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": " One of the most prevalent symptoms among the elderly population, dementia,\ncan be detected by classifiers trained on linguistic features extracted from\nnarrative transcripts. However, these linguistic features are impacted in a\nsimilar but different fashion by the normal aging process. Aging is therefore a\nconfounding factor, whose effects have been hard for machine learning\nclassifiers to isolate. In this paper, we show that deep neural network (DNN)\nclassifiers can infer ages from linguistic features, which is an entanglement\nthat could lead to unfairness across age groups. We show this problem is caused\nby undesired activations of v-structures in causality diagrams, and it could be\naddressed with fair representation learning. We build neural network\nclassifiers that learn low-dimensional representations reflecting the impacts\nof dementia yet discarding the effects of age. To evaluate these classifiers,\nwe specify a model-agnostic score $\\Delta_{eo}^{(N)}$ measuring how classifier\nresults are disentangled from age. Our best models outperform baseline neural\nnetwork classifiers in disentanglement, while compromising accuracy by as\nlittle as 2.56\\% and 2.25\\% on DementiaBank and the Famous People dataset\nrespectively.\n",
"title": "Isolating effects of age with fair representation learning when assessing dementia"
}
| null | null | null | null | true | null |
11601
| null |
Default
| null | null |
null |
{
"abstract": " Nonconvex penalty methods for sparse modeling in linear regression have been\na topic of fervent interest in recent years. Herein, we study a family of\nnonconvex penalty functions that we call the trimmed Lasso and that offers\nexact control over the desired level of sparsity of estimators. We analyze its\nstructural properties and in doing so show the following:\n1) Drawing parallels between robust statistics and robust optimization, we\nshow that the trimmed-Lasso-regularized least squares problem can be viewed as\na generalized form of total least squares under a specific model of\nuncertainty. In contrast, this same model of uncertainty, viewed instead\nthrough a robust optimization lens, leads to the convex SLOPE (or OWL) penalty.\n2) Further, in relating the trimmed Lasso to commonly used sparsity-inducing\npenalty functions, we provide a succinct characterization of the connection\nbetween trimmed-Lasso- like approaches and penalty functions that are\ncoordinate-wise separable, showing that the trimmed penalties subsume existing\ncoordinate-wise separable penalties, with strict containment in general.\n3) Finally, we describe a variety of exact and heuristic algorithms, both\nexisting and new, for trimmed Lasso regularized estimation problems. We include\na comparison between the different approaches and an accompanying\nimplementation of the algorithms.\n",
"title": "The Trimmed Lasso: Sparsity and Robustness"
}
| null | null | null | null | true | null |
11602
| null |
Default
| null | null |
null |
{
"abstract": " Mobile phones identification through their built in components has been\ndemonstrated in literature for various types of sensors including the camera,\nmicrophones and accelerometers. The identification is performed by the\nexploitation of the small but significant differences in the electronic\ncircuits generated during the production process. Thus, these differences\nbecome an intrinsic property of the electronic components, which can be\ndetected and become an unique fingerprint of the component and of the mobile\nphone. In this paper, we investigate the identification of mobile phones\nthrough their builtin magnetometers, which has not been reported in literature\nyet. Magnetometers are stimulated with different waveforms using a solenoid\nconnected to a computer s audio board. The identification is performed\nanalyzing the digital output of the magnetometer through the use of statistical\nfeatures and the Support Vector Machine (SVM) machine learning algorithm. We\nprove that this technique can distinguish different models and brands with very\nhigh accuracy but it can only distinguish phones of the same model with limited\naccuracy.\n",
"title": "Mobile phone identification through the built-in magnetometers"
}
| null | null | null | null | true | null |
11603
| null |
Default
| null | null |
null |
{
"abstract": " This paper demonstrates how to apply machine learning algorithms to\ndistinguish good stocks from the bad stocks. To this end, we construct 244\ntechnical and fundamental features to characterize each stock, and label stocks\naccording to their ranking with respect to the return-to-volatility ratio.\nAlgorithms ranging from traditional statistical learning methods to recently\npopular deep learning method, e.g. Logistic Regression (LR), Random Forest\n(RF), Deep Neural Network (DNN), and the Stacking, are trained to solve the\nclassification task. Genetic Algorithm (GA) is also used to implement feature\nselection. The effectiveness of the stock selection strategy is validated in\nChinese stock market in both statistical and practical aspects, showing that:\n1) Stacking outperforms other models reaching an AUC score of 0.972; 2) Genetic\nAlgorithm picks a subset of 114 features and the prediction performances of all\nmodels remain almost unchanged after the selection procedure, which suggests\nsome features are indeed redundant; 3) LR and DNN are radical models; RF is\nrisk-neutral model; Stacking is somewhere between DNN and RF. 4) The portfolios\nconstructed by our models outperform market average in back tests.\n",
"title": "A Machine Learning Framework for Stock Selection"
}
| null | null |
[
"Statistics",
"Quantitative Finance"
] | null | true | null |
11604
| null |
Validated
| null | null |
null |
{
"abstract": " The article outlines in memoriam Prof. Pavel Zampa's concepts of system\ntheory which enable to devise a measurement in dynamic systems independently of\nthe particular system behaviour. From the point of view of Zampa's theory,\nterms like system time, system attributes, system link, system element, input,\noutput, subsystems, and state variables are defined. In Conclusions, Zampa's\ntheory is discussed together with another mathematical approaches of\nqualitative dynamics known since the 19th century. In Appendices, we present\napplications of Zampa's technical approach to measurement of complex dynamical\n(chemical and biological) systems at the Institute of Complex Systems,\nUniversity of South Bohemia in Ceske Budejovice.\n",
"title": "Zampa's systems theory: a comprehensive theory of measurement in dynamic systems"
}
| null | null |
[
"Computer Science"
] | null | true | null |
11605
| null |
Validated
| null | null |
null |
{
"abstract": " We consider the problem of classifying data manifolds where each manifold\nrepresents invariances that are parameterized by continuous degrees of freedom.\nConventional data augmentation methods rely upon sampling large numbers of\ntraining examples from these manifolds; instead, we propose an iterative\nalgorithm called M_{CP} based upon a cutting-plane approach that efficiently\nsolves a quadratic semi-infinite programming problem to find the maximum margin\nsolution. We provide a proof of convergence as well as a polynomial bound on\nthe number of iterations required for a desired tolerance in the objective\nfunction. The efficiency and performance of M_{CP} are demonstrated in\nhigh-dimensional simulations and on image manifolds generated from the ImageNet\ndataset. Our results indicate that M_{CP} is able to rapidly learn good\nclassifiers and shows superior generalization performance compared with\nconventional maximum margin methods using data augmentation methods.\n",
"title": "Learning Data Manifolds with a Cutting Plane Method"
}
| null | null | null | null | true | null |
11606
| null |
Default
| null | null |
null |
{
"abstract": " The magnetism of ordered and disordered La$_2$NiMnO$_6$ is explained using a\nmodel involving double exchange and superexchange. The concept of majority spin\nhybridization in the large coupling limit is used to explain the ferromagnetism\nof La$_2$NiMnO$_6$ as compared to the ferrimagnetism of Sr$_{2}$FeMoO$_{6}$.\nThe ferromagnetic insulating ground state in the ordered phase is explained.\nThe essential role played by the Ni-Mn superexchange between the Ni $e_{g}$\nelectron spins and the Mn $t_{2g}$ core electron spins in realizing this ground\nstate, is outlined. In presence of antisite disorder, the model system is found\nto exhibit a tendency of becoming a spin-glass at low temperatures, while it\ncontinues to retain a ferromagnetic transition at higher temperatures, similar\nto recent experimental observations [D. Choudhury .et.al., Phys. Rev. Lett.\n108, 127201 (2012)]. This reentrant spin-glass or reentrant ferromagnetic\nbehaviour is explained in terms of the competition of the ferromagnetic double\nexchange between the Ni $e_{g}$ and the Mn $e_{g}$ electrons, and the\nferromagnetic Ni-Mn superexchange, with the antiferromagnetic antisite Mn-Mn\nsuperexchange.\n",
"title": "Theory of magnetism in La$_2$NiMnO$_6$"
}
| null | null | null | null | true | null |
11607
| null |
Default
| null | null |
null |
{
"abstract": " Partially Observable Markov Decision Process (POMDP) is widely used to model\nprobabilistic behavior for complex systems. Compared with MDPs, POMDP models a\nsystem more accurate but solving a POMDP generally takes exponential time in\nthe size of its state space. This makes the formal verification and synthesis\nproblems much more challenging for POMDPs, especially when multiple system\ncomponents are involved. As a promising technique to reduce the verification\ncomplexity, the abstraction method tries to find an abstract system with a\nsmaller state space but preserves enough properties for the verification\npurpose. While abstraction based verification has been explored extensively for\nMDPs, in this paper, we present the first result of POMDP abstraction and its\nrefinement techniques. The main idea follows the counterexample-guided\nabstraction refinement (CEGAR) framework. Starting with a coarse guess for the\nPOMDP abstraction, we iteratively use counterexamples from formal verification\nto refine the abstraction until the abstract system can be used to infer the\nverification result for the original POMDP. Our main contributions have two\nfolds: 1) we propose a novel abstract system model for POMDP and a new\nsimulation relation to capture the partial observability then prove the\npreservation on a fragment of Probabilistic Computation Tree Logic (PCTL); 2)\nto find a proper abstract system that can prove or disprove the satisfaction\nrelation on the concrete POMDP, we develop a novel refinement algorithm. Our\nwork leads to a sound and complete CEGAR framework for POMDP.\n",
"title": "Counterexample-guided Abstraction Refinement for POMDPs"
}
| null | null | null | null | true | null |
11608
| null |
Default
| null | null |
null |
{
"abstract": " Multi-task learning (MTL) has led to successes in many applications of\nmachine learning, from natural language processing and speech recognition to\ncomputer vision and drug discovery. This article aims to give a general\noverview of MTL, particularly in deep neural networks. It introduces the two\nmost common methods for MTL in Deep Learning, gives an overview of the\nliterature, and discusses recent advances. In particular, it seeks to help ML\npractitioners apply MTL by shedding light on how MTL works and providing\nguidelines for choosing appropriate auxiliary tasks.\n",
"title": "An Overview of Multi-Task Learning in Deep Neural Networks"
}
| null | null | null | null | true | null |
11609
| null |
Default
| null | null |
null |
{
"abstract": " Large-area ($\\sim$cm$^2$) films of vertical heterostructures formed by\nalternating graphene and transition-metal dichalcogenide(TMD) alloys are\nobtained by wet chemical routes followed by a thermal treatment at low\ntemperature (300 $^\\circ$C). In particular, we synthesized stacked graphene and\nW$_x$Mo$_{1-x}$S$_2$ alloy phases that were used as hydrogen evolution\ncatalysts. We observed a Tafel slope of 38.7 mV dec$^{-1}$ and 96 mV onset\npotential (at current density of 10 mA cm$^{-2}$) when the heterostructure\nalloy is annealed at 300 $^o$C. These results indicate that heterostructure\nformed by graphene and W$_{0.4}$Mo$_{0.6}$S$_2$ alloys are far more efficient\nthan WS$_2$ and MoS$_2$ by at least a factor of two, and it is superior than\nany other reported TMD system. This strategy offers a cheap and low temperature\nsynthesis alternative able to replace Pt in the hydrogen evolution reaction\n(HER). Furthermore, the catalytic activity of the alloy is stable over time,\ni.e. the catalytic activity does not experience a significant change even after\n1000 cycles. Using density functional theory calculations, we found that this\nenhanced hydrogen evolution in the W$_x$Mo$_{1-x}$S$_2$ alloys is mainly due to\nthe lower energy barrier created by a favorable overlap of the d-orbitals from\nthe transition metals and the s-orbitals of H$_2$, with the lowest energy\nbarrier occurring for W$_{0.4}$Mo$_{0.6}$S$_2$ alloy. Thus, it is now possible\nto further improve the performance of the \"inert\" TMD basal plane via metal\nalloying, in addition to the previously reported strategies of creation of\npoint defects, vacancies and edges. The synthesis of\ngraphene/W$_{0.4}$Mo$_{0.6}$S$_2$ produced at relatively low temperatures is\nscalable and could be used as an effective low cost Pt-free catalyst.\n",
"title": "Low temperature synthesis of heterostructures of transition metal dichalcogenide alloys (WxMo1-xS2) and graphene with superior catalytic performance for hydrogen evolution"
}
| null | null |
[
"Physics"
] | null | true | null |
11610
| null |
Validated
| null | null |
null |
{
"abstract": " We conjecture a formula for the generating function of virtual\n$\\chi_y$-genera of moduli spaces of rank 2 sheaves on arbitrary surfaces with\nholomorphic 2-form. Specializing the conjecture to minimal surfaces of general\ntype and to virtual Euler characteristics, we recover (part of) a formula of C.\nVafa and E. Witten.\nThese virtual $\\chi_y$-genera can be written in terms of descendent Donaldson\ninvariants. Using T. Mochizuki's formula, the latter can be expressed in terms\nof Seiberg-Witten invariants and certain explicit integrals over Hilbert\nschemes of points. These integrals are governed by seven universal functions,\nwhich are determined by their values on $\\mathbb{P}^2$ and $\\mathbb{P}^1 \\times\n\\mathbb{P}^1$. Using localization we calculate these functions up to some\norder, which allows us to check our conjecture in many cases.\nIn an appendix by H. Nakajima and the first named author, the virtual Euler\ncharacteristic specialization of our conjecture is extended to include\n$\\mu$-classes, thereby interpolating between Vafa-Witten's formula and Witten's\nconjecture for Donaldson invariants.\n",
