that are worse than those obtained by Monte-Carlo\nregression, even in the simple case of on-policy evaluation. To increase our\nunderstanding of the problem, we investigate the issue of approximation errors\nin areas of sharp discontinuities of the value function being further\npropagated by bootstrap updates. We show empirical evidence of this leakage\npropagation, and show analytically that it must occur, in a simple Markov\nchain, when function approximation errors are present. For reversible policies,\nthe result can be interpreted as the tension between two terms of the loss\nfunction that TD minimises, as recently described by [Ollivier, 2018]. We show\nthat the upper bounds from [Tsitsiklis and Van Roy, 1997] hold, but they do not\nimply that leakage propagation occurs and under what conditions. Finally, we\ntest whether the problem could be mitigated with a better state representation,\nand whether it can be learned in an unsupervised manner, without rewards or\nprivileged information.\n",
"title": "Temporal Difference Learning with Neural Networks - Study of the Leakage Propagation Problem"
}
| null | null | null | null | true | null |
11179
| null |
Default
| null | null |
null |
{
"abstract": " We show that the first order theory of the lattice of open sets in some\nnatural topological spaces is $m$-equivalent to second order arithmetic. We\nalso show that for many natural computable metric spaces and computable domains\nthe first order theory of the lattice of effectively open sets is undecidable.\nMoreover, for several important spaces (e.g., $\\mathbb{R}^n$, $n\\geq1$, and the\ndomain $P\\omega$) this theory is $m$-equivalent to first order arithmetic.\n",
"title": "First Order Theories of Some Lattices of Open Sets"
}
| null | null | null | null | true | null |
11180
| null |
Default
| null | null |
null |
{
"abstract": " A cloud server spent a lot of time, energy and money to train a Viola-Jones\ntype object detector with high accuracy. Clients can upload their photos to the\ncloud server to find objects. However, the client does not want the leakage of\nthe content of his/her photos. In the meanwhile, the cloud server is also\nreluctant to leak any parameters of the trained object detectors. 10 years ago,\nAvidan & Butman introduced Blind Vision, which is a method for securely\nevaluating a Viola-Jones type object detector. Blind Vision uses standard\ncryptographic tools and is painfully slow to compute, taking a couple of hours\nto scan a single image. The purpose of this work is to explore an efficient\nmethod that can speed up the process. We propose the Random Base Image (RBI)\nRepresentation. The original image is divided into random base images. Only the\nbase images are submitted randomly to the cloud server. Thus, the content of\nthe image can not be leaked. In the meanwhile, a random vector and the secure\nMillionaire protocol are leveraged to protect the parameters of the trained\nobject detector. The RBI makes the integral-image enable again for the great\nacceleration. The experimental results reveal that our method can retain the\ndetection accuracy of that of the plain vision algorithm and is significantly\nfaster than the traditional blind vision, with only a very low probability of\nthe information leakage theoretically.\n",
"title": "Efficient Privacy Preserving Viola-Jones Type Object Detection via Random Base Image Representation"
}
| null | null | null | null | true | null |
11181
| null |
Default
| null | null |
null |
{
"abstract": " We study small-scale and high-frequency turbulent fluctuations in\nthree-dimensional flows under Fourier-mode reduction. The Navier-Stokes\nequations are evolved on a restricted set of modes, obtained as a projection on\na fractal or homogeneous Fourier set. We find a strong sensitivity (reduction)\nof the high-frequency variability of the Lagrangian velocity fluctuations on\nthe degree of mode decimation, similarly to what is already reported for\nEulerian statistics. This is quantified by a tendency towards a quasi-Gaussian\nstatistics, i.e., to a reduction of intermittency, at all scales and\nfrequencies. This can be attributed to a strong depletion of vortex filaments\nand of the vortex stretching mechanism. Nevertheless, we found that Eulerian\nand Lagrangian ensembles are still connected by a dimensional bridge-relation\nwhich is independent of the degree of Fourier-mode decimation.\n",
"title": "Lagrangian Statistics for Navier-Stokes Turbulence under Fourier-mode reduction: Fractal and Homogeneous Decimations"
