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": " Lu and Boutilier proposed a novel approach based on \"minimax regret\" to use\nclassical score based voting rules in the setting where preferences can be any\npartial (instead of complete) orders over the set of alternatives. We show here\nthat such an approach is vulnerable to a new kind of manipulation which was not\npresent in the classical (where preferences are complete orders) world of\nvoting. We call this attack \"manipulative elicitation.\" More specifically, it\nmay be possible to (partially) elicit the preferences of the agents in a way\nthat makes some distinguished alternative win the election who may not be a\nwinner if we elicit every preference completely. More alarmingly, we show that\nthe related computational task is polynomial time solvable for a large class of\nvoting rules which includes all scoring rules, maximin, Copeland$^\\alpha$ for\nevery $\\alpha\\in[0,1]$, simplified Bucklin voting rules, etc. We then show that\nintroducing a parameter per pair of alternatives which specifies the minimum\nnumber of partial preferences where this pair of alternatives must be\ncomparable makes the related computational task of manipulative elicitation\n\\NPC for all common voting rules including a class of scoring rules which\nincludes the plurality, $k$-approval, $k$-veto, veto, and Borda voting rules,\nmaximin, Copeland$^\\alpha$ for every $\\alpha\\in[0,1]$, and simplified Bucklin\nvoting rules. Hence, in this work, we discover a fundamental vulnerability in\nusing minimax regret based approach in partial preferential setting and propose\na novel way to tackle it.\n",
"title": "Manipulative Elicitation -- A New Attack on Elections with Incomplete Preferences"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5601
| null |
Validated
| null | null |
null |
{
"abstract": " Event sequence, asynchronously generated with random timestamp, is ubiquitous\namong applications. The precise and arbitrary timestamp can carry important\nclues about the underlying dynamics, and has lent the event data fundamentally\ndifferent from the time-series whereby series is indexed with fixed and equal\ntime interval. One expressive mathematical tool for modeling event is point\nprocess. The intensity functions of many point processes involve two\ncomponents: the background and the effect by the history. Due to its inherent\nspontaneousness, the background can be treated as a time series while the other\nneed to handle the history events. In this paper, we model the background by a\nRecurrent Neural Network (RNN) with its units aligned with time series indexes\nwhile the history effect is modeled by another RNN whose units are aligned with\nasynchronous events to capture the long-range dynamics. The whole model with\nevent type and timestamp prediction output layers can be trained end-to-end.\nOur approach takes an RNN perspective to point process, and models its\nbackground and history effect. For utility, our method allows a black-box\ntreatment for modeling the intensity which is often a pre-defined parametric\nform in point processes. Meanwhile end-to-end training opens the venue for\nreusing existing rich techniques in deep network for point process modeling. We\napply our model to the predictive maintenance problem using a log dataset by\nmore than 1000 ATMs from a global bank headquartered in North America.\n",
"title": "Modeling The Intensity Function Of Point Process Via Recurrent Neural Networks"
}
| null | null | null | null | true | null |
5602
| null |
Default
| null | null |
null |
{
"abstract": " Let $M$ be a finite von Neumann algebra (resp. a type II$_{1}$ factor) and\nlet $N\\subset M$ be a II$_{1}$ factor (resp. $N\\subset M$ have an atomic part).\nWe prove that the inclusion $N\\subset M$ is amenable implies the identity map\non $M$ has an approximate factorization through $M_m(\\mathbb{C})\\otimes N $ via\ntrace preserving normal unital completely positive maps, which is a\ngeneralization of a result of Haagerup. We also prove two permanence properties\nfor amenable inclusions. One is weak Haagerup property, the other is weak\nexactness.\n",
"title": "A note on relative amenable of finite von Neumann algebras"
}
| null | null | null | null | true | null |
5603
| null |
Default
| null | null |
null |
{