"title": "Virtual refinements of the Vafa-Witten formula"
}
| null | null | null | null | true | null |
11611
| null |
Default
| null | null |
null |
{
"abstract": " Using variational Bayes neural networks, we develop an algorithm capable of\naccumulating knowledge into a prior from multiple different tasks. The result\nis a rich and meaningful prior capable of few-shot learning on new tasks. The\nposterior can go beyond the mean field approximation and yields good\nuncertainty on the performed experiments. Analysis on toy tasks shows that it\ncan learn from significantly different tasks while finding similarities among\nthem. Experiments of Mini-Imagenet yields the new state of the art with 74.5%\naccuracy on 5 shot learning. Finally, we provide experiments showing that other\nexisting methods can fail to perform well in different benchmarks.\n",
"title": "Uncertainty in Multitask Transfer Learning"
}
| null | null | null | null | true | null |
11612
| null |
Default
| null | null |
null |
{
"abstract": " Theorems and techniques to form different types of transformationally\ninvariant processing and to produce the same output quantitatively based on\neither transformationally invariant operators or symmetric operations have\nrecently been introduced by the authors. In this study, we further propose to\ncompose a geared rotationally identical CNN system (GRI-CNN) with a small step\nangle by connecting networks of participated processes at the first flatten\nlayer. Using an ordinary CNN structure as a base, requirements for constructing\na GRI-CNN include the use of either symmetric input vector or kernels with an\nangle increment that can form a complete cycle as a \"gearwheel\". Four basic\nGRI-CNN structures were studied. Each of them can produce quantitatively\nidentical output results when a rotation angle of the input vector is evenly\ndivisible by the step angle of the gear. Our study showed when an input vector\nrotated with an angle does not match to a step angle, the GRI-CNN can also\nproduce a highly consistent result. With a design of using an ultra-fine\ngear-tooth step angle (e.g., 1 degree or 0.1 degree), all four GRI-CNN systems\ncan be constructed virtually isotropically.\n",
"title": "Geared Rotationally Identical and Invariant Convolutional Neural Network Systems"
}
| null | null | null | null | true | null |
11613
| null |
Default
| null | null |
null |
{
"abstract": " Suppose that Alice and Bob are located in distant laboratories, which are\nconnected by an ideal quantum channel. Suppose further that they share many\ncopies of a quantum state $\\rho_{ABE}$, such that Alice possesses the $A$\nsystems and Bob the $BE$ systems. In our model, there is an identifiable part\nof Bob's laboratory that is insecure: a third party named Eve has infiltrated\nBob's laboratory and gained control of the $E$ systems. Alice, knowing this,\nwould like use their shared state and the ideal quantum channel to communicate\na message in such a way that Bob, who has access to the whole of his laboratory\n($BE$ systems), can decode it, while Eve, who has access only to a sector of\nBob's laboratory ($E$ systems) and the ideal quantum channel connecting Alice\nto Bob, cannot learn anything about Alice's transmitted message. We call this\ntask the conditional one-time pad, and in this paper, we prove that the optimal\nrate of secret communication for this task is equal to the conditional quantum\nmutual information $I(A;B|E)$ of their shared state. We thus give the\nconditional quantum mutual information an operational meaning that is different\nfrom those given in prior works, via state redistribution, conditional erasure,\nor state deconstruction. We also generalize the model and method in several\nways, one of which demonstrates that the negative tripartite interaction\ninformation $-I_{3}(A;B;E) = I(A;BE)-I(A;B)-I(A;E)$ of a tripartite state\n$\\rho_{ABE}$ is an achievable rate for a secret-sharing task, i.e., the case in\nwhich Alice's message should be secure from someone possessing only the $AB$ or\n$AE$ systems but should be decodable by someone possessing all systems $A$,\n$B$, and $E$.\n",
"title": "Conditional quantum one-time pad"
}
| null | null | null | null | true | null |
11614
| null |
Default
| null | null |
null |
{
"abstract": " We study the Bratteli diagram of 2-Sylow subgroups of symmetric groups. We\nshow that it is simple, has a recursive structure, and self-similarities at all\nscales. We contrast its subgraph of one-dimensional representations with the\nMacdonald tree.\n",
"title": "The Representation Theory of 2-Sylow Subgroups of the Symmetric Group"
}
| null | null | null | null | true | null |
11615
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we prove that under proper conditions, bootstrap can further\ndebias the debiased Lasso estimator for statistical inference of\nlow-dimensional parameters in high-dimensional linear regression. We prove that\nthe required sample size for inference with bootstrapped debiased Lasso, which\ninvolves the number of small coefficients, can be of smaller order than the\nexisting ones for the debiased Lasso. Therefore, our results reveal the\nbenefits of having strong signals. Our theory is supported by results of\nsimulation experiments, which compare coverage probabilities and lengths of\nconfidence intervals with and without bootstrap, with and without debiasing.\n",
"title": "Debiasing the Debiased Lasso with Bootstrap"
}
| null | null | null | null | true | null |
11616
| null |
Default
| null | null |
null |
{
"abstract": " Recently, He and Owen (2016) proposed the use of Hilbert's space filling\ncurve (HSFC) in numerical integration as a way of reducing the dimension from\n$d>1$ to $d=1$. This paper studies the asymptotic normality of the HSFC-based\nestimate when using scrambled van der Corput sequence as input. We show that\nthe estimate has an asymptotic normal distribution for functions in\n$C^1([0,1]^d)$, excluding the trivial case of constant functions. The\nasymptotic normality also holds for discontinuous functions under mild\nconditions. It was previously known only that scrambled $(0,m,d)$-net\nquadratures enjoy the asymptotic normality for smooth enough functions, whose\nmixed partial gradients satisfy a Hölder condition. As a by-product, we find\nlower bounds for the variance of the HSFC-based estimate. Particularly, for\nnontrivial functions in $C^1([0,1]^d)$, the low bound is of order $n^{-1-2/d}$,\nwhich matches the rate of the upper bound established in He and Owen (2016).\n",
"title": "Asymptotic Normality of Extensible Grid Sampling"
}
| null | null | null | null | true | null |
11617
| null |
Default
| null | null |
null |
{
"abstract": " Stratum, the de-facto mining communication protocol used by blockchain based\ncryptocurrency systems, enables miners to reliably and efficiently fetch jobs\nfrom mining pool servers. In this paper we exploit Stratum's lack of encryption\nto develop passive and active attacks on Bitcoin's mining protocol, with\nimportant implications on the privacy, security and even safety of mining\nequipment owners. We introduce StraTap and ISP Log attacks, that infer miner\nearnings if given access to miner communications, or even their logs. We\ndevelop BiteCoin, an active attack that hijacks shares submitted by miners, and\ntheir associated payouts. We build BiteCoin on WireGhost, a tool we developed\nto hijack and surreptitiously maintain Stratum connections. Our attacks reveal\nthat securing Stratum through pervasive encryption is not only undesirable (due\nto large overheads), but also ineffective: an adversary can predict miner\nearnings even when given access to only packet timestamps. Instead, we devise\nBedrock, a minimalistic Stratum extension that protects the privacy and\nsecurity of mining participants. We introduce and leverage the mining cookie\nconcept, a secret that each miner shares with the pool and includes in its\npuzzle computations, and that prevents attackers from reconstructing or\nhijacking the puzzles. We have implemented our attacks and collected 138MB of\nStratum protocol traffic from mining equipment in the US and Venezuela. We show\nthat Bedrock is resilient to active attacks even when an adversary breaks the\ncrypto constructs it uses. Bedrock imposes a daily overhead of 12.03s on a\nsingle pool server that handles mining traffic from 16,000 miners.\n",
"title": "Hardening Stratum, the Bitcoin Pool Mining Protocol"
}
| null | null | null | null | true | null |
11618
| null |
Default
| null | null |
null |
{
"abstract": " This paper provides a comparison between the $K$-structure of unipotent\nrepresentations and regular sections of bundles on nilpotent orbits for complex\ngroups of type $D$. Precisely, let $ G_ 0 =Spin(2n,\\mathbb C)$ be the Spin\ncomplex group viewed as a real group, and $K\\cong G_0$ be the complexification\nof the maximal compact subgroup of $G_0$. We compute $K$-spectra of the regular\nfunctions on some small nilpotent orbits $\\mathcal O$ transforming according to\ncharacters $\\psi$ of $C_{ K}(\\mathcal O)$ trivial on the connected component of\nthe identity $C_{ K}(\\mathcal O)^0$. We then match them with the ${K}$-types of\nthe genuine (i.e. representations which do not factor to $SO(2n,\\mathbb C)$)\nunipotent representations attached to $\\mathcal O$.\n",
"title": "Representations associated to small nilpotent orbits for complex Spin groups"
}
| null | null | null | null | true | null |
11619
| null |
Default
| null | null |
null |
{
"abstract": " This paper introduces a new free library for the Python programming language,\nwhich provides a collection of structured linear transforms, that are not\nrepresented as explicit two dimensional arrays but in a more efficient way by\nexploiting the structural knowledge.\nThis allows fast and memory savy forward and backward transformations while\nalso provding a clean but still flexible interface to these effcient\nalgorithms, thus making code more readable, scable and adaptable.\nWe first outline the goals of this library, then how they were achieved and\nlastly we demonstrate the performance compared to current state of the art\npackages available for Python.\nThis library is released and distributed under a free license.\n",
"title": "Fast Linear Transformations in Python"
}
| null | null | null | null | true | null |
11620
| null |
Default
| null | null |
null |
{
"abstract": " Objectivity is often considered as an ideal for scientific description of\nnature. When we describe physical phenomena, thus, we have exclusively taken an\nobjective viewpoint by excluding a subject. Here we consider how nature can be\ndescribed from a subjective viewpoint and how it is related to the objective\ndescription. To this end, we introduce a viewpoint-shift operation that sets\nperspective within a system, and subject the system to the laws of\nthermodynamics. We consider a situation in which the activation of an active\npart of the system starts to influence an objective part at t = 0, bringing the\nsystem into non-equilibrium. We find that the perspective alters physical state\nfunctions of the system, or leaves a viewpoint-dependent physical trace that is\ndetectable. In the system-internal viewpoint, a system in the heat bath\nself-organizes to maximize the free energy, creating order. The active part\nkeeps increasing a gap from an initial equilibrium state as long as the energy\nis available, forming a memory in the form of organized matter. This outcome of\na system-internal viewpoint in physics matches our intuition coming from our\ndaily-life experience that our subjective action leads to a change of our\nenvironment. This suggests that this system-internal viewpoint may provide a\nclue to understand a long-standing problem on the physical meaning to be\nsubjective.\n",
"title": "Physical description of nature from a system-internal viewpoint"
}
| null | null | null | null | true | null |
11621
| null |
Default
| null | null |
null |
{
"abstract": " In this work we report the synthesis and structural, electronic and magnetic\nproperties of La1.5Ca0.5CoMnO6 double-perovskite. This is a re-entrant spin\ncluster material which exhibits a non-negligible negative exchange bias effect\nwhen it is cooled in zero magnetic field from an unmagnetized state down to low\ntemperature. X-ray powder diffraction, X-ray photoelectron spectroscopy and\nmagnetometry results indicate mixed valence state at Co site, leading to\ncompeting magnetic phases and uncompensated spins at the magnetic interfaces.\nWe compare the results for this Ca-doped material with those reported for the\nresemblant compound La1.5Sr0.5CoMnO6, and discuss the much smaller spontaneous\nexchange bias effect observed for the former in terms of its structural and\nmagnetic particularities. For La1.5Ca0.5CoMnO6, when successive magnetization\nloops are carried, the spontaneous exchange bias field inverts its sign from\nnegative to positive from the first to the second measurement. We discuss this\nbehavior based on the disorder at the magnetic interfaces, related to the\npresence of a glassy phase. This compound also exhibits a large conventional\nexchange bias, for which there is no sign inversion of the exchange bias field\nfor consecutive cycles.\n",
"title": "Training-induced inversion of spontaneous exchange bias field on La1.5Ca0.5CoMnO6"
}
| null | null | null | null | true | null |
11622
| null |
Default
| null | null |
null |
{
"abstract": " We consider the linear regression problem under semi-supervised settings\nwherein the available data typically consists of: (i) a small or moderate sized\n'labeled' data, and (ii) a much larger sized 'unlabeled' data. Such data arises\nnaturally from settings where the outcome, unlike the covariates, is expensive\nto obtain, a frequent scenario in modern studies involving large databases like\nelectronic medical records (EMR). Supervised estimators like the ordinary least\nsquares (OLS) estimator utilize only the labeled data. It is often of interest\nto investigate if and when the unlabeled data can be exploited to improve\nestimation of the regression parameter in the adopted linear model.\nIn this paper, we propose a class of 'Efficient and Adaptive Semi-Supervised\nEstimators' (EASE) to improve estimation efficiency. The EASE are two-step\nestimators adaptive to model mis-specification, leading to improved (optimal in\nsome cases) efficiency under model mis-specification, and equal (optimal)\nefficiency under a linear model. This adaptive property, often unaddressed in\nthe existing literature, is crucial for advocating 'safe' use of the unlabeled\ndata. The construction of EASE primarily involves a flexible\n'semi-non-parametric' imputation, including a smoothing step that works well\neven when the number of covariates is not small; and a follow up 'refitting'\nstep along with a cross-validation (CV) strategy both of which have useful\npractical as well as theoretical implications towards addressing two important\nissues: under-smoothing and over-fitting. We establish asymptotic results\nincluding consistency, asymptotic normality and the adaptive properties of\nEASE. We also provide influence function expansions and a 'double' CV strategy\nfor inference. The results are further validated through extensive simulations,\nfollowed by application to an EMR study on auto-immunity.\n",