}
| null | null | null | null | true | null |
11182
| null |
Default
| null | null |
null |
{
"abstract": " Hormozgan Province, located in the south of Iran, faces several challenges\nregarding water resources management. The first one is the discharge of a\nmassive volume of water to the Persian Gulf because of the concentration of the\nannual rainfalls in a short period of time and the narrow distance between the\nheadwater and the coast. The second one is the unbalanced development of\neconomic sectors in comparison with distribution of fresh water resources.\nFinally, long-term drought is also common in this area. The construction of a\ncarry-over dam (Esteghlal Dam) and several conveyance pipelines and withdrawing\nof the surface water and groundwater resources were considered as the solution\nto deal with those challenges. During recent drought, severe overdraft and\ninefficient use of recourses confirmed the fact that all done before are not\nenough. During this period, there was a tendency to store water in reservoir in\norder to meet the demand of the urban sector. Therefore, the agricultural\ndemand was the first victim of water allocation policy. It caused over\nexploitation of the groundwater resources (to meet the agricultural demand) and\nconsiderable losses (evaporation and leakage) from the reservoir. All of the\nabove-mentioned problems confirm the necessity of the development of a\nconjunctive use policy. In this paper, all demand related to the Esteghlal Dam\nand the Minab Aquifer (Bandarabbas City and agriculture in the Minab Plain)\nwere considered as the case study. The main objective was to find the best\napplicable conjunctive policy which as well guarantees the conservation of the\nMinab Aquifer. Alternative water allocation policies have been developed based\non the present capacities and the experience of local operating staff.\n",
"title": "Conjunctive management of surface and groundwater under severe drought: A case study in southern Iran"
}
| null | null |
[
"Physics"
] | null | true | null |
11183
| null |
Validated
| null | null |
null |
{
"abstract": " Biological and cellular systems are often modeled as graphs in which vertices\nrepresent objects of interest (genes, proteins, drugs) and edges represent\nrelational ties among these objects (binds-to, interacts-with, regulates). This\napproach has been highly successful owing to the theory, methodology and\nsoftware that support analysis and learning on graphs. Graphs, however, often\nsuffer from information loss when modeling physical systems due to their\ninability to accurately represent multiobject relationships. Hypergraphs, a\ngeneralization of graphs, provide a framework to mitigate information loss and\nunify disparate graph-based methodologies. In this paper, we present a\nhypergraph-based approach for modeling physical systems and formulate vertex\nclassification, edge classification and link prediction problems on\n(hyper)graphs as instances of vertex classification on (extended, dual)\nhypergraphs in a semi-supervised setting. We introduce a novel kernel method on\nvertex- and edge-labeled (colored) hypergraphs for analysis and learning. The\nmethod is based on exact and inexact (via hypergraph edit distances)\nenumeration of small simple hypergraphs, referred to as hypergraphlets, rooted\nat a vertex of interest. We extensively evaluate this method and show its\npotential use in a positive-unlabeled setting to estimate the number of missing\nand false positive links in protein-protein interaction networks.\n",
"title": "Classification in biological networks with hypergraphlet kernels"
}
| null | null | null | null | true | null |
11184
| null |
Default
| null | null |
null |
{
"abstract": " We study empirical statistical and gap distributions of several important\ntilings of the plane. In particular, we consider the slope distributions, the\nangle distributions, pair correlation, squared-distance pair correlation, angle\ngap distributions, and slope gap distributions for the Ammann Chair tiling, the\nrecently discovered fifteenth pentagonal tiling, and a few pertinent tilings\nrelated to these famous examples. We also consider the spatial statistics of\ngeneralized Ulam sets in two dimensions. Additionally, we carefully prove a\ntight asymptotic formula for the time steps in which Ulam set points at certain\nprescribed geometric positions in their plots in the plane formally enter the\nrecursively-defined sets.\nThe software we have developed to these generate numerical approximations to\nthe distributions for the tilings we consider here is written in Python under\nthe Sage environment and is released as open-source software which is available\nfreely on our websites. In addition to the small subset of tilings and other\npoint sets in the plane we study within the article, our program supports many\nother tiling variants and is easily extended for researchers to explore related\ntilings and iterative sets.\n",