"abstract": " Let $q$ be a positive integer. Recently, Niu and Liu proved that if $n\\ge\n\\max\\{q,1198-q\\}$, then the product $(1^3+q^3)(2^3+q^3)\\cdots (n^3+q^3)$ is not\na powerful number. In this note, we prove that (i) for any odd prime power\n$\\ell$ and $n\\ge \\max\\{q,11-q\\}$, the product\n$(1^{\\ell}+q^{\\ell})(2^{\\ell}+q^{\\ell})\\cdots (n^{\\ell}+q^{\\ell})$ is not a\npowerful number; (2) for any positive odd integer $\\ell$, there exists an\ninteger $N_{q,\\ell}$ such that for any positive integer $n\\ge N_{q,\\ell}$, the\nproduct $(1^{\\ell}+q^{\\ell})(2^{\\ell}+q^{\\ell})\\cdots (n^{\\ell}+q^{\\ell})$ is\nnot a powerful number.\n",
"title": "Powerful numbers in $(1^{\\ell}+q^{\\ell})(2^{\\ell}+q^{\\ell})\\cdots (n^{\\ell}+q^{\\ell})$"
}
| null | null | null | null | true | null |
5604
| null |
Default
| null | null |
null |
{
"abstract": " The growing popularity of autonomous systems creates a need for reliable and\nefficient metric pose retrieval algorithms. Currently used approaches tend to\nrely on nearest neighbor search of binary descriptors to perform the 2D-3D\nmatching and guarantee realtime capabilities on mobile platforms. These methods\nstruggle, however, with the growing size of the map, changes in viewpoint or\nappearance, and visual aliasing present in the environment. The rigidly defined\ndescriptor patterns only capture a limited neighborhood of the keypoint and\ncompletely ignore the overall visual context.\nWe propose LandmarkBoost - an approach that, in contrast to the conventional\n2D-3D matching methods, casts the search problem as a landmark classification\ntask. We use a boosted classifier to classify landmark observations and\ndirectly obtain correspondences as classifier scores. We also introduce a\nformulation of visual context that is flexible, efficient to compute, and can\ncapture relationships in the entire image plane. The original binary\ndescriptors are augmented with contextual information and informative features\nare selected by the boosting framework. Through detailed experiments, we\nevaluate the retrieval quality and performance of LandmarkBoost, demonstrating\nthat it outperforms common state-of-the-art descriptor matching methods.\n",
"title": "LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization"
}
| null | null | null | null | true | null |
5605
| null |
Default
| null | null |
null |
{
"abstract": " The paper discusses stably trivial torsors for spin and orthogonal groups\nover smooth affine schemes over infinite perfect fields of characteristic\nunequal to 2. We give a complete description of all the invariants relevant for\nthe classification of such objects over schemes of dimension at most $3$, along\nwith many examples. The results are based on the\n$\\mathbb{A}^1$-representability theorem for torsors and transfer of known\ncomputations of $\\mathbb{A}^1$-homotopy sheaves along the sporadic isomorphisms\nto spin groups.\n",
"title": "On stably trivial spin torsors over low-dimensional schemes"
}
| null | null | null | null | true | null |
5606
| null |
Default
| null | null |
null |
{
"abstract": " Liu et al. (2017) provide a comprehensive account of research on dependency\ndistance in human languages. While the article is a very rich and useful report\non this complex subject, here I will expand on a few specific issues where\nresearch in computational linguistics (specifically natural language\nprocessing) can inform DDM research, and vice versa. These aspects have not\nbeen explored much in the article by Liu et al. or elsewhere, probably due to\nthe little overlap between both research communities, but they may provide\ninteresting insights for improving our understanding of the evolution of human\nlanguages, the mechanisms by which the brain processes and understands\nlanguage, and the construction of effective computer systems to achieve this\ngoal.\n",
"title": "On the relation between dependency distance, crossing dependencies, and parsing. Comment on \"Dependency distance: a new perspective on syntactic patterns in natural languages\" by Haitao Liu et al"
}
| null | null | null | null | true | null |
5607
| null |
Default
| null | null |
null |
{
"abstract": " We prove several results about chordal graphs and weighted chordal graphs by\nfocusing on exposed edges. These are edges that are properly contained in a\nsingle maximal complete subgraph. This leads to a characterization of chordal\ngraphs via deletions of a sequence of exposed edges from a complete graph. Most\ninteresting is that in this context the connected components of the\nedge-induced subgraph of exposed edges are 2-edge connected. We use this latter\nfact in the weighted case to give a modified version of Kruskal's second\nalgorithm for finding a minimum spanning tree in a weighted chordal graph. This\nmodified algorithm benefits from being local in an important sense.\n",
"title": "Edge Erasures and Chordal Graphs"
}