"title": "Efficient and Adaptive Linear Regression in Semi-Supervised Settings"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
11623
| null |
Validated
| null | null |
null |
{
"abstract": " Five simple soft sensor methodologies with two update conditions were\ncompared on two experimentally-obtained datasets and one simulated dataset. The\nsoft sensors investigated were moving window partial least squares regression\n(and a recursive variant), moving window random forest regression, the mean\nmoving window of $y$, and a novel random forest partial least squares\nregression ensemble (RF-PLS), all of which can be used with small sample sizes\nso that they can be rapidly placed online. It was found that, on two of the\ndatasets studied, small window sizes led to the lowest prediction errors for\nall of the moving window methods studied. On the majority of datasets studied,\nthe RF-PLS calibration method offered the lowest one-step-ahead prediction\nerrors compared to those of the other methods, and it demonstrated greater\npredictive stability at larger time delays than moving window PLS alone. It was\nfound that both the random forest and RF-PLS methods most adequately modeled\nthe datasets that did not feature purely monotonic increases in property\nvalues, but that both methods performed more poorly than moving window PLS\nmodels on one dataset with purely monotonic property values. Other data\ndependent findings are presented and discussed.\n",
"title": "Small Moving Window Calibration Models for Soft Sensing Processes with Limited History"
}
| null | null | null | null | true | null |
11624
| null |
Default
| null | null |
null |
{
"abstract": " We have developed a new data-driven paradigm for the rapid inference,\nmodeling and simulation of the physics of transport phenomena by deep learning.\nUsing conditional generative adversarial networks (cGAN), we train models for\nthe direct generation of solutions to steady state heat conduction and\nincompressible fluid flow purely on observation without knowledge of the\nunderlying governing equations. Rather than using iterative numerical methods\nto approximate the solution of the constitutive equations, cGANs learn to\ndirectly generate the solutions to these phenomena, given arbitrary boundary\nconditions and domain, with high test accuracy (MAE$<$1\\%) and state-of-the-art\ncomputational performance. The cGAN framework can be used to learn causal\nmodels directly from experimental observations where the underlying physical\nmodel is complex or unknown.\n",
"title": "Deep Learning the Physics of Transport Phenomena"
}
| null | null | null | null | true | null |
11625
| null |
Default
| null | null |
null |
{
"abstract": " Deep neural networks are the state-of-the-art methods for many real-world\ntasks, such as computer vision, natural language processing and speech\nrecognition. For all its popularity, deep neural networks are also criticized\nfor consuming a lot of memory and draining battery life of devices during\ntraining and inference. This makes it hard to deploy these models on mobile or\nembedded devices which have tight resource constraints. Quantization is\nrecognized as one of the most effective approaches to satisfy the extreme\nmemory requirements that deep neural network models demand. Instead of adopting\n32-bit floating point format to represent weights, quantized representations\nstore weights using more compact formats such as integers or even binary\nnumbers. Despite a possible degradation in predictive performance, quantization\nprovides a potential solution to greatly reduce the model size and the energy\nconsumption. In this survey, we give a thorough review of different aspects of\nquantized neural networks. Current challenges and trends of quantized neural\nnetworks are also discussed.\n",
"title": "A Survey on Methods and Theories of Quantized Neural Networks"
}
| null | null | null | null | true | null |
11626
| null |
Default
| null | null |
null |
{
"abstract": " In the present note we consider the problem of constructing honest and\nadaptive confidence sets for the matrix completion problem. For the Bernoulli\nmodel with known variance of the noise we provide a realizable method for\nconstructing confidence sets that adapt to the unknown rank of the true matrix.\n",
"title": "Constructing confidence sets for the matrix completion problem"
}
| null | null | null | null | true | null |
11627
| null |
Default
| null | null |
null |
{
"abstract": " This is the second companion paper of arXiv:1601.03586. We consider the\nmorphism from the variety of triples introduced in arXiv:1601.03586 to the\naffine Grassmannian. The direct image of the dualizing complex is a ring object\nin the equivariant derived category on the affine Grassmannian (equivariant\nderived Satake category). We show that various constructions in\narXiv:1601.03586 work for an arbitrary commutative ring object.\nThe second purpose of this paper is to study Coulomb branches associated with\nstar shaped quivers, which are expected to be conjectural Higgs branches of\n$3d$ Sicilian theories in type $A$ by arXiv:1007.0992.\n",
"title": "Ring objects in the equivariant derived Satake category arising from Coulomb branches (with an appendix by Gus Lonergan)"
}
| null | null | null | null | true | null |
11628
| null |
Default
| null | null |
null |
{
"abstract": " We consider a two-phase flow of two incompressible, viscous and immiscible\nfluids which are separated by a sharp interface in the case of a simple phase\ntransition. In this model the interface is no longer material and its evolution\nis governed by a convective mean curvature flow equation, which is coupled to a\ntwo-phase Navier-Stokes equation with Young-Laplace law. The problem arises as\na sharp interface limit of a diffuse interface model, which consists of a\nNavier-Stokes system coupled with an Allen-Cahn equation. We prove existence of\nstrong solutions for sufficiently small times and regular initial data.\n",
"title": "Well-Posedness of a Navier-Stokes/Mean Curvature Flow system"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11629
| null |
Validated
| null | null |
null |
{
"abstract": " MOBAs represent a huge segment of online gaming and are growing as both an\neSport and a casual genre. The natural starting point for AI researchers\ninterested in MOBAs is to develop an AI to play the game better than a human -\nbut MOBAs have many more challenges besides adversarial AI. In this paper we\nintroduce the reader to the wider context of MOBA culture, propose a range of\nchallenges faced by the community today, and posit concrete AI projects that\ncan be undertaken to begin solving them.\n",
"title": "Off The Beaten Lane: AI Challenges In MOBAs Beyond Player Control"
}
| null | null | null | null | true | null |
11630
| null |
Default
| null | null |
null |
{
"abstract": " We study two-player inclusion games played over word-generating higher-order\nrecursion schemes. While inclusion checks are known to capture verification\nproblems, two-player games generalize this relationship to program synthesis.\nIn such games, non-terminals of the grammar are controlled by opposing players.\nThe goal of the existential player is to avoid producing a word that lies\noutside of a regular language of safe words.\nWe contribute a new domain that provides a representation of the winning\nregion of such games. Our domain is based on (functions over) potentially\ninfinite Boolean formulas with words as atomic propositions. We develop an\nabstract interpretation framework that we instantiate to abstract this domain\ninto a domain where the propositions are replaced by states of a finite\nautomaton. This second domain is therefore finite and we obtain, via standard\nfixed-point techniques, a direct algorithm for the analysis of two-player\ninclusion games. We show, via a second instantiation of the framework, that our\nfinite domain can be optimized, leading to a (k+1)EXP algorithm for order-k\nrecursion schemes. We give a matching lower bound, showing that our approach is\noptimal. Since our approach is based on standard Kleene iteration, existing\ntechniques and tools for fixed-point computations can be applied.\n",
"title": "Domains for Higher-Order Games"
}
| null | null |
[
"Computer Science"
] | null | true | null |
11631
| null |
Validated
| null | null |
null |
{
"abstract": " The migration of planets on nearly circular, non-inclined orbits in\nprotoplanetary discs is entirely described by the disc's torque. This torque is\na complex function of the disc parameters, and essentially amounts to the sum\nof two components: the Lindblad torque and the corotation torque. Known torque\nformulae do not reproduce accurately the torque actually experienced in\nnumerical simulations by low- and intermediate- mass planets in radiative\ndiscs. One of the main reasons for this inaccuracy is that these formulae have\nbeen worked out in two-dimensional analyses. Here we revisit the torque formula\nand update many of its dimensionless coefficients by means of tailored, three-\ndimensional numerical simulations. In particular, we derive the dependence of\nthe Lindblad torque on the temperature gradient, the dependence of the\ncorotation torque on the radial entropy gradient (and work out a suitable\nexpression of this gradient in a three-dimensional disc). We also work out the\ndependence of the corotation torque on the radial temperature gradient,\noverlooked so far. Corotation torques are known to scale very steeply with the\nwidth of the horseshoe region. We extend the expression of this width to the\ndomain of intermediate mass planets, so that our updated torque formula remains\nvalid for planets up to typically several tens of Earth masses, provided these\nrelatively massive planets do not significantly deplete their coorbital region.\nOur torque expression can be applied to low- and intermediate-mass planets in\noptically thick protoplanetary discs, as well as protomoons embedded in\ncircumplanetary discs.\n",
"title": "Improved torque formula for low and intermediate mass planetary migration"
}
| null | null | null | null | true | null |
11632
| null |
Default
| null | null |
null |
{
"abstract": " We have researched the motion of gas in the subnanochannel with functional\nsurface which wettability has a gradient for the fluid by using molecular\ndynamics simulation. The results show that the gas is driven to flow under a\nsingle heat source and without any other work or energy applied to the system.\nThe driving source is owed to the potential gradient of the functional face\nwhich keeps the fluid running in the subnanochannel. The width of the channel\nand the pressure of the reservoir has a significant influence on the flow\nvelocity, which, respectively, has an optimal value for the maximum velocity.\nThe flow velocity grows with the increasing temperature.\n",
"title": "The fluid running in the subnanochannel with functional surface"
}
| null | null |
[
"Physics"
] | null | true | null |
11633
| null |
Validated
| null | null |
null |
{
"abstract": " The implementation of optimal power flow (OPF) methods to perform voltage and\npower flow regulation in electric networks is generally believed to require\ncommunication. We consider distribution systems with multiple controllable\nDistributed Energy Resources (DERs) and present a data-driven approach to learn\ncontrol policies for each DER to reconstruct and mimic the solution to a\ncentralized OPF problem from solely locally available information.\nCollectively, all local controllers closely match the centralized OPF solution,\nproviding near-optimal performance and satisfaction of system constraints. A\nrate distortion framework facilitates the analysis of how well the resulting\nfully decentralized control policies are able to reconstruct the OPF solution.\nOur methodology provides a natural extension to decide what buses a DER should\ncommunicate with to improve the reconstruction of its individual policy. The\nmethod is applied on both single- and three-phase test feeder networks using\ndata from real loads and distributed generators. It provides a framework for\nDistribution System Operators to efficiently plan and operate the contributions\nof DERs to active distribution networks.\n",
"title": "Data-Driven Decentralized Optimal Power Flow"
}
| null | null | null | null | true | null |
11634
| null |
Default
| null | null |
null |
{
"abstract": " We present theoretical calculations to interpret optical and mechanical\nproperties of Ag@Fe3O4 nanoflowers. The microstructures and nature of optical\npeaks of nanoflowers are determined by means of the Mie theory associated with\neffective dielectric approximation and the experimental absorption spectrum.\nUnder laser illumination, the thermal strain fields inside and outside the\nstructure due to the absorbed optical energy are studied using continuum\nmechanics approach. Our findings provide simple but comprehensive description\nof the elastic behaviors of previous experiments.\n",
"title": "Thermal-induced stress of plasmonic magnetic nanocomposites"
}
| null | null |
[
"Physics"
] | null | true | null |
11635
| null |
Validated
| null | null |
null |
{
"abstract": " The photoelectron spectrum of water has been recorded in the vicinity of the\n${\\mathrm {\\tilde A^+}}$ $\\leftarrow$ $\\tilde{\\mathrm{X}}$ transition between\n112 000 and 116 000 cm$^{-1}$ (13.89-14.38 eV). The high-resolution allowed the\nobservation of the rotational structure of several bands. Rotational\nassignments of the transitions involving the $\\Pi(080)$, $\\Sigma(070)$ and\n$\\Pi(060)$ vibronic states of the $\\tilde{\\mathrm{A}}^+$ electronic state are\ndeduced from previous studies of the $\\tilde{\\mathrm{A}}^+ -\n\\tilde{\\mathrm{X}}^+$ band system of H$_2$O$^+$ (Lew, Can. J. Phys. 54, 2028\n(1976) and Huet et al., J. Chem. Phys. 107, 5645 (1997)) and photoionization\nselection rules. The transition to the $\\Sigma(030)$ vibronic state is\ntentatively assigned.\n",