"title": "Pair Correlation and Gap Distributions for Substitution Tilings and Generalized Ulam Sets in the Plane"
}
| null | null | null | null | true | null |
11185
| null |
Default
| null | null |
null |
{
"abstract": " Mobile payment systems are increasingly used to simplify the way in which\nmoney transfers and transactions can be performed. We argue that, to achieve\ntheir full potential as economic boosters in developing countries, mobile\npayment systems need to rely on new metaphors suitable for the business models,\nlifestyle, and technology availability conditions of the targeted communities.\nThe Pay-with-a-Group-Selfie (PGS) project, funded by the Melinda & Bill Gates\nFoundation, has developed a micro-payment system that supports everyday small\ntransactions by extending the reach of, rather than substituting, existing\npayment frameworks. PGS is based on a simple gesture and a readily\nunderstandable metaphor. The gesture - taking a selfie - has become part of the\nlifestyle of mobile phone users worldwide, including non-technology-savvy ones.\nThe metaphor likens computing two visual shares of the selfie to ripping a\nbanknote in two, a technique used for decades for delayed payment in cash-only\nmarkets. PGS is designed to work with devices with limited computational power\nand when connectivity is patchy or not always available. Thanks to visual\ncryptography techniques PGS uses for computing the shares, the original selfie\ncan be recomposed simply by stacking the shares, preserving the analogy with\nre-joining the two parts of the banknote.\n",
"title": "Pay-with-a-Selfie, a human-centred digital payment system"
}
| null | null | null | null | true | null |
11186
| null |
Default
| null | null |
null |
{
"abstract": " Deep convolutional Neural Networks (CNN) are the state-of-the-art performers\nfor object detection task. It is well known that object detection requires more\ncomputation and memory than image classification. Thus the consolidation of a\nCNN-based object detection for an embedded system is more challenging. In this\nwork, we propose LCDet, a fully-convolutional neural network for generic object\ndetection that aims to work in embedded systems. We design and develop an\nend-to-end TensorFlow(TF)-based model. Additionally, we employ 8-bit\nquantization on the learned weights. We use face detection as a use case. Our\nTF-Slim based network can predict different faces of different shapes and sizes\nin a single forward pass. Our experimental results show that the proposed\nmethod achieves comparative accuracy comparing with state-of-the-art CNN-based\nface detection methods, while reducing the model size by 3x and memory-BW by\n~4x comparing with one of the best real-time CNN-based object detector such as\nYOLO. TF 8-bit quantized model provides additional 4x memory reduction while\nkeeping the accuracy as good as the floating point model. The proposed model\nthus becomes amenable for embedded implementations.\n",
"title": "LCDet: Low-Complexity Fully-Convolutional Neural Networks for Object Detection in Embedded Systems"
}
| null | null | null | null | true | null |
11187
| null |
Default
| null | null |
null |
{
"abstract": " We study spin-2 deformed-AKLT models on the square lattice, specifically a\ntwo-parameter family of $O(2)$-symmetric ground-state wavefunctions as defined\nby Niggemann, Klümper, and Zittartz, who found previously that the phase\ndiagram consists of a Néel-ordered phase and a disordered phase which\ncontains the AKLT point. Using tensor-network methods, we not only confirm the\nNéel phase but also find an XY phase with quasi-long-range order and a region\nadjacent to it, within the AKLT phase, with very large correlation length, and\ninvestigate the consequences of a perfectly factorizable point at the corner of\nthat phase.\n",
"title": "Phase transitions of a 2D deformed-AKLT model"
}
| null | null | null | null | true | null |
11188
| null |
Default
| null | null |
null |
{