| null | null | null | null | true | null |
5608
| null |
Default
| null | null |
null |
{
"abstract": " Hotspots of surface-enhanced resonance Raman scattering (SERRS) are localized\nwithin 1 nm at gaps or crevices of plasmonic nanoparticle (NP) dimers. We\ndemonstrate SERRS hotspots with volumes that are extended in one dimension tens\nof thousand times compared to standard zero-dimensional hotspots using gaps or\ncrevices of plasmonic nanowire (NW) dimers.\n",
"title": "One-dimensional plasmonic hotspots located between silver nanowire dimers evaluated by surface-enhanced resonance Raman scattering"
}
| null | null | null | null | true | null |
5609
| null |
Default
| null | null |
null |
{
"abstract": " Complex computer codes are often too time expensive to be directly used to\nperform uncertainty, sensitivity, optimization and robustness analyses. A\nwidely accepted method to circumvent this problem consists in replacing\ncpu-time expensive computer models by cpu inexpensive mathematical functions,\ncalled metamodels. For example, the Gaussian process (Gp) model has shown\nstrong capabilities to solve practical problems , often involving several\ninterlinked issues. However, in case of high dimensional experiments (with\ntypically several tens of inputs), the Gp metamodel building process remains\ndifficult, even unfeasible, and application of variable selection techniques\ncannot be avoided. In this paper, we present a general methodology allowing to\nbuild a Gp metamodel with large number of inputs in a very efficient manner.\nWhile our work focused on the Gp metamodel, its principles are fully generic\nand can be applied to any types of metamodel. The objective is twofold:\nestimating from a minimal number of computer experiments a highly predictive\nmetamodel. This methodology is successfully applied on an industrial computer\ncode.\n",
"title": "An efficient methodology for the analysis and modeling of computer experiments with large number of inputs"
}
| null | null | null | null | true | null |
5610
| null |
Default
| null | null |
null |
{
"abstract": " Domain adaptation is crucial in many real-world applications where the\ndistribution of the training data differs from the distribution of the test\ndata. Previous Deep Learning-based approaches to domain adaptation need to be\ntrained jointly on source and target domain data and are therefore unappealing\nin scenarios where models need to be adapted to a large number of domains or\nwhere a domain is evolving, e.g. spam detection where attackers continuously\nchange their tactics.\nTo fill this gap, we propose Knowledge Adaptation, an extension of Knowledge\nDistillation (Bucilua et al., 2006; Hinton et al., 2015) to the domain\nadaptation scenario. We show how a student model achieves state-of-the-art\nresults on unsupervised domain adaptation from multiple sources on a standard\nsentiment analysis benchmark by taking into account the domain-specific\nexpertise of multiple teachers and the similarities between their domains.\nWhen learning from a single teacher, using domain similarity to gauge\ntrustworthiness is inadequate. To this end, we propose a simple metric that\ncorrelates well with the teacher's accuracy in the target domain. We\ndemonstrate that incorporating high-confidence examples selected by this metric\nenables the student model to achieve state-of-the-art performance in the\nsingle-source scenario.\n",
"title": "Knowledge Adaptation: Teaching to Adapt"
}
| null | null | null | null | true | null |
5611
| null |
Default
| null | null |
null |
{
"abstract": " Microblogging sites are the direct platform for the users to express their\nviews. It has been observed from previous studies that people are viable to\nflaunt their emotions for events (eg. natural catastrophes, sports, academics\netc.), for persons (actor/actress, sports person, scientist) and for the places\nthey visit. In this study we focused on a sport event, particularly the cricket\ntournament and collected the emotions of the fans for their favorite players\nusing their tweets. Further, we acquired the stock market performance of the\nbrands which are either endorsing the players or sponsoring the match in the\ntournament. It has been observed that performance of the player triggers the\nusers to flourish their emotions over social media therefore, we observed\ncorrelation between players performance and fans' emotions. Therefore, we found\nthe direct connection between player's performance with brand's behavior on\nstock market.\n",
"title": "A Datamining Approach for Emotions Extraction and Discovering Cricketers performance from Stadium to Sensex"
}