"title": "High-resolution photoelectron-spectroscopic investigation of the H$_2$O$^+$ cation in its ${\\mathrm {\\tilde A^+}}$ electronic state"
}
| null | null | null | null | true | null |
11636
| null |
Default
| null | null |
null |
{
"abstract": " By virtue of Balmer's celebrated theorem, the classification of thick tensor\nideals of a tensor triangulated category $\\T$ is equivalent to the topological\nstructure of its Balmer spectrum $\\spc \\T$. Motivated by this theorem, we\ndiscuss connectedness and noetherianity of the Balmer spectrum of a right\nbounded derived category of finitely generated modules over a commutative ring.\n",
"title": "Connectedness of the Balmer spectra of right bounded derived categories"
}
| null | null | null | null | true | null |
11637
| null |
Default
| null | null |
null |
{
"abstract": " We study the underdamped Langevin diffusion when the log of the target\ndistribution is smooth and strongly concave. We present a MCMC algorithm based\non its discretization and show that it achieves $\\varepsilon$ error (in\n2-Wasserstein distance) in $\\mathcal{O}(\\sqrt{d}/\\varepsilon)$ steps. This is a\nsignificant improvement over the best known rate for overdamped Langevin MCMC,\nwhich is $\\mathcal{O}(d/\\varepsilon^2)$ steps under the same\nsmoothness/concavity assumptions.\nThe underdamped Langevin MCMC scheme can be viewed as a version of\nHamiltonian Monte Carlo (HMC) which has been observed to outperform overdamped\nLangevin MCMC methods in a number of application areas. We provide quantitative\nrates that support this empirical wisdom.\n",
"title": "Underdamped Langevin MCMC: A non-asymptotic analysis"
}
| null | null | null | null | true | null |
11638
| null |
Default
| null | null |
null |
{
"abstract": " Direct numerical simulation is performed to study compressible, viscous flow\naround a circular cylinder. The present study considers two-dimensional,\nshock-free continuum flow by varying the Reynolds number between 20 and 100 and\nthe freestream Mach number between 0 and 0.5. The results indicate that\ncompressibility effects elongate the near wake for cases above and below the\ncritical Reynolds number for two-dimensional flow under shedding. The wake\nelongation becomes more pronounced as the Reynolds number approaches this\ncritical value. Moreover, we determine the growth rate and frequency of linear\ninstability for cases above the critical Reynolds number. From the analysis, it\nis observed that the frequency of the Bénard-von Kármán vortex street in\nthe time-periodic, fully-saturated flow increases from the dominant unstable\nfrequency found from the linear stability analysis as the Reynolds number\nincreases from its critical value, even for the low range of Reynolds numbers\nconsidered. We also notice that the compressibility effects reduce the growth\nrate and dominant frequency in the linear growth stage. Semi-empirical\nfunctional relationships for the growth rate and the dominant frequency in\nlinearized flow around the cylinder in terms of the Reynolds number and\nfreestream Mach number are presented.\n",
"title": "Two-dimensional compressible viscous flow around a circular cylinder"
}
| null | null | null | null | true | null |
11639
| null |
Default
| null | null |
null |
{
"abstract": " The development of spiking neural network simulation software is a critical\ncomponent enabling the modeling of neural systems and the development of\nbiologically inspired algorithms. Existing software frameworks support a wide\nrange of neural functionality, software abstraction levels, and hardware\ndevices, yet are typically not suitable for rapid prototyping or application to\nproblems in the domain of machine learning. In this paper, we describe a new\nPython package for the simulation of spiking neural networks, specifically\ngeared towards machine learning and reinforcement learning. Our software,\ncalled BindsNET, enables rapid building and simulation of spiking networks and\nfeatures user-friendly, concise syntax. BindsNET is built on top of the PyTorch\ndeep neural networks library, enabling fast CPU and GPU computation for large\nspiking networks. The BindsNET framework can be adjusted to meet the needs of\nother existing computing and hardware environments, e.g., TensorFlow. We also\nprovide an interface into the OpenAI gym library, allowing for training and\nevaluation of spiking networks on reinforcement learning problems. We argue\nthat this package facilitates the use of spiking networks for large-scale\nmachine learning experimentation, and show some simple examples of how we\nenvision BindsNET can be used in practice. BindsNET code is available at\nthis https URL\n",
"title": "BindsNET: A machine learning-oriented spiking neural networks library in Python"
}
| null | null |
[
"Quantitative Biology"
] | null | true | null |
11640
| null |
Validated
| null | null |
null |
{
"abstract": " We propose an algorithm to separate simultaneously speaking persons from each\nother, the \"cocktail party problem\", using a single microphone. Our approach\ninvolves a deep recurrent neural networks regression to a vector space that is\ndescriptive of independent speakers. Such a vector space can embed empirically\ndetermined speaker characteristics and is optimized by distinguishing between\nspeaker masks. We call this technique source-contrastive estimation. The\nmethodology is inspired by negative sampling, which has seen success in natural\nlanguage processing, where an embedding is learned by correlating and\nde-correlating a given input vector with output weights. Although the matrix\ndetermined by the output weights is dependent on a set of known speakers, we\nonly use the input vectors during inference. Doing so will ensure that source\nseparation is explicitly speaker-independent. Our approach is similar to recent\ndeep neural network clustering and permutation-invariant training research; we\nuse weighted spectral features and masks to augment individual speaker\nfrequencies while filtering out other speakers. We avoid, however, the severe\ncomputational burden of other approaches with our technique. Furthermore, by\ntraining a vector space rather than combinations of different speakers or\ndifferences thereof, we avoid the so-called permutation problem during\ntraining. Our algorithm offers an intuitive, computationally efficient response\nto the cocktail party problem, and most importantly boasts better empirical\nperformance than other current techniques.\n",
"title": "Monaural Audio Speaker Separation with Source Contrastive Estimation"
}
| null | null | null | null | true | null |
11641
| null |
Default
| null | null |
null |
{
"abstract": " The quest for biologically plausible deep learning is driven, not just by the\ndesire to explain experimentally-observed properties of biological neural\nnetworks, but also by the hope of discovering more efficient methods for\ntraining artificial networks. In this paper, we propose a new algorithm named\nVariational Probably Flow (VPF), an extension of minimum probability flow for\ntraining binary Deep Boltzmann Machines (DBMs). We show that weight updates in\nVPF are local, depending only on the states and firing rates of the adjacent\nneurons. Unlike contrastive divergence, there is no need for Gibbs\nconfabulations; and unlike backpropagation, alternating feedforward and\nfeedback phases are not required. Moreover, the learning algorithm is effective\nfor training DBMs with intra-layer connections between the hidden nodes.\nExperiments with MNIST and Fashion MNIST demonstrate that VPF learns reasonable\nfeatures quickly, reconstructs corrupted images more accurately, and generates\nsamples with a high estimated log-likelihood. Lastly, we note that,\ninterestingly, if an asymmetric version of VPF exists, the weight updates\ndirectly explain experimental results in Spike-Timing-Dependent Plasticity\n(STDP).\n",
"title": "Variational Probability Flow for Biologically Plausible Training of Deep Neural Networks"
}
| null | null | null | null | true | null |
11642
| null |
Default
| null | null |
null |
{
"abstract": " Signed networks are a crucial tool when modeling friend and foe\nrelationships. In contrast to classical undirected, weighted graphs, the edge\nweights for signed graphs are positive and negative. Crucial network properties\nare often derived from the study of the associated graph Laplacians. We here\nstudy several different signed network Laplacians with a focus on the task of\nclassifying the nodes of the graph. We here extend a recently introduced\ntechnique based on a partial differential equation defined on the signed\nnetwork, namely the Allen-Cahn-equation, to classify the nodes into two or more\nclasses. We illustrate the performance of this approach on several real-world\nnetworks.\n",
"title": "Node classification for signed networks using diffuse interface methods"
}
| null | null | null | null | true | null |
11643
| null |
Default
| null | null |
null |
{
"abstract": " Recently we proposed a general, ensemble-based feature engineering wrapper\n(FEW) that was paired with a number of machine learning methods to solve\nregression problems. Here, we adapt FEW for supervised classification and\nperform a thorough analysis of fitness and survival methods within this\nframework. Our tests demonstrate that two fitness metrics, one introduced as an\nadaptation of the silhouette score, outperform the more commonly used Fisher\ncriterion. We analyze survival methods and demonstrate that $\\epsilon$-lexicase\nsurvival works best across our test problems, followed by random survival which\noutperforms both tournament and deterministic crowding. We conduct a benchmark\ncomparison to several classification methods using a large set of problems and\nshow that FEW can improve the best classifier performance in several cases. We\nshow that FEW generates consistent, meaningful features for a biomedical\nproblem with different ML pairings.\n",
"title": "Ensemble representation learning: an analysis of fitness and survival for wrapper-based genetic programming methods"
}
| null | null | null | null | true | null |
11644
| null |
Default
| null | null |
null |
{
"abstract": " TaSb$_{2}$ has been predicted theoretically and proposed through\nmagnetotransport experiment to be a topological semimetal. In earlier reports,\nthe Shubnikov-de Haas oscillation has been analyzed to probe the Fermi surface,\nwith magnetic field along a particular crystallographic axis only. By employing\na sample rotator, we reveal highly anisotropic transverse magnetoresistance by\nrotating the magnetic field along different crystallographic directions. To\nprobe the anisotropy in the Fermi surface, we have performed magnetization\nmeasurements and detected strong de Haas-van Alphen (dHvA) oscillations for the\nmagnetic field applied along \\textbf{b} and \\textbf{c} axes as well as\nperpendicular to \\textbf{bc} plane of the crystals. Three Fermi pockets have\nbeen identified by analyzing the dHvA oscillations. Hall measurement reveals\nelectron as the only charge carrier, i.e., all the three Fermi pockets are\nelectron type. With the application of magnetic field along different crystal\ndirections, the cross sectional areas of the Fermi pockets have been found\nsignificantly different. Other physical parameters, such as the effective mass\nof the charge carrier and Fermi velocity have also been calculated using the\nLifshitz-Kosevich formula.\n",
"title": "Anisotropic Fermi surface probed by the de Haas-van Alphen oscillation in proposed Dirac Semimetal TaSb$_{2}$"
}
| null | null | null | null | true | null |
11645
| null |
Default
| null | null |
null |
{
"abstract": " The training of Generative Adversarial Networks is a difficult task mainly\ndue to the nature of the networks. One such issue is when the generator and\ndiscriminator start oscillating, rather than converging to a fixed point.\nAnother case can be when one agent becomes more adept than the other which\nresults in the decrease of the other agent's ability to learn, reducing the\nlearning capacity of the system as a whole. Additionally, there exists the\nproblem of Mode Collapse which involves the generators output collapsing to a\nsingle sample or a small set of similar samples. To train GANs a careful\nselection of the architecture that is used along with a variety of other\nmethods to improve training. Even when applying these methods there is low\nstability of training in relation to the parameters that are chosen. Stochastic\nensembling is suggested as a method for improving the stability while training\nGANs.\n",
"title": "Stochastic Deconvolutional Neural Network Ensemble Training on Generative Pseudo-Adversarial Networks"
}
| null | null | null | null | true | null |
11646
| null |
Default
| null | null |
null |
{
"abstract": " This work is concerned with the prime factor decomposition (PFD) of strong\nproduct graphs. A new quasi-linear time algorithm for the PFD with respect to\nthe strong product for arbitrary, finite, connected, undirected graphs is\nderived. Moreover, since most graphs are prime although they can have a\nproduct-like structure, also known as approximate graph products, the practical\napplication of the well-known \"classical\" prime factorization algorithm is\nstrictly limited. This new PFD algorithm is based on a local approach that\ncovers a graph by small factorizable subgraphs and then utilizes this\ninformation to derive the global factors. Therefore, we can take advantage of\nthis approach and derive in addition a method for the recognition of\napproximate graph products.\n",
"title": "A Local Prime Factor Decomposition Algorithm for Strong Product Graphs"
}
| null | null |
[
"Computer Science"
] | null | true | null |
11647
| null |
Validated
| null | null |
null |
{
"abstract": " The purpose of this note is to prove dispersive estimates for the wave\nequation outside a ball in R^d. If d = 3, we show that the linear flow\nsatisfies the dispersive estimates as in R^3. In higher dimensions d $\\ge$ 4 we\nshow that losses in dispersion do appear and this happens at the Poisson spot.\n",
"title": "Dispersion for the wave equation outside a ball and counterexamples"
}
| null | null |
[
"Mathematics"
] | null | true | null |
11648
| null |
Validated
| null | null |
null |
{
"abstract": " If $M$ is a finite volume complete hyperbolic $3$-manifold, the quantity\n$\\mathcal A_1(M)$ is defined as the infimum of the areas of closed minimal\nsurfaces in $M$. In this paper we study the continuity property of the\nfunctional $\\mathcal A_1$ with respect to the geometric convergence of\nhyperbolic manifolds. We prove that it is lower semi-continuous and even\ncontinuous if $\\mathcal A_1(M)$ is realized by a minimal surface satisfying\nsome hypotheses. Understanding the interaction between minimal surfaces and\nshort geodesics in $M$ is the main theme of this paper\n",