"abstract": " We consider statistical estimation of superhedging prices using historical\nstock returns in a frictionless market with d traded assets. We introduce a\nsimple plugin estimator based on empirical measures, show it is consistent but\nlacks suitable robustness. This is addressed by our improved estimators which\nuse a larger set of martingale measures defined through a tradeoff between the\nradius of Wasserstein balls around the empirical measure and the allowed norm\nof martingale densities. We also study convergence rates, convergence of\nsuperhedging strategies, and our study extends, in part, to the case of a\nmarket with traded options and to a multiperiod setting.\n",
"title": "Statistical estimation of superhedging prices"
}
| null | null | null | null | true | null |
11189
| null |
Default
| null | null |
null |
{
"abstract": " This paper proves that an irreducible subfactor planar algebra with a\ndistributive biprojection lattice admits a minimal 2-box projection generating\nthe identity biprojection. It is a generalization of a theorem of Ore on\nintervals of finite groups, conjectured by the author since 2013. We deduce a\nlink between combinatorics and representations in finite groups theory, related\nto an open problem of K.S. Brown in algebraic combinatorics.\n",
"title": "Ore's theorem on subfactor planar algebras"
}
| null | null | null | null | true | null |
11190
| null |
Default
| null | null |
null |
{
"abstract": " The latest techniques from Neural Networks and Support Vector Machines (SVM)\nare used to investigate geometric properties of Complete Intersection\nCalabi-Yau (CICY) threefolds, a class of manifolds that facilitate string model\nbuilding. An advanced neural network classifier and SVM are employed to (1)\nlearn Hodge numbers and report a remarkable improvement over previous efforts,\n(2) query for favourability, and (3) predict discrete symmetries, a highly\nimbalanced problem to which both Synthetic Minority Oversampling Technique\n(SMOTE) and permutations of the CICY matrix are used to decrease the class\nimbalance and improve performance. In each case study, we employ a genetic\nalgorithm to optimise the hyperparameters of the neural network. We demonstrate\nthat our approach provides quick diagnostic tools capable of shortlisting\nquasi-realistic string models based on compactification over smooth CICYs and\nfurther supports the paradigm that classes of problems in algebraic geometry\ncan be machine learned.\n",
"title": "Machine Learning CICY Threefolds"
}
| null | null | null | null | true | null |
11191
| null |
Default
| null | null |
null |
{
"abstract": " Weak gravitational lensing alters the apparent separations between observed\nsources, potentially affecting clustering statistics. We derive a general\nexpression for the lensing deflection which is valid for any three-point\nstatistic, and investigate its effect on the three-point clustering correlation\nfunction. We find that deflection of the clustering correlation function is\ngreatest at around $z=2$. It is most prominent in regions where the correlation\nfunction varies rapidly, in particular at the baryon acoustic oscillation scale\nwhere it smooths out the peaks and troughs, reducing the peak-to-trough\ndifference by about 0.1 percent at $z=1$ and around 2.3 percent at $z=10$. The\nmodification due to lensing deflection is typically at the per cent level of\nthe expected errors in a Euclid-like survey and therefore undetectable.\n",
"title": "Weak lensing deflection of three-point correlation functions"
}
| null | null | null | null | true | null |
11192
| null |
Default
| null | null |
null |
{
"abstract": " Solitons are of the important significant in many fields of nonlinear science\nsuch as nonlinear optics, Bose-Einstein condensates, plamas physics, biology,\nfluid mechanics, and etc.. The stable solitons have been captured not only\ntheoretically and experimentally in both linear and nonlinear Schrodinger (NLS)\nequations in the presence of non-Hermitian potentials since the concept of the\nparity-time (PT)-symmetry was introduced in 1998. In this paper, we present\nnovel bright solitons of the NLS equation with third-order dispersion in some\ncomplex PT-symmetric potentials (e.g., physically relevant PT-symmetric\nScarff-II-like and harmonic-Gaussian potentials). We find stable nonlinear\nmodes even if the respective linear PT-symmetric phases are broken. Moreover,\nwe also use the adiabatic changes of the control parameters to excite the\ninitial modes related to exact solitons to reach stable nonlinear modes. The\nelastic interactions of two solitons are exhibited in the third-order NLS\nequation with PT-symmetric potentials. Our results predict the dynamical\nphenomena of soliton equations in the presence of third-order dispersion and\nPT-symmetric potentials arising in nonlinear fiber optics and other physically\nrelevant fields.\n",