| null | null | null | null | true | null |
5612
| null |
Default
| null | null |
null |
{
"abstract": " This paper discusses the local linear smoothing to estimate the unknown first\nand second infinitesimal moments in second-order jump-diffusion model based on\nGamma asymmetric kernels. Under the mild conditions, we obtain the weak\nconsistency and the asymptotic normality of these estimators for both interior\nand boundary design points. Besides the standard properties of the local linear\nestimation such as simple bias representation and boundary bias correction, the\nlocal linear smoothing using Gamma asymmetric kernels possess some extra\nadvantages such as variable bandwidth, variance reduction and resistance to\nsparse design, which is validated through finite sample simulation study.\nFinally, we employ the estimators for the return of some high frequency\nfinancial data.\n",
"title": "Local Nonparametric Estimation for Second-Order Jump-Diffusion Model Using Gamma Asymmetric Kernels"
}
| null | null | null | null | true | null |
5613
| null |
Default
| null | null |
null |
{
"abstract": " Among the ergodic actions of a compact quantum group $\\mathbb{G}$ on possibly\nnon-commutative spaces, those that are {\\it embeddable} are the natural\nanalogues of actions of a compact group on its homogeneous spaces. These can be\nrealized as {\\it coideal subalgebras} of the function algebra\n$\\mathcal{O}(\\mathbb{G})$ attached to the compact quantum group.\nWe classify the embeddable ergodic actions of the compact quantum group\n$O_{-1}(2)$, basing our analysis on the bijective correspondence between all\nergodic actions of the classical group $O(2)$ and those of its quantum twist\nresulting from the monoidal equivalence between their respective tensor\ncategories of unitary representations.\nIn the last section we give counterexamples showing that in general we cannot\nexpect a bijective correspondence between embeddable ergodic actions of two\nmonoidally equivalent compact quantum groups.\n",
"title": "Ergodic actions of the compact quantum group $O_{-1}(2)$"
}
| null | null | null | null | true | null |
5614
| null |
Default
| null | null |
null |
{
"abstract": " We formulate the so called \"VARMA covariance matching problem\" and\ndemonstrate the existence of a solution using the degree theory from\ndifferential topology.\n",
"title": "On Vector ARMA Models Consistent with a Finite Matrix Covariance Sequence"
}
| null | null | null | null | true | null |
5615
| null |
Default
| null | null |
null |
{
"abstract": " We devise an approach to the calculation of scaling dimensions of generic\noperators describing scattering within multi-channel Luttinger liquid. The\nlocal impurity scattering in an arbitrary configuration of conducting and\ninsulating channels is investigated and the problem is reduced to a single\nalgebraic matrix equation. In particular, the solution to this equation is\nfound for a finite array of chains described by Luttinger liquid models. It is\nfound that for a weak inter-chain hybridisation and intra-channel\nelectron-electron attraction the edge wires are robust against disorder whereas\nbulk wires, on contrary, become insulating. Thus, the edge state may exist in a\nfinite sliding Luttinger liquid without time-reversal symmetry breaking\n(quantum Hall systems) or spin-orbit interaction (topological insulators).\n",
"title": "Edge states in non-Fermi liquids"
}
| null | null | null | null | true | null |
5616
| null |
Default
| null | null |
null |
{
"abstract": " The problem of automatically generating a computer program from some\nspecification has been studied since the early days of AI. Recently, two\ncompeting approaches for automatic program learning have received significant\nattention: (1) neural program synthesis, where a neural network is conditioned\non input/output (I/O) examples and learns to generate a program, and (2) neural\nprogram induction, where a neural network generates new outputs directly using\na latent program representation.\nHere, for the first time, we directly compare both approaches on a\nlarge-scale, real-world learning task. We additionally contrast to rule-based\nprogram synthesis, which uses hand-crafted semantics to guide the program\ngeneration. Our neural models use a modified attention RNN to allow encoding of\nvariable-sized sets of I/O pairs. Our best synthesis model achieves 92%\naccuracy on a real-world test set, compared to the 34% accuracy of the previous\nbest neural synthesis approach. The synthesis model also outperforms a\ncomparable induction model on this task, but we more importantly demonstrate\nthat the strength of each approach is highly dependent on the evaluation metric\nand end-user application. Finally, we show that we can train our neural models\nto remain very robust to the type of noise expected in real-world data (e.g.,\ntypos), while a highly-engineered rule-based system fails entirely.\n",