"title": "Minimal surfaces near short geodesics in hyperbolic $3$-manifolds"
}
| null | null | null | null | true | null |
11649
| null |
Default
| null | null |
null |
{
"abstract": " Solids deform and fluids flow, but soft glassy materials, such as emulsions,\nfoams, suspensions, and pastes, exhibit an intricate mix of solid and\nliquid-like behavior. While much progress has been made to understand their\nelastic (small strain) and flow (infinite strain) properties, such\nunderstanding is lacking for the softening and yielding phenomena that connect\nthese asymptotic regimes. Here we present a comprehensive framework for\nsoftening and yielding of soft glassy materials, based on extensive numerical\nsimulations of oscillatory rheological tests, and show that two distinct\nscenarios unfold depending on the material's packing density. For dense\nsystems, there is a single, pressure-independent strain where the elastic\nmodulus drops and the particle motion becomes diffusive. In contrast, for\nweakly jammed systems, a two-step process arises: at an intermediate softening\nstrain, the elastic and loss moduli both drop down and then reach a new plateau\nvalue, whereas the particle motion becomes diffusive at the distinctly larger\nyield strain. We show that softening is associated with an extensive number of\nmicroscopic contact changes leading to a non-analytic rheological signature.\nMoreover, the scaling of the softening strain with pressure suggest the\nexistence of a novel pressure scale above which softening and yielding\ncoincide, and we verify the existence of this crossover scale numerically. Our\nfindings thus evidence the existence of two distinct classes of soft glassy\nmaterials -- jamming dominated and dense -- and show how these can be\ndistinguished by their rheological fingerprint.\n",
"title": "Softening and Yielding of Soft Glassy Materials"
}
| null | null | null | null | true | null |
11650
| null |
Default
| null | null |
null |
{
"abstract": " We describe the technical effort used to process a voluminous high value\nhuman neuroimaging dataset on the Open Science Grid with opportunistic use of\nidle HPC resources to boost computing capacity more than 5-fold. With minimal\nsoftware development effort and no discernable competitive interference with\nother HPC users, this effort delivered 15,000,000 core hours over 7 months.\n",
"title": "Seemless Utilization of Heterogeneous XSede Resources to Accelerate Processing of a High Value Functional Neuroimaging Dataset"
}
| null | null | null | null | true | null |
11651
| null |
Default
| null | null |
null |
{
"abstract": " We show that the recently introduced iterative backflow renormalization can\nbe interpreted as a general neural network in continuum space with non-linear\nfunctions in the hidden units. We use this wave function within Variational\nMonte Carlo for liquid $^4$He in two and three dimensions, where we typically\nfind a tenfold increase in accuracy over currently used wave functions.\nFurthermore, subsequent stages of the iteration procedure define a set of\nincreasingly good wave functions, each with its own variational energy and\nvariance of the local energy: extrapolation of these energies to zero variance\ngives values in close agreement with the exact values. For two dimensional\n$^4$He, we also show that the iterative backflow wave function can describe\nboth the liquid and the solid phase with the same functional form -a feature\nshared with the Shadow Wave Function, but now joined by much higher accuracy.\nWe also achieve significant progress for liquid $^3$He in three dimensions,\nimproving previous variational and fixed-node energies for this very\nchallenging fermionic system.\n",
"title": "Nonlinear Network description for many-body quantum systems in continuous space"
}
| null | null | null | null | true | null |
11652
| null |
Default
| null | null |
null |
{
"abstract": " We prove that for any winding number $m>0$ pattern $P$ and winding number\n$-m$ pattern $Q$, there exist knots $K$ such that the minimal genus of a\ncobordism between $P(K)$ and $Q(K)$ is arbitrarily large. This answers a\nquestion posed by Cochran-Harvey [CH17] and generalizes a result of\nKim-Livingston [KL05].\n",
"title": "Winding number $m$ and $-m$ patterns acting on concordance"
}
| null | null | null | null | true | null |
11653
| null |
Default
| null | null |
null |
{
"abstract": " Let $A_f(1,n)$ be the normalized Fourier coefficients of a $GL(3)$ Maass cusp\nform $f$ and let $a_g(n)$ be the normalized Fourier coefficients of a $GL(2)$\ncusp form $g$. Let $\\lambda(n)$ be either $A_f(1,n)$ or the triple divisor\nfunction $d_3(n)$. It is proved that for any $\\epsilon>0$, any integer $r\\geq\n1$ and $r^{5/2}X^{1/4+7\\delta/2}\\leq H\\leq X$ with $\\delta>0$, $$\n\\frac{1}{H}\\sum_{h\\geq 1}W\\left(\\frac{h}{H}\\right) \\sum_{n\\geq\n1}\\lambda(n)a_g(rn+h)V\\left(\\frac{n}{X}\\right)\\ll X^{1-\\delta+\\epsilon}, $$\nwhere $V$ and $W$ are smooth compactly supported functions, and the implied\nconstants depend only on the associated forms and $\\epsilon$.\n",
"title": "Averages of shifted convolution sums for $GL(3) \\times GL(2)$"
}
| null | null | null | null | true | null |
11654
| null |
Default
| null | null |
null |
{
"abstract": " The present letter to the editor is one in a series of publications\ndiscussing the formulation of hypotheses (propositions) for the evaluation of\nstrength of forensic evidence. In particular, the discussion focusses on the\nissue of what information may be used to define the relevant population\nspecified as part of the different-speaker hypothesis in forensic voice\ncomparison. The previous publications in the series are: Hicks et al. 2015\n<this http URL>; Morrison et al. (2016)\n<this http URL>; Hicks et al. (2017)\n<this http URL>. The latter letter to the\neditor mostly resolves the apparent disagreement between the two groups of\nauthors. We briefly discuss one outstanding point of apparent disagreement, and\nattempt to correct a misinterpretation of our earlier remarks. We believe that\nat this point there is no actual disagreement, and that both groups of authors\nare calling for greater collaboration in order to reduce the likelihood of\nfuture misunderstandings.\n",
"title": "Reply to Hicks et al 2017, Reply to Morrison et al 2016 Refining the relevant population in forensic voice comparison, Reply to Hicks et al 2015 The importance of distinguishing info from evidence/observations when formulating propositions"
}
| null | null | null | null | true | null |
11655
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we give explicit expressions of differential-difference\noperators appeared in the hypothesis of the general Fourier multiplier theorem\nassociated to the Heisenberg groups proved by Mauceri and De Micheal for one\ndimension and C. Lin for higher dimension. We also give a much shorter proof of\nthe above-mentioned theorem. Then we obtain a sharp weighted estimate for\nFourier multipliers on the Heisenberg groups.\n",
"title": "Fourier Multipliers on the Heisenberg groups revisited"
}
| null | null | null | null | true | null |
11656
| null |
Default
| null | null |
null |
{
"abstract": " We construct a new family of high genus examples of free boundary minimal\nsurfaces in the Euclidean unit 3-ball by desingularizing the intersection of a\ncoaxial pair of a critical catenoid and an equatorial disk. The surfaces are\nconstructed by singular perturbation methods and have three boundary\ncomponents. They are the free boundary analogue of the Costa-Hoffman-Meeks\nsurfaces and the surfaces constructed by Kapouleas by desingularizing coaxial\ncatenoids and planes. It is plausible that the minimal surfaces we constructed\nhere are the same as the ones obtained recently by Ketover using the min-max\nmethod.\n",
"title": "Free Boundary Minimal Surfaces in the Unit Three-Ball via Desingularization of the Critical Catenoid and the Equatorial Disk"
}
| null | null | null | null | true | null |
11657
| null |
Default
| null | null |
null |
{
"abstract": " This thesis presents the design, analysis, and validation of a hierarchical\ntransactive control system that engages demand response resources to enhance\nthe integration of renewable electricity generation resources. This control\nsystem joins energy, capacity and regulation markets together in a unified\nhomeostatic and economically efficient electricity operation that increases\ntotal surplus while improving reliability and decreasing carbon emissions from\nfossil-based generation resources.\nThe work encompasses: (1) the derivation of a short-term demand response\nmodel suitable for transactive control systems and its validation with field\ndemonstration data; (2) an aggregate load model that enables effective control\nof large populations of thermal loads using a new type of thermostat (discrete\ntime with zero deadband); (3) a methodology for optimally controlling response\nto frequency deviations while tracking schedule area exports in areas that have\nhigh penetration of both intermittent renewable resources and fast-acting\ndemand response; and (4) the development of a system-wide (continental\ninterconnection) scale strategy for optimal power trajectory and resource\ndispatch based on a shift from primarily energy cost-based approach to a\nprimarily ramping cost-based one.\nThe results show that multi-layer transactive control systems can be\nconstructed, will enhance renewable resource utilization, and will operate in a\ncoordinated manner with bulk power systems that include both regions with and\nwithout organized power markets. Estimates of Western Electric Coordionating\nCouncil (WECC) system cost savings under target renewable energy generation\nlevels resulting from the proposed system exceed US$150B annually by the year\n2024, when compared to the existing control system.\n",
"title": "Multi-scale Transactive Control In Interconnected Bulk Power Systems Under High Renewable Energy Supply and High Demand Response Scenarios"
}
| null | null | null | null | true | null |
11658
| null |
Default
| null | null |
null |
{
"abstract": " Recent results have suggested that active galactic nuclei (AGN) could provide\nenough photons to reionise the Universe. We assess the viability of this\nscenario using a semi-numerical framework for modeling reionisation, to which\nwe add a quasar contribution by constructing a Quasar Halo Occupation\nDistribution (QHOD) based on Giallongo et al. observations. Assuming a constant\nQHOD, we find that an AGN-only model cannot simultaneously match observations\nof the optical depth $\\tau_e$, neutral fraction, and ionising emissivity. Such\na model predicts $\\tau_e$ too low by $\\sim 2\\sigma$ relative to Planck\nconstraints, and reionises the Universe at $z\\lesssim 5$. Arbitrarily\nincreasing the AGN emissivity to match these results yields a strong mismatch\nwith the observed ionising emissivity at $z\\sim 5$. If we instead assume a\nredshift-independent AGN luminosity function yielding an emissivity evolution\nlike that assumed in Madau & Haardt model, then we can match $\\tau_e$ albeit\nwith late reionisation, however such evolution is inconsistent with\nobservations at $z\\sim 4-6$ and poorly motivated physically. These results\narise because AGN are more biased towards massive halos than typical reionising\ngalaxies, resulting in stronger clustering and later formation times.\nAGN-dominated models produce larger ionising bubbles that are reflected in\n$\\sim\\times 2$ more 21cm power on all scales. A model with equal parts galaxies\nand AGN contribution is still (barely) consistent with observations, but could\nbe distinguished using next-generation 21cm experiments HERA and SKA-low. We\nconclude that, even with recent claims of more faint AGN than previously\nthought, AGN are highly unlikely to dominate the ionising photon budget for\nreionisation.\n",
"title": "Constraining the contribution of active galactic nuclei to reionisation"
}
| null | null |
[
"Physics"
] | null | true | null |
11659
| null |
Validated
| null | null |
null |
{
"abstract": " Degree ssortativity is the tendency for nodes of high degree (resp.low\ndegree) in a graph to be connected to high degree nodes (resp. to low degree\nones). It is sually quantified by the Pearson correlation coefficient of the\ndegree-degree correlation. Here we extend this concept to account for the\neffect of second neighbours to a given node in a graph. That is, we consider\nthe two-walks degree of a node as the sum of all the degrees of its adjacent\nnodes. The two-walks degree assortativity of a graph is then the Pearson\ncorrelation coefficient of the two-walks degree-degree correlation. We found\nhere analytical expression for this two-walks degree assortativity index as a\nfunction of contributing subgraphs. We then study all the 261,000 connected\ngraphs with 9 nodes and observe the existence of assortative-assortative and\ndisassortative-disassortative graphs according to degree and two-walks degree,\nrespectively. More surprinsingly, we observe a class of graphs which are degree\ndisassortative and two-walks degree assortative. We explain the existence of\nsome of these graphs due to the presence of certain topological features, such\nas a node of low-degree connected to high-degree ones. More importantly, we\nstudy a series of 49 real-world networks, where we observe the existence of the\ndisassortative-assortative class in several of them. In particular, all\nbiological networks studied here were in this class. We also conclude that no\ngraphs/networks are possible with assortative-disassortative structure.\n",
"title": "Two-walks degree assortativity in graphs and networks"
}
| null | null | null | null | true | null |
11660
| null |
Default
| null | null |
null |
{
"abstract": " We propose a unified framework for establishing existence of nonparametric\nM-estimators, computing the corresponding estimates, and proving their strong\nconsistency when the class of functions is exceptionally rich. In particular,\nthe framework addresses situations where the class of functions is complex\ninvolving information and assumptions about shape, pointwise bounds, location\nof modes, height at modes, location of level-sets, values of moments, size of\nsubgradients, continuity, distance to a `prior' function, multivariate total\npositivity, and any combination of the above. The class might be engineered to\nperform well in a specific setting even in the presence of little data. The\nframework views the class of functions as a subset of a particular metric space\nof upper semicontinuous functions under the Attouch-Wets distance. In addition\nto allowing a systematic treatment of numerous M-estimators, the framework\nyields consistency of plug-in estimators of modes of densities, maximizers of\nregression functions, and related quantities, and also enables computation by\nmeans of approximating parametric classes. We establish consistency through a\none-sided law of large numbers, here extended to sieves, that relaxes\nassumptions of uniform laws, while ensuring global approximations even under\nmodel misspecification.\n",