"title": "Solitonic dynamics and excitations of the nonlinear Schrodinger equation with third-order dispersion in non-Hermitian PT-symmetric potentials"
}
| null | null | null | null | true | null |
11193
| null |
Default
| null | null |
null |
{
"abstract": " We explore several problems related to ruled polygons. Given a ruling of a\npolygon $P$, we consider the Reeb graph of $P$ induced by the ruling. We define\nthe Reeb complexity of $P$, which roughly equates to the minimum number of\npoints necessary to support $P$. We give asymptotically tight bounds on the\nReeb complexity that are also tight up to a small additive constant. When\nrestricted to the set of parallel rulings, we show that the Reeb complexity can\nbe computed in polynomial time.\n",
"title": "Supporting Ruled Polygons"
}
| null | null | null | null | true | null |
11194
| null |
Default
| null | null |
null |
{
"abstract": " The recent Nobel-prize-winning detections of gravitational waves from merging\nblack holes and the subsequent detection of the collision of two neutron stars\nin coincidence with electromagnetic observations have inaugurated a new era of\nmultimessenger astrophysics. To enhance the scope of this emergent field of\nscience, we pioneered the use of deep learning with convolutional neural\nnetworks, that take time-series inputs, for rapid detection and\ncharacterization of gravitational wave signals. This approach, Deep Filtering,\nwas initially demonstrated using simulated LIGO noise. In this article, we\npresent the extension of Deep Filtering using real data from LIGO, for both\ndetection and parameter estimation of gravitational waves from binary black\nhole mergers using continuous data streams from multiple LIGO detectors. We\ndemonstrate for the first time that machine learning can detect and estimate\nthe true parameters of real events observed by LIGO. Our results show that Deep\nFiltering achieves similar sensitivities and lower errors compared to\nmatched-filtering while being far more computationally efficient and more\nresilient to glitches, allowing real-time processing of weak time-series\nsignals in non-stationary non-Gaussian noise with minimal resources, and also\nenables the detection of new classes of gravitational wave sources that may go\nunnoticed with existing detection algorithms. This unified framework for data\nanalysis is ideally suited to enable coincident detection campaigns of\ngravitational waves and their multimessenger counterparts in real-time.\n",
"title": "Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation: Results with Advanced LIGO Data"
}
| null | null | null | null | true | null |
11195
| null |
Default
| null | null |
null |
{
"abstract": " A self-doping effect between outer and inner CuO$_2$ planes (OPs and IPs) in\nmulti-layer cuprate superconductors is studied. When one considers a\nthree-layer tight-binding model of the Hg-based three-layer cuprate derived\nfrom the first principle calculations, the electron concentration gets to be\nlarge in the OP compared to IP. This is inconsistent with the experimental fact\nthat more hole carriers tend to be introduced into the OP than IP.We\ninvestigate a three-layer Hubbard model with the two-particle self-consistent\napproach for multi-layer systems to incorporate electron correlations. We\nobserve that the double occupancy (antiferromagnetic instability) in the IP\ndecreases (increases) more than the OP, and also reveal that more electrons\ntend to be introduced into the IP than OP to obtain the energy gain from the\non-site Hubbard interaction. These results are consistent with the experimental\nfacts, and this electron distribution between the OP and IP can be interpreted\nas a self-doping effect arising from strong electron correlations.\n",
"title": "Self-doping effect arising from electron correlations in multi-layer cuprates"
}
| null | null | null | null | true | null |
11196
| null |
Default
| null | null |
null |
{