"title": "RobustFill: Neural Program Learning under Noisy I/O"
}
| null | null | null | null | true | null |
5617
| null |
Default
| null | null |
null |
{
"abstract": " Flexible and transparent electronics presents a new era of electronic\ntechnologies. Ubiquitous applications involve wearable electronics, biosensors,\nflexible transparent displays, radio-frequency identifications (RFIDs),\netc.Zinc oxide (ZnO) and related materials are the most commonly used inorganic\nsemiconductors in flexible and transparent devices, owing to their high\nelectrical performance, together with low processing temperature and good\noptical transparency.In this paper, we review recent advances in flexible and\ntransparent thin-film transistors (TFTs) based on ZnO and related\nmaterials.After a brief introduction, the main progresses on the preparation of\neach component (substrate, electrodes, channel and dielectrics) are summarized\nand discussed. Then, the effect of mechanical bending on electrical performance\nwas highlighted. Finally, we suggest the challenges and opportunities in future\ninvestigations.\n",
"title": "Review of flexible and transparent thin-film transistors based on zinc oxide and related materials"
}
| null | null | null | null | true | null |
5618
| null |
Default
| null | null |
null |
{
"abstract": " The Decay-At-rest Experiment for delta-CP violation At a Laboratory for\nUnderground Science (DAEdALUS) and the Isotope Decay-At-Rest experiment\n(IsoDAR) are proposed experiments to search for CP violation in the neutrino\nsector, and \"sterile\" neutrinos, respectively. In order to be decisive within 5\nyears, the neutrino flux and, consequently, the driver beam current (produced\nby chained cyclotrons) must be high. H2+ was chosen as primary beam ion in\norder to reduce the electrical current and thus space charge. This has the\nadded advantage of allowing for stripping extraction at the exit of the\nDAEdALUS Superconducting Ring Cyclotron (DSRC). The primary beam current is\nhigher than current cyclotrons have demonstrated which has led to a substantial\nR&D effort of our collaboration in the last years. We present the results of\nthis research, including tests of prototypes and highly realistic beam\nsimulations, which led to the latest physics-based design. The presented\nresults suggest that it is feasible, albeit challenging, to accelerate 5 mA of\nH2+ to 60 MeV/amu in a compact cyclotron and boost it to 800 MeV/amu in the\nDSRC with clean extraction in both cases.\n",
"title": "Updated physics design of the DAEdALUS and IsoDAR coupled cyclotrons for high intensity H2+ beam production"
}
| null | null | null | null | true | null |
5619
| null |
Default
| null | null |
null |
{
"abstract": " We present a kernel-independent method that applies hierarchical matrices to\nthe problem of maximum likelihood estimation for Gaussian processes. The\nproposed approximation provides natural and scalable stochastic estimators for\nits gradient and Hessian, as well as the expected Fisher information matrix,\nthat are computable in quasilinear $O(n \\log^2 n)$ complexity for a large range\nof models. To accomplish this, we (i) choose a specific hierarchical\napproximation for covariance matrices that enables the computation of their\nexact derivatives and (ii) use a stabilized form of the Hutchinson stochastic\ntrace estimator. Since both the observed and expected information matrices can\nbe computed in quasilinear complexity, covariance matrices for MLEs can also be\nestimated efficiently. After discussing the associated mathematics, we\ndemonstrate the scalability of the method, discuss details of its\nimplementation, and validate that the resulting MLEs and confidence intervals\nbased on the inverse Fisher information matrix faithfully approach those\nobtained by the exact likelihood.\n",
"title": "Scalable Gaussian Process Computations Using Hierarchical Matrices"
}
| null | null | null | null | true | null |
5620
| null |
Default
| null | null |
null |
{