"title": "Variational Analysis of Constrained M-Estimators"
}
| null | null | null | null | true | null |
11661
| null |
Default
| null | null |
null |
{
"abstract": " Aggregate analysis, such as comparing country-wise sales versus global market\nshare across product categories, is often complicated by the unavailability of\ncommon join attributes, e.g., category, across diverse datasets from different\ngeographies or retail chains, even after disparate data is technically ingested\ninto a common data lake. Sometimes this is a missing data issue, while in other\ncases it may be inherent, e.g., the records in different geographical databases\nmay actually describe different product 'SKUs', or follow different norms for\ncategorization. Record linkage techniques can be used to automatically map\nproducts in different data sources to a common set of global attributes,\nthereby enabling federated aggregation joins to be performed. Traditional\nrecord-linkage techniques are typically unsupervised, relying textual\nsimilarity features across attributes to estimate matches. In this paper, we\npresent an ensemble model combining minimal supervision using Bayesian network\nmodels together with unsupervised textual matching for automating such\n'attribute fusion'. We present results of our approach on a large volume of\nreal-life data from a market-research scenario and compare with a standard\nrecord matching algorithm. Finally we illustrate how attribute fusion using\nmachine learning could be included as a data-lake management feature,\nespecially as our approach also provides confidence values for matches,\nenabling human intervention, if required.\n",
"title": "Minimally-Supervised Attribute Fusion for Data Lakes"
}
| null | null | null | null | true | null |
11662
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we investigate the behavior of the thermoelectric DC\nconductivities in the presence of Weyl corrections with momentum dissipation in\nthe incoherent limit. Moreover, we compute the butterfly velocity and study the\ncharge and energy diffusion with broken translational symmetry. Our results\nshow that the Weyl coupling $\\gamma$, violates the bounds on the charge and\nenergy diffusivity. It is also shown that the Weyl corrections violate the\nbound on the DC electrical conductivity in the incoherent limit.\n",
"title": "Diffusivities bounds in the presence of Weyl corrections"
}
| null | null | null | null | true | null |
11663
| null |
Default
| null | null |
null |
{
"abstract": " Among several quantitative invariants found in evolutionary genomics, one of\nthe most striking is the scaling of the overall abundance of proteins, or\nprotein domains, sharing a specific functional annotation across genomes of\ngiven size. The size of these functional categories change, on average, as\npower-laws in the total number of protein-coding genes. Here, we show that such\nregularities are not restricted to the overall behavior of high-level\nfunctional categories, but also exist systematically at the level of single\nevolutionary families of protein domains. Specifically, the number of proteins\nwithin each family follows family-specific scaling laws with genome size.\nFunctionally similar sets of families tend to follow similar scaling laws, but\nthis is not always the case. To understand this systematically, we provide a\ncomprehensive classification of families based on their scaling properties.\nAdditionally, we develop a quantitative score for the heterogeneity of the\nscaling of families belonging to a given category or predefined group. Under\nthe common reasonable assumption that selection is driven solely or mainly by\nbiological function, these findings point to fine-tuned and interdependent\nfunctional roles of specific protein domains, beyond our current functional\nannotations. This analysis provides a deeper view on the links between\nevolutionary expansion of protein families and the functional constraints\nshaping the gene repertoire of bacterial genomes.\n",
"title": "Family-specific scaling laws in bacterial genomes"
}
| null | null | null | null | true | null |
11664
| null |
Default
| null | null |
null |
{
"abstract": " We present a novel method for obtaining high-quality, domain-targeted\nmultiple choice questions from crowd workers. Generating these questions can be\ndifficult without trading away originality, relevance or diversity in the\nanswer options. Our method addresses these problems by leveraging a large\ncorpus of domain-specific text and a small set of existing questions. It\nproduces model suggestions for document selection and answer distractor choice\nwhich aid the human question generation process. With this method we have\nassembled SciQ, a dataset of 13.7K multiple choice science exam questions\n(Dataset available at this http URL). We demonstrate that the\nmethod produces in-domain questions by providing an analysis of this new\ndataset and by showing that humans cannot distinguish the crowdsourced\nquestions from original questions. When using SciQ as additional training data\nto existing questions, we observe accuracy improvements on real science exams.\n",
"title": "Crowdsourcing Multiple Choice Science Questions"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
11665
| null |
Validated
| null | null |
null |
{
"abstract": " Bulk sensitive hard x-ray photoelectron spectroscopy data of the Ce 3$p$ core\nlevel of CeRu$_4$Sn$_6$ are presented. Using a combination of full multiplet\nand configuration iteration model we were able to obtain an accurate lineshape\nanalysis of the data, thereby taking into account correlations for the strong\nplasmon intensities. We conclude that CeRu$_4$Sn$_6$ is a moderately mixed\nvalence compound with a weight of 8% for the Ce $f^0$ configuration in the\nground state.\n",
"title": "Ce 3$p$ hard x-ray photoelectron spectroscopy study of the topological Kondo insulator CeRu$_4$Sn$_6$"
}
| null | null |
[
"Physics"
] | null | true | null |
11666
| null |
Validated
| null | null |
null |
{
"abstract": " Consider the graph that has as vertices all bitstrings of length $2n+1$ with\nexactly $n$ or $n+1$ entries equal to 1, and an edge between any two bitstrings\nthat differ in exactly one bit. The well-known middle levels conjecture asserts\nthat this graph has a Hamilton cycle for any $n\\geq 1$. In this paper we\npresent a new proof of this conjecture, which is much shorter and more\naccessible than the original proof.\n",
"title": "A short proof of the middle levels theorem"
}
| null | null | null | null | true | null |
11667
| null |
Default
| null | null |
null |
{
"abstract": " After being trained, classifiers must often operate on data that has been\ncorrupted by noise. In this paper, we consider the impact of such noise on the\nfeatures of binary classifiers. Inspired by tools for classifier robustness, we\nintroduce the same classification probability (SCP) to measure the resulting\ndistortion on the classifier outputs. We introduce a low-complexity estimate of\nthe SCP based on quantization and polynomial multiplication. We also study\nchannel coding techniques based on replication error-correcting codes. In\ncontrast to the traditional channel coding approach, where error-correction is\nmeant to preserve the data and is agnostic to the application, our schemes\nspecifically aim to maximize the SCP (equivalently minimizing the distortion of\nthe classifier output) for the same redundancy overhead.\n",
"title": "Don't Fear the Bit Flips: Optimized Coding Strategies for Binary Classification"
}
| null | null | null | null | true | null |
11668
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we find a condition under which a Finsler space with Kropina\nchange of mth-root metric is projectively related to a mth-root metric and also\nwe find a condition under which this Kropina transformed mth-root metric is\nlocally dually flat. Moreover we find the condition for its Projective\nflatness.\n",
"title": "Kropina change of a Finsler space with m-th root metric"
}
| null | null | null | null | true | null |
11669
| null |
Default
| null | null |
null |
{
"abstract": " We grow nearly freestanding single-layer 1T'-WTe2 on graphitized 6H-SiC(0001)\nby using molecular beam epitaxy (MBE), and characterize its electronic\nstructure with scanning tunneling microscopy / spectroscopy (STM/STS). We\ndemonstrate the existence of topological edge states at the periphery of\nsingle-layer WTe2 islands. Surprisingly, we also find a band gap in the bulk\nand the semiconducting behaviors of the single-layer WTe2 at low temperature,\nwhich is likely resulted from an incommensurate charge density wave (CDW)\ntransition. The realization of two-dimensional topological insulators (2D TIs)\nin single-layer transition metal dichalcogenide (TMD) thus provides a promising\nplatform for further exploration of the 2D TIs' physics and related\napplications.\n",
"title": "Direct Visualization of 2D Topological Insulator in Single-layer 1T'-WTe2"
}
| null | null | null | null | true | null |
11670
| null |
Default
| null | null |
null |
{
"abstract": " This paper is concerned with structured machine learning, in a supervised\nmachine learning context. It discusses how to make joint structured learning on\ninterdependent objects of different nature, as well as how to enforce logical\ncon-straints when predicting labels. We explain how this need arose in a\nDocument Understanding task. We then discuss a general extension to Conditional\nRandom Field (CRF) for this purpose and present the contributed open source\nimplementation on top of the open source PyStruct library. We evaluate its\nperformance on a publicly available dataset.\n",
"title": "Joint Structured Learning and Predictions under Logical Constraints in Conditional Random Fields"
}
| null | null | null | null | true | null |
11671
| null |
Default
| null | null |
null |
{
"abstract": " This paper investigates to identify the requirement and the development of\nmachine learning-based mobile big data analysis through discussing the insights\nof challenges in the mobile big data (MBD). Furthermore, it reviews the\nstate-of-the-art applications of data analysis in the area of MBD. Firstly, we\nintroduce the development of MBD. Secondly, the frequently adopted methods of\ndata analysis are reviewed. Three typical applications of MBD analysis, namely\nwireless channel modeling, human online and offline behavior analysis, and\nspeech recognition in the internet of vehicles, are introduced respectively.\nFinally, we summarize the main challenges and future development directions of\nmobile big data analysis.\n",
"title": "Mobile big data analysis with machine learning"
}
| null | null | null | null | true | null |
11672
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the galaxy overdensity around proto-cluster scale quasar pairs\nat high (z>3) and low (z~1) redshift based on the unprecedentedly wide and deep\noptical survey of the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP).\nUsing the first-year survey data covering effectively ~121 deg^2 with the\n5sigma depth of i~26.4 and the SDSS DR12Q catalog, we find two luminous pairs\nat z~3.3 and 3.6 which reside in >5sigma overdense regions of g-dropout\ngalaxies at i<25. The projected separations of the two pairs are R_perp=1.75\nand 1.04 proper Mpc, and their velocity offsets are Delta V=692 and 1448 km\ns^{-1}, respectively. This result is in clear contrast to the average z~4\nquasar environments as discussed in Uchiyama et al. (2017) and implies that the\nquasar activities of the pair members are triggered via major mergers in\nproto-clusters, unlike the vast majority of isolated quasars in general fields\nthat may turn on via non-merger events such as bar and disk instabilities. At\nz~1, we find 37 pairs with R_perp<2 pMpc and Delta V<2300 km s^{-1} in the\ncurrent HSC-Wide coverage, including four from Hennawi et al. (2006). The\ndistribution of the peak overdensity significance within two arcminutes around\nthe pairs has a long tail toward high density (>4sigma) regions. Thanks to the\nlarge sample size, we find a statistical evidence that this excess is unique to\nthe pair environments when compared to single quasar and randomly selected\ngalaxy environments at the same redshift range. Moreover, there are nine\nsmall-scale (R_perp<1 pMpc) pairs, two of which are found to reside in cluster\nfields. Our results demonstrate that <2 pMpc-scale quasar pairs at both\nredshift range tend to occur in massive haloes, although perhaps not the most\nmassive ones, and that they are useful to search for rare density peaks.\n",
"title": "Enhancement of Galaxy Overdensity around Quasar Pairs at z<3.6 based on the Hyper Suprime-Cam Subaru Strategic Program Survey"
}
| null | null | null | null | true | null |
11673
| null |
Default
| null | null |
null |
{
"abstract": " The dark matter search project by means of ultra high purity NaI(Tl)\nscintillator is now underdevelopment. An array of large volume NaI(Tl)\ndetectors whose volume is 12.7 cm$\\phi\\times$12.7 cm is applied to search for\ndark matter signal. To remove radioactive impurities in NaI(Tl) crystal is one\nof the most important task to find small number of dark matter signals. We have\ndeveloped high purity NaI(Tl) crystal which contains small amounts of\nradioactive impurities, $<4$ ppb of $^{nat}$K, 0.3 ppt of Th chain, 58\n$\\mu$Bq/kg of $^{226}$Ra and 30 $\\mu$Bq/kg of $^{210}$Pb. Future prospects to\nsearch for dark matter by means of a large volume and high purity NaI(Tl)\nscintillator is discussed.\n",
"title": "Search for cosmic dark matter by means of ultra high purity NaI(Tl) scintillator"
}
| null | null | null | null | true | null |
11674
| null |
Default
| null | null |
null |
{