"abstract": " Player selection is one the most important tasks for any sport and cricket is\nno exception. The performance of the players depends on various factors such as\nthe opposition team, the venue, his current form etc. The team management, the\ncoach and the captain select 11 players for each match from a squad of 15 to 20\nplayers. They analyze different characteristics and the statistics of the\nplayers to select the best playing 11 for each match. Each batsman contributes\nby scoring maximum runs possible and each bowler contributes by taking maximum\nwickets and conceding minimum runs. This paper attempts to predict the\nperformance of players as how many runs will each batsman score and how many\nwickets will each bowler take for both the teams. Both the problems are\ntargeted as classification problems where number of runs and number of wickets\nare classified in different ranges. We used naïve bayes, random forest,\nmulticlass SVM and decision tree classifiers to generate the prediction models\nfor both the problems. Random Forest classifier was found to be the most\naccurate for both the problems.\n",
"title": "Increased Prediction Accuracy in the Game of Cricket using Machine Learning"
}
| null | null | null | null | true | null |
11197
| null |
Default
| null | null |
null |
{
"abstract": " Probabilistic modeling enables combining domain knowledge with learning from\ndata, thereby supporting learning from fewer training instances than purely\ndata-driven methods. However, learning probabilistic models is difficult and\nhas not achieved the level of performance of methods such as deep neural\nnetworks on many tasks. In this paper, we attempt to address this issue by\npresenting a method for learning the parameters of a probabilistic program\nusing backpropagation. Our approach opens the possibility to building deep\nprobabilistic programming models that are trained in a similar way to neural\nnetworks.\n",
"title": "Learning Probabilistic Programs Using Backpropagation"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
11198
| null |
Validated
| null | null |
null |
{
"abstract": " While supermassive black holes are known to co-evolve with their host galaxy,\nthe precise nature and origin of this co-evolution is not clear. We here\nexplore the possible connection between star formation and black hole growth in\nthe circumnuclear disk (CND) to probe this connection in the vicinity close to\nthe black hole. We adopt here the circumnuclear disk model developed by\nKawakatu & Wada (2008) and Wutschik et al. (2013), and explore both the\ndependence on the star formation recipe as well as the role of the\ngravitational field, which can be dominated by the central black hole, the CND\nitself or the host galaxy. A specific emphasis is put on the turbulence\nregulated star formation model by Krumholz et al. (2005) to explore the impact\nof a realistic star formation recipe. It is shown that this model helps to\nintroduce realistic fluctuations in the black hole and star formation rate,\nwithout overestimating them. Consistent with previous works, we show that the\nfinal black hole masses are rather insensitive to the masses of the initial\nseeds, even for seed masses of up to 10^6 M_sol. In addition, we apply our\nmodel to the formation of high-redshift quasars, as well as to the nearby\nsystem NGC 6951, where a tentative comparison is made in spite of the presence\nof a bar in the galaxy. We show that our model can reproduce the high black\nhole masses of the high-redshift quasars within a sufficiently short time,\nprovided a high mass supply rate from the host galaxy. In addition, it\nreproduces several of the properties observed in NGC 6951. With respect to the\nlatter system, our analysis suggests that supernova feedback may be important\nto create the observed fluctuations in the star formation history as a result\nof negative feedback effects.\n",
"title": "Turbulent gas accretion between supermassive black holes and star-forming rings in the circumnuclear disk"
}
| null | null | null | null | true | null |
11199
| null |
Default
| null | null |
null |
{
"abstract": " Many different approaches for neural network based hash functions have been\nproposed. Statistical analysis must correlate security of them. This paper\nproposes novel neural hashing approach for gray scale image authentication. The\nsuggested system is rapid, robust, useful and secure. Proposed hash function\ngenerates hash values using neural network one-way property and non-linear\ntechniques. As a result security and performance analysis are performed and\nsatisfying results are achieved. These features are dominant reasons for\npreferring against traditional ones.\n",
"title": "Grayscale Image Authentication using Neural Hashing"
}
| null | null | null | null | true | null |
11200
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.