"abstract": " Anyone in need of a data system today is confronted with numerous complex\noptions in terms of system architectures, such as traditional relational\ndatabases, NoSQL and NewSQL solutions as well as several sub-categories like\ncolumn-stores, row-stores etc. This overwhelming array of choices makes\nbootstrapping data-driven applications difficult and time consuming, requiring\nexpertise often not accessible due to cost issues (e.g., to scientific labs or\nsmall businesses). In this paper, we present the vision of evolutionary data\nsystems that free systems architects and application designers from the\ncomplex, cumbersome and expensive process of designing and tuning specialized\ndata system architectures that fit only a single, static application scenario.\nSetting up an evolutionary system is as simple as identifying the data. As new\ndata and queries come in, the system automatically evolves so that its\narchitecture matches the properties of the incoming workload at all times.\nInspired by the theory of evolution, at any given point in time, an\nevolutionary system may employ multiple competing solutions down at the low\nlevel of database architectures -- characterized as combinations of data\nlayouts, access methods and execution strategies. Over time, \"the fittest wins\"\nand becomes the dominant architecture until the environment (workload) changes.\nIn our initial prototype, we demonstrate solutions that can seamlessly evolve\n(back and forth) between a key-value store and a column-store architecture in\norder to adapt to changing workloads.\n",
"title": "Evolutionary Data Systems"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5621
| null |
Validated
| null | null |
null |
{
"abstract": " We consider the problem of sequentially making decisions that are rewarded by\n\"successes\" and \"failures\" which can be predicted through an unknown\nrelationship that depends on a partially controllable vector of attributes for\neach instance. The learner takes an active role in selecting samples from the\ninstance pool. The goal is to maximize the probability of success in either\noffline (training) or online (testing) phases. Our problem is motivated by\nreal-world applications where observations are time-consuming and/or expensive.\nWe develop a knowledge gradient policy using an online Bayesian linear\nclassifier to guide the experiment by maximizing the expected value of\ninformation of labeling each alternative. We provide a finite-time analysis of\nthe estimated error and show that the maximum likelihood estimator based\nproduced by the KG policy is consistent and asymptotically normal. We also show\nthat the knowledge gradient policy is asymptotically optimal in an offline\nsetting. This work further extends the knowledge gradient to the setting of\ncontextual bandits. We report the results of a series of experiments that\ndemonstrate its efficiency.\n",
"title": "Optimal Learning for Sequential Decision Making for Expensive Cost Functions with Stochastic Binary Feedbacks"
}
| null | null | null | null | true | null |
5622
| null |
Default
| null | null |
null |
{
"abstract": " Consider the problem of modeling hysteresis for finite-state random walks\nusing higher-order Markov chains. This Letter introduces a Bayesian framework\nto determine, from data, the number of prior states of recent history upon\nwhich a trajectory is statistically dependent. The general recommendation is to\nuse leave-one-out cross validation, using an easily-computable formula that is\nprovided in closed form. Importantly, Bayes factors using flat model priors are\nbiased in favor of too-complex a model (more hysteresis) when a large amount of\ndata is present and the Akaike information criterion (AIC) is biased in favor\nof too-sparse a model (less hysteresis) when few data are present.\n",
"title": "Determination of hysteresis in finite-state random walks using Bayesian cross validation"
}
| null | null | null | null | true | null |
5623
| null |
Default
| null | null |
null |
{
"abstract": " We establish effective mean-value estimates for a wide class of\nmultiplicative arithmetic functions, thereby providing (essentially optimal)\nquantitative versions of Wirsing's classical estimates and extending those of\nHalász. Several applications are derived, including: estimates for the\ndifference of mean-values of so-called pretentious functions, local laws for\nthe distribution of prime factors in an arbitrary set, and weighted\ndistribution of additive functions.\n",
"title": "Moyennes effectives de fonctions multiplicatives complexes"
}
| null | null | null | null | true | null |
5624
| null |
Default
| null | null |
null |
{