"abstract": " Recurrent neural networks have been the dominant models for many speech and\nlanguage processing tasks. However, we understand little about the behavior and\nthe class of functions recurrent networks can realize. Moreover, the heuristics\nused during training complicate the analyses. In this paper, we study recurrent\nnetworks' ability to learn long-term dependency in the context of speech\nrecognition. We consider two decoding approaches, online and batch decoding,\nand show the classes of functions to which the decoding approaches correspond.\nWe then draw a connection between batch decoding and a popular training\napproach for recurrent networks, truncated backpropagation through time.\nChanging the decoding approach restricts the amount of past history recurrent\nnetworks can use for prediction, allowing us to analyze their ability to\nremember. Empirically, we utilize long-term dependency in subphonetic states,\nphonemes, and words, and show how the design decisions, such as the decoding\napproach, lookahead, context frames, and consecutive prediction, characterize\nthe behavior of recurrent networks. Finally, we draw a connection between\nMarkov processes and vanishing gradients. These results have implications for\nstudying the long-term dependency in speech data and how these properties are\nlearned by recurrent networks.\n",
"title": "On Training Recurrent Networks with Truncated Backpropagation Through Time in Speech Recognition"
}
| null | null | null | null | true | null |
11675
| null |
Default
| null | null |
null |
{
"abstract": " The physical mechanisms of the laser-induced periodic surface structures\n(LIPSS) formation are studied in this paper for single-pulse irradiation\nregimes. The change in the LIPSS period with wavelength of incident laser\nradiation is investigated experimentally, using a picosecond laser system,\nwhich provides 7-ps pulses in near-IR, visible, and UV spectral ranges. The\nexperimental results are compared with predictions made under the assumption\nthat the surface-scattered waves are involved in the LIPSS formation.\nConsiderable disagreement suggests that hydrodynamic mechanisms can be\nresponsible for the observed pattern periodicity.\n",
"title": "Wavelength Dependence of Picosecond Laser-Induced Periodic Surface Structures on Copper"
}
| null | null | null | null | true | null |
11676
| null |
Default
| null | null |
null |
{
"abstract": " SSDs are currently replacing magnetic disks in many application areas. A\nchallenge of the underlying flash technology is that data cannot be updated\nin-place. A block consisting of many pages must be completely erased before a\nsingle page can be rewritten. This victim block can still contain valid pages\nwhich need to be copied to other blocks before erasure. The objective of\ngarbage collection strategies is to minimize write amplification induced by\ncopying valid pages from victim blocks while minimizing the performance\noverhead of the victim selection. Victim selection strategies minimizing write\namplification, like the cost-benefit approach, have linear runtime, while the\nwrite amplifications of time-efficient strategies, like the greedy strategy,\nsignificantly reduce the lifetime of SSDs. In this paper, we propose two\nstrategies which optimize the performance of cost-benefit, while (almost)\npreserving its write amplification. Trace-driven simulations for single- and\nmulti-channel SSDs show that the optimizations help to keep the write\namplification low while improving the runtime by up to 24-times compared to the\noriginal cost-benefit strategy, so that the new strategies can be used in\nmulti-TByte SSDs.\n",
"title": "Time-efficient Garbage Collection in SSDs"
}
| null | null |
[
"Computer Science"
] | null | true | null |
11677
| null |
Validated
| null | null |
null |
{
"abstract": " In a recent paper [15], Giardin{à}, Giberti, Hofstad, Prioriello have\nproved a law of large number and a central limit theorem with respect to the\nannealed measure for the magnetization of the Ising model on some random graphs\nincluding the random 2-regular graph. We present a new proof of their results,\nwhich applies to all random regular graphs. In addition, we prove the existence\nof annealed pressure in the case of configuration model random graphs.\n",
"title": "Annealed limit theorems for the ising model on random regular graphs"
}
| null | null | null | null | true | null |
11678
| null |
Default
| null | null |
null |
{
"abstract": " Keyphrase boundary classification (KBC) is the task of detecting keyphrases\nin scientific articles and labelling them with respect to predefined types.\nAlthough important in practice, this task is so far underexplored, partly due\nto the lack of labelled data. To overcome this, we explore several auxiliary\ntasks, including semantic super-sense tagging and identification of multi-word\nexpressions, and cast the task as a multi-task learning problem with deep\nrecurrent neural networks. Our multi-task models perform significantly better\nthan previous state of the art approaches on two scientific KBC datasets,\nparticularly for long keyphrases.\n",
"title": "Multi-Task Learning of Keyphrase Boundary Classification"
}
| null | null | null | null | true | null |
11679
| null |
Default
| null | null |
null |
{
"abstract": " The aim of this article is the construction of the interior motive of a\nPicard variety. Those are Shimura varieties of PEL type. Our result is an\napplication of the strategy developed by Wildeshaus to construct a\nHecke-invariant motive whose realizations correspond to interior cohomology.\n",
"title": "On the interior motive of certain Shimura varieties : the case of Picard varieties"
}
| null | null | null | null | true | null |
11680
| null |
Default
| null | null |
null |
{
"abstract": " We present a new topic model that generates documents by sampling a topic for\none whole sentence at a time, and generating the words in the sentence using an\nRNN decoder that is conditioned on the topic of the sentence. We argue that\nthis novel formalism will help us not only visualize and model the topical\ndiscourse structure in a document better, but also potentially lead to more\ninterpretable topics since we can now illustrate topics by sampling\nrepresentative sentences instead of bag of words or phrases. We present a\nvariational auto-encoder approach for learning in which we use a factorized\nvariational encoder that independently models the posterior over topical\nmixture vectors of documents using a feed-forward network, and the posterior\nover topic assignments to sentences using an RNN. Our preliminary experiments\non two different datasets indicate early promise, but also expose many\nchallenges that remain to be addressed.\n",
"title": "SenGen: Sentence Generating Neural Variational Topic Model"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
11681
| null |
Validated
| null | null |
null |
{
"abstract": " In this article, a few problems related to multiscale modelling of magnetic\nmaterials at finite temperatures and possible ways of solving these problems\nare discussed. The discussion is mainly centred around two established\nmultiscale concepts: the partitioned domain and the upscaling-based\nmethodologies. The major challenge for both multiscale methods is to capture\nthe correct value of magnetisation length accurately, which is affected by a\nrandom temperature-dependent force. Moreover, general limitations of these\nmultiscale techniques in application to spin systems are discussed.\n",
"title": "Atomistic-continuum multiscale modelling of magnetisation dynamics at non-zero temperature"
}
| null | null |
[
"Physics",
"Mathematics"
] | null | true | null |
11682
| null |
Validated
| null | null |
null |
{
"abstract": " We revisit the problem of characterizing the eigenvalue distribution of the\nDirichlet-Laplacian on bounded open sets $\\Omega\\subset\\mathbb{R}$ with fractal\nboundaries. It is well-known from the results of Lapidus and Pomerance\n\\cite{LapPo1} that the asymptotic second term of the eigenvalue counting\nfunction can be described in terms of the Minkowski content of the boundary of\n$\\Omega$ provided it exists. He and Lapidus \\cite{HeLap2} discussed a\nremarkable extension of this characterization to sets $\\Omega$ with boundaries\nthat are not necessarily Minkowski measurable. They employed so-called\ngeneralized Minkowski contents given in terms of gauge functions more general\nthan the usual power functions. The class of valid gauge functions in their\ntheory is characterized by some technical conditions, the geometric meaning and\nnecessity of which is not obvious. Therefore, it is not completely clear how\ngeneral the approach is and which sets $\\Omega$ are covered. Here we revisit\nthese results and put them in the context of regularly varying functions. Using\nKaramata theory, it is possible to get rid of most of the technical conditions\nand simplify the proofs given by He and Lapidus, revealing thus even more of\nthe beauty of their results. Further simplifications arise from\ncharacterization results for Minkowski contents obtained in \\cite{RW13}. We\nhope our new point of view on these spectral problems will initiate some\nfurther investigations of this beautiful theory.\n",
"title": "Regularly Varying Functions, Generalized contents, and the spectrum of fractal strings"
}
| null | null | null | null | true | null |
11683
| null |
Default
| null | null |
null |
{
"abstract": " The $\\mathcal{G}_I^0$ distribution is able to characterize different regions\nin monopolarized SAR imagery. It is indexed by three parameters: the number of\nlooks (which can be estimated in the whole image), a scale parameter and a\ntexture parameter. This paper presents a new proposal for feature extraction\nand region discrimination in SAR imagery, using the geodesic distance as a\nmeasure of dissimilarity between $\\mathcal{G}_I^0$ models. We derive geodesic\ndistances between models that describe several practical situations, assuming\nthe number of looks known, for same and different texture and for same and\ndifferent scale. We then apply this new tool to the problems of (i)~identifying\nedges between regions with different texture, and (ii)~quantify the\ndissimilarity between pairs of samples in actual SAR data. We analyze the\nadvantages of using the geodesic distance when compared to stochastic\ndistances.\n",
"title": "The Geodesic Distance between $\\mathcal{G}_I^0$ Models and its Application to Region Discrimination"
}
| null | null | null | null | true | null |
11684
| null |
Default
| null | null |
null |
{
"abstract": " We present a novel optimization method, named the Combined Optimization\nMethod (COM), for the joint optimization of two or more cost functions. Unlike\nthe conventional joint optimization schemes, which try to find minima in a\nweighted sum of cost functions, the COM explores search space for common minima\nshared by all the cost functions. Given a set of multiple cost functions that\nhave qualitatively different distributions of local minima with each other, the\nproposed method finds the common minima with a high success rate without the\nhelp of any metaheuristics. As a demonstration, we apply the COM to the crystal\nstructure prediction in materials science. By introducing the concept of data\nassimilation, i.e., adopting the theoretical potential energy of the crystal\nand the crystallinity, which characterizes the agreement with the theoretical\nand experimental X-ray diffraction patterns, as cost functions, we show that\nthe correct crystal structures of Si diamond, low quartz, and low cristobalite\ncan be predicted with significantly higher success rates than the previous\nmethods.\n",
"title": "Search for Common Minima in Joint Optimization of Multiple Cost Functions"
}
| null | null | null | null | true | null |
11685
| null |
Default
| null | null |
null |
{
"abstract": " Information forms the basis for all human behavior, including the ubiquitous\ndecision-making that people constantly perform in their every day lives. It is\nthus the mission of researchers to understand how humans process information to\nreach decisions. In order to facilitate this task, this work proposes a novel\nmethod of studying the reception of granular expressions in natural language.\nThe approach utilizes LASSO regularization as a statistical tool to extract\ndecisive words from textual content and draw statistical inferences based on\nthe correspondence between the occurrences of words and an exogenous response\nvariable. Accordingly, the method immediately suggests significant implications\nfor social sciences and Information Systems research: everyone can now identify\ntext segments and word choices that are statistically relevant to authors or\nreaders and, based on this knowledge, test hypotheses from behavioral research.\nWe demonstrate the contribution of our method by examining how authors\ncommunicate subjective information through narrative materials. This allows us\nto answer the question of which words to choose when communicating negative\ninformation. On the other hand, we show that investors trade not only upon\nfacts in financial disclosures but are distracted by filler words and\nnon-informative language. Practitioners - for example those in the fields of\ninvestor communications or marketing - can exploit our insights to enhance\ntheir writings based on the true perception of word choice.\n",
"title": "Statistical Inferences for Polarity Identification in Natural Language"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
11686
| null |
Validated
| null | null |
null |
{
"abstract": " The paper evaluates the influence of the maximum vehicle acceleration and\nvariable proportions of ACC/CACC vehicles on the throughput of an intersection.\nTwo cases are studied: (1) free road downstream of the intersection; and (2)\nred light at some distance downstream of the intersection. Simulation of a\n4-mile stretch of an arterial with 13 signalized intersections is used to\nevaluate the impact of (C)ACC vehicles on the mean and standard deviation of\ntravel time as the proportion of (C)ACC vehicles is increased. The results\nsuggest a very high urban mobility benefit of (C)ACC vehicles at little or no\ncost in infrastructure.\n",
"title": "Effect of Adaptive and Cooperative Adaptive Cruise Control on Throughput of Signalized Arterials"
}
| null | null | null | null | true | null |
11687
| null |
Default
| null | null |
null |
{
"abstract": " Let Z_n be the finite commutative ring of residue classes modulo n and\nGamma(Z_n) be its zero-divisor graph. The nilradical graph and non-nilradical\ngraph of Z_n are denoted by N(Z_n) and Omega(Z_n) respectively. In 2012, Haynes\net al. [5] introduced the concept of very cost effective graph. For a graph G =\n(V,E) and a set of vertices S subset of V, a vertex v in S is said to be very\ncost effective if it is adjacent to more vertices in V§than in S. A\nbipartition Pi = {S, V§} is called very cost effective if both S and V§are\nvery cost effective sets [5,6]. In this paper, we investigate the very cost\neffective bipartition of Gamma(Z_n), where n = p_1 p_2 ... p_m, here all p_i's\nare distinct primes. In addition, we discuss the cases in which N(Z_n) and\nOmega(Z_n) graphs have very cost effective bipartition for different n.\nFinally, we derive some results for very cost effective bipartition of the Line\ngraph and Total graph of Gamma(Z_n), denoted by L(Gamma(Z_n)) and T(Gamma(Z_n))\nrespectively.\n",
"title": "Very cost effective bipartition in Gamma(Z_n)"
}
| null | null | null | null | true | null |
11688
| null |
Default
| null | null |
null |
{
"abstract": " We demonstrate the existence of a novel quasiparticle: an exciton in a\nsemiconductor doubly dressed with two photons of different wavelengths: near\ninfrared cavity photon and terahertz (THz) photon, with the THz coupling\nstrength approaching the ultra-strong coupling regime. This quasiparticle is\ncomposed of three different bosons, being a mixture of a matter-light\nquasiparticle. Our observations are confirmed by a detailed theoretical\nanalysis, treating quantum mechanically all three bosonic fields. The doubly\ndressed quasiparticles retain the bosonic nature of their constituents, but\ntheir internal quantum structure strongly depends on the intensity of the\napplied terahertz field.\n",
"title": "Doubly dressed bosons - exciton-polaritons in a strong terahertz field"
}
| null | null | null | null | true | null |
11689
| null |
Default
| null | null |
null |
{
"abstract": " Future grid scenario analysis requires a major departure from conventional\npower system planning, where only a handful of most critical conditions is\ntypically analyzed. To capture the inter-seasonal variations in renewable\ngeneration of a future grid scenario necessitates the use of computationally\nintensive time-series analysis. In this paper, we propose a planning framework\nfor fast stability scanning of future grid scenarios using a novel feature\nselection algorithm and a novel self-adaptive PSO-k-means clustering algorithm.\nTo achieve the computational speed-up, the stability analysis is performed only\non small number of representative cluster centroids instead of on the full set\nof operating conditions. As a case study, we perform small-signal stability and\nsteady-state voltage stability scanning of a simplified model of the Australian\nNational Electricity Market with significant penetration of renewable\ngeneration. The simulation results show the effectiveness of the proposed\napproach. Compared to an exhaustive time series scanning, the proposed\nframework reduced the computational burden up to ten times, with an acceptable\nlevel of accuracy.\n",
"title": "Fast Stability Scanning for Future Grid Scenario Analysis"
}
| null | null | null | null | true | null |
11690
| null |
Default
| null | null |
null |
{
"abstract": " Pump-probe electron energy-loss spectroscopy (EELS) with femtosecond temporal\nresolution will be a transformative research tool for studying non-equilibrium\nchemistry and electronic dynamics of matter. In this paper, we propose a new\nconcept of femtosecond EELS utilizing mega-electron-volt electron beams from a\nradio-frequency (rf) photocathode source. The high acceleration gradient and\nhigh beam energy of the rf gun are critical to the generation of 10-femtosecond\nelectron beams, which enables improvement of the temporal resolution by more\nthan one order of magnitude beyond the state of the art. The major innovation\nin our proposal - the `reference-beam technique', relaxes the energy stability\nrequirement on the rf power source by roughly two orders of magnitude.\nRequirements on the electron beam quality, photocathode, spectrometer and\ndetector are also discussed. Supported by particle-tracking simulations, we\ndemonstrate the feasibility of achieving sub-electron-volt energy resolution\nand ~10-femtosecond temporal resolution with existing or near-future hardware\nperformances.\n",
"title": "Femtosecond Mega-electron-volt Electron Energy-Loss Spectroscopy"
}
| null | null | null | null | true | null |
11691
| null |
Default
| null | null |
null |
{
"abstract": " We consider families of symmetric linear programs (LPs) that decide a\nproperty of graphs (or other relational structures) in the sense that, for each\nsize of graph, there is an LP defining a polyhedral lift that separates the\ninteger points corresponding to graphs with the property from those\ncorresponding to graphs without the property. We show that this is equivalent,\nwith at most polynomial blow-up in size, to families of symmetric Boolean\ncircuits with threshold gates. In particular, when we consider polynomial-size\nLPs, the model is equivalent to definability in a non-uniform version of\nfixed-point logic with counting (FPC). Known upper and lower bounds for FPC\napply to the non-uniform version. In particular, this implies that the class of\ngraphs with perfect matchings has polynomial-size symmetric LPs while we obtain\nan exponential lower bound for symmetric LPs for the class of Hamiltonian\ngraphs. We compare and contrast this with previous results (Yannakakis 1991)\nshowing that any symmetric LPs for the matching and TSP polytopes have\nexponential size. As an application, we establish that for random, uniformly\ndistributed graphs, polynomial-size symmetric LPs are as powerful as general\nBoolean circuits. We illustrate the effect of this on the well-studied\nplanted-clique problem.\n",
"title": "On the Power of Symmetric Linear Programs"
}
| null | null | null | null | true | null |
11692
| null |
Default
| null | null |
null |
{
"abstract": " The CHIME telescope (the Canadian Hydrogen Intensity Mapping Experiment)\nrecently built in Penticton, Canada, is currently being commissioned.\nOriginally designed as a cosmology experiment, it was soon recognized that\nCHIME has the potential to simultaneously serve as an incredibly useful radio\ntelescope for pulsar science. CHIME operates across a wide bandwidth of 400-800\nMHz and will have a collecting area and sensitivity comparable to that of the\n100-m class radio telescopes. CHIME has a huge field of view of ~250 square\ndegrees. It will be capable of observing 10 pulsars simultaneously, 24-hours\nper day, every day, while still accomplishing its missions to study Baryon\nAcoustic Oscillations and Fast Radio Bursts. It will carry out daily monitoring\nof roughly half of all pulsars in the northern hemisphere, including all\nNANOGrav pulsars employed in the Pulsar Timing Array project. It will cycle\nthrough all pulsars in the northern hemisphere with a range of cadence of no\nmore than 10 days.\n",
"title": "Pulsar science with the CHIME telescope"
}
| null | null | null | null | true | null |
11693
| null |
Default
| null | null |
null |
{
"abstract": " In the present article, we analyse the behaviour of a new family of\nKantorovich type sampling operators $(K_w^{\\varphi}f)_{w>0}.$ First, we give a\nVoronovskaya type theorem for these Kantorovich generalized sampling series and\na corresponding quantitative version in terms of the first order of modulus of\ncontinuity. Further, we study the order of approximation in $C({\\mathbb{R}})$\n(the set of all uniformly continuous and bounded functions on ${\\mathbb{R}}$)\nfor the family $(K_w^{\\varphi}f)_{w>0}.$ Finally, we give some examples of\nkernels such as B-spline kernels and Blackman-Harris kernel to which the theory\ncan be applied.\n",
"title": "Approximation by generalized Kantorovich sampling type series"
}
| null | null | null | null | true | null |
11694
| null |
Default
| null | null |
null |
{
"abstract": " We theoretically investigate the mechanical stability of three-dimensional\n(3D) foam geometry in a cell sheet and apply its understandings to epithelial\nintegrity and cell delamination. Analytical calculations revealed that the\nmonolayer integrity of cell sheet is lost to delamination by a spontaneous\nsymmetry breaking, inherently depending on the 3D foam geometry of cells; i.e.,\nthe instability spontaneously appears when the cell density in the sheet plane\nincreases and/or when the number of neighboring cells decreases, as observed in\nvivo. The instability is also facilitated by the delaminated cell-specific\nforce generation upon lateral surfaces, which are driven by cell-intrinsic\ngenetic programs during cell invasion and apoptosis in physiology. In\nprinciple, this instability emerges from the force balance on the lateral\nboundaries among cells. Additionally, taking into account the cell-intrinsic\nforce generation on apical and basal sides, which are also broadly observed in\nmorphogenesis, homeostasis, and carcinogenesis, we found apically/basally\ndirected cell delaminations and pseudostratified structures, which could\nuniversally explain mechanical regulations of epithelial geometries in both\nphysiology and pathophysiology.\n",
"title": "Mechanical Instability Leading Epithelial Cell Delamination"
}
| null | null | null | null | true | null |
11695
| null |
Default
| null | null |
null |
{
"abstract": " A two-layer shallow water type model is proposed to describe bedload sediment\ntransport. The upper layer is filled by water and the lower one by sediment.\nThe key point falls on the definition of the friction laws between the two\nlayers, which are a generalization of those introduced in Fernández-Nieto et\nal. (ESAIM: M2AN, 51:115-145, 2017). This definition allows to apply properly\nthe two-layer shallow water model for the case of intense and slow bedload\nsediment transport. Moreover, we prove that the two-layer model converges to a\nSaint-Venant-Exner system (SVE) including gravitational effects when the ratio\nbetween the hydrodynamic and morphodynamic time scales is small. The SVE with\ngravitational effects is a degenerated nonlinear parabolic system. This means\nthat its numerical approximation is very expensive from a computational point\nof view, see for example T. Morales de Luna et al. (J. Sci. Comp., 48(1):\n258-273, 2011). In this work, gravitational effects are introduced into the\ntwo-layer system without such extra computational cost. Finally, we also\nconsider a generalization of the model that includes a non-hydrostatic pressure\ncorrection for the fluid layer and the boundary condition at the sediment\nsurface. Numerical tests show that the model provides promising results and\nbehave well in low transport rate regimes as well as in many other situations.\n",
"title": "A two-layer shallow water model for bedload sediment transport: convergence to Saint-Venant-Exner model"
}
| null | null |
[
"Physics"
] | null | true | null |
11696
| null |
Validated
| null | null |
null |
{
"abstract": " Chondrules are primitive materials in the Solar System. They are formed in\nthe first about 3 Myr of the Solar System's history. This timescale is longer\nthan that of Mars formation, and it is conceivable that protoplanets,\nplanetesimals and chondrules might have existed simultaneously in the solar\nnebula. Due to protoplanets perturbation on the planetesimal dynamics and\nchondrule accretion on them, all the formed chondrules are unlikely to be\naccreted by planetesimals. We investigate the amount of chondrules accreted by\nplanetesimals in such a condition. We assume that a protoplanet is in\noligarchic growth, and we perform analytical calculations of chondrule\naccretion both by a protoplanet and by planetesimals. Through the oligarchic\ngrowth stage, planetesimals accrete about half of the formed chondrules. The\nsmallest planetesimals get the largest amount of the chondrules, compared with\nthe amount accreted by more massive planetesimals. We perform a parameter study\nand find that this fraction is not largely changed for a wide range of\nparameter sets.\n",
"title": "Chondrule Accretion with a Growing Protoplanet"
}
| null | null | null | null | true | null |
11697
| null |
Default
| null | null |
null |
{
"abstract": " The edit distance under the DCJ model can be computed in linear time for\ngenomes with equal content or with Indels. But it becomes NP-Hard in the\npresence of duplications, a problem largely unsolved especially when Indels are\nconsidered. In this paper, we compare two mainstream methods to deal with\nduplications and associate them with Indels: one by deletion, namely\nDCJ-Indel-Exemplar distance; versus the other by gene matching, namely\nDCJ-Indel-Matching distance. We design branch-and-bound algorithms with set of\noptimization methods to compute exact distances for both. Furthermore, median\nproblems are discussed in alignment with both of these distance methods, which\nare to find a median genome that minimizes distances between itself and three\ngiven genomes. Lin-Kernighan (LK) heuristic is leveraged and powered up by\nsub-graph decomposition and search space reduction technologies to handle\nmedian computation. A wide range of experiments are conducted on synthetic data\nsets and real data sets to show pros and cons of these two distance metrics per\nse, as well as putting them in the median computation scenario.\n",
"title": "Exemplar or Matching: Modeling DCJ Problems with Unequal Content Genome Data"
}
| null | null | null | null | true | null |
11698
| null |
Default
| null | null |
null |
{
"abstract": " Learning-based hashing methods are widely used for nearest neighbor\nretrieval, and recently, online hashing methods have demonstrated good\nperformance-complexity trade-offs by learning hash functions from streaming\ndata. In this paper, we first address a key challenge for online hashing: the\nbinary codes for indexed data must be recomputed to keep pace with updates to\nthe hash functions. We propose an efficient quality measure for hash functions,\nbased on an information-theoretic quantity, mutual information, and use it\nsuccessfully as a criterion to eliminate unnecessary hash table updates. Next,\nwe also show how to optimize the mutual information objective using stochastic\ngradient descent. We thus develop a novel hashing method, MIHash, that can be\nused in both online and batch settings. Experiments on image retrieval\nbenchmarks (including a 2.5M image dataset) confirm the effectiveness of our\nformulation, both in reducing hash table recomputations and in learning\nhigh-quality hash functions.\n",
"title": "MIHash: Online Hashing with Mutual Information"
}
| null | null | null | null | true | null |
11699
| null |
Default
| null | null |
null |
{
"abstract": " Given a 0-dimensional scheme $\\mathbb{X}$ in a projective space\n$\\mathbb{P}^n_K$ over a field $K$, we characterize the Cayley-Bacharach\nproperty of $\\mathbb{X}$ in terms of the algebraic structure of the Dedekind\ndifferent of its homogeneous coordinate ring. Moreover, we characterize\nCayley-Bacharach schemes by Dedekind's formula for the conductor and the\ncomplementary module, we study schemes with minimal Dedekind different using\nthe trace of the complementary module, and we prove various results about\nalmost Gorenstein and nearly Gorenstein schemes.\n",
"title": "On the Dedekind different of a Cayley-Bacharach scheme"
}
| null | null | null | null | true | null |
11700
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.