"abstract": " We report on terahertz spectroscopy of quantum spin dynamics in\n$\\alpha$-RuCl$_3$, a system proximate to the Kitaev honeycomb model, as a\nfunction of temperature and magnetic field. An extended magnetic continuum\ndevelops below the structural phase transition at $T_{s2}=62$K. With the onset\nof a long-range magnetic order at $T_N=6.5$K, spectral weight is transferred to\na well-defined magnetic excitation at $\\hbar \\omega_1 = 2.48$meV, which is\naccompanied by a higher-energy band at $\\hbar \\omega_2 = 6.48$meV. Both\nexcitations soften in magnetic field, signaling a quantum phase transition at\n$B_c=7$T where we find a broad continuum dominating the dynamical response.\nAbove $B_c$, the long-range order is suppressed, and on top of the continuum,\nvarious emergent magnetic excitations evolve. These excitations follow clear\nselection rules and exhibit distinct field dependencies, characterizing the\ndynamical properties of the field-induced quantum spin liquid.\n",
"title": "Magnetic Excitations and Continuum of a Field-Induced Quantum Spin Liquid in $α$-RuCl$_3$"
}
| null | null | null | null | true | null |
5625
| null |
Default
| null | null |
null |
{
"abstract": " In this work, we study the benefit of partial relay cooperation. We consider\na two-node system consisting of one source and one relay node transmitting\ninformation to a common destination. The source and the relay have external\ntraffic and in addition, the relay is equipped with a flow controller to\nregulate the incoming traffic from the source node. The cooperation is\nperformed at the network level. A collision channel with erasures is\nconsidered. We provide an exact characterization of the stability region of the\nsystem and we also prove that the system with partial cooperation is always\nbetter or at least equal to the system without the flow controller.\n",
"title": "Wireless Network-Level Partial Relay Cooperation: A Stable Throughput Analysis"
}
| null | null | null | null | true | null |
5626
| null |
Default
| null | null |
null |
{
"abstract": " Test-Driven Development (TDD), an agile development approach that enforces\nthe construction of software systems by means of successive micro-iterative\ntesting coding cycles, has been widely claimed to increase external software\nquality. In view of this, some managers at Paf-a Nordic gaming entertainment\ncompany-were interested in knowing how would TDD perform at their premises.\nEventually, if TDD outperformed their traditional way of coding (i.e., YW,\nshort for Your Way), it would be possible to switch to TDD considering the\nempirical evidence achieved at the company level. We conduct an experiment at\nPaf to evaluate the performance of TDD, YW and the reverse approach of TDD\n(i.e., ITL, short for Iterative-Test Last) on external quality. TDD outperforms\nYW and ITL at Paf. Despite the encouraging results, we cannot recommend Paf to\nimmediately adopt TDD as the difference in performance between YW and TDD is\nsmall. However, as TDD looks promising at Paf, we suggest to move some\ndevelopers to TDD and to run a future experiment to compare the performance of\nTDD and YW. TDD slightly outperforms ITL in controlled experiments for TDD\nnovices. However, more industrial experiments are still needed to evaluate the\nperformance of TDD in real-life contexts.\n",
"title": "Improving Development Practices through Experimentation: an Industrial TDD Case"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5627
| null |
Validated
| null | null |
null |
{
"abstract": " Let $p\\equiv 4,7\\mod 9$ be a rational prime number such that $3\\mod p$ is not\na cubic residue. In this paper we prove the 3-part of the product of the full\nBSD conjectures for $E_p$ and $E_{3p^3}$ is true using an explicit Gross-Zagier\nformula, where $E_p: x^3+y^3=p$ and $E_{3p^2}: x^3+y^3=3p^2$ are the elliptic\ncurves related to the Sylvester conjecture and cube sum problems.\n",
"title": "An explicit Gross-Zagier formula related to the Sylvester Conjecture"
}
| null | null | null | null | true | null |
5628
| null |
Default
| null | null |
null |
{
"abstract": " The aim of our study is to investigate the dynamics of possible comets in the\nHD 10180 system. This investigation is motivated by the discovery of exocomets\nin various systems, especially $\\beta$ Pictoris, as well as in at least ten\nother systems. Detailed theoretical studies about the formation and evolution\nof star--planet systems indicate that exocomets should be quite common. Further\nobservational results are expected in the foreseeable future, in part due to\nthe availability of the Large Synoptic Survey Telescope. Nonetheless, the Solar\nSystem represents the best studied example for comets, thus serving as a prime\nmotivation for investigating comets in HD 10180 as well. HD 10180 is strikingly\nsimilar to the Sun. This system contains six confirmed planets and (at least)\ntwo additional planets subject to final verification. In our studies, we\nconsider comets of different inclinations and eccentricities and find an array\nof different outcomes such as encounters with planets, captures, and escapes.\nComets with relatively large eccentricities are able to enter the inner region\nof the system facing early planetary encounters. Stable comets experience\nlong-term evolution of orbital elements, as expected. We also tried to\ndistinguish cometary families akin to our Solar System but no clear distinction\nbetween possible families was found. Generally, theoretical and observational\nstudies of exoplanets have a large range of ramifications, involving the\norigin, structure and evolution of systems as well as the proliferation of\nwater and prebiotic compounds to terrestrial planets, which will increase their\nchances of being habitable.\n",
"title": "Case Studies of Exocomets in the System of HD 10180"
}
| null | null | null | null | true | null |
5629
| null |
Default
| null | null |
null |
{
"abstract": " Single-Particle Reconstruction (SPR) in Cryo-Electron Microscopy (cryo-EM) is\nthe task of estimating the 3D structure of a molecule from a set of noisy 2D\nprojections, taken from unknown viewing directions. Many algorithms for SPR\nstart from an initial reference molecule, and alternate between refining the\nestimated viewing angles given the molecule, and refining the molecule given\nthe viewing angles. This scheme is called iterative refinement. Reliance on an\ninitial, user-chosen reference introduces model bias, and poor initialization\ncan lead to slow convergence. Furthermore, since no ground truth is available\nfor an unsolved molecule, it is difficult to validate the obtained results.\nThis creates the need for high quality ab initio models that can be quickly\nobtained from experimental data with minimal priors, and which can also be used\nfor validation. We propose a procedure to obtain such an ab initio model\ndirectly from raw data using Kam's autocorrelation method. Kam's method has\nbeen known since 1980, but it leads to an underdetermined system, with missing\northogonal matrices. Until now, this system has been solved only for special\ncases, such as highly symmetric molecules or molecules for which a homologous\nstructure was already available. In this paper, we show that knowledge of just\ntwo clean projections is sufficient to guarantee a unique solution to the\nsystem. This system is solved by an optimization-based heuristic. For the first\ntime, we are then able to obtain a low-resolution ab initio model of an\nasymmetric molecule directly from raw data, without 2D class averaging and\nwithout tilting. Numerical results are presented on both synthetic and\nexperimental data.\n",
"title": "3D ab initio modeling in cryo-EM by autocorrelation analysis"
}
| null | null | null | null | true | null |
5630
| null |
Default
| null | null |
null |
{
"abstract": " No real-world reward function is perfect. Sensory errors and software bugs\nmay result in RL agents observing higher (or lower) rewards than they should.\nFor example, a reinforcement learning agent may prefer states where a sensory\nerror gives it the maximum reward, but where the true reward is actually small.\nWe formalise this problem as a generalised Markov Decision Problem called\nCorrupt Reward MDP. Traditional RL methods fare poorly in CRMDPs, even under\nstrong simplifying assumptions and when trying to compensate for the possibly\ncorrupt rewards. Two ways around the problem are investigated. First, by giving\nthe agent richer data, such as in inverse reinforcement learning and\nsemi-supervised reinforcement learning, reward corruption stemming from\nsystematic sensory errors may sometimes be completely managed. Second, by using\nrandomisation to blunt the agent's optimisation, reward corruption can be\npartially managed under some assumptions.\n",
"title": "Reinforcement Learning with a Corrupted Reward Channel"
}
| null | null | null | null | true | null |
5631
| null |
Default
| null | null |
null |
{
"abstract": " Let $s(\\cdot)$ denote the sum-of-proper-divisors function, that is, $s(n) =\n\\sum_{d\\mid n,~d<n}d$. Erdős-Granville-Pomerance-Spiro conjectured that for\nany set $\\mathcal{A}$ of asymptotic density zero, the preimage set\n$s^{-1}(\\mathcal{A})$ also has density zero. We prove a weak form of this\nconjecture: If $\\epsilon(x)$ is any function tending to $0$ as $x\\to\\infty$,\nand $\\mathcal{A}$ is a set of integers of cardinality at most\n$x^{\\frac12+\\epsilon(x)}$, then the number of integers $n\\le x$ with $s(n) \\in\n\\mathcal{A}$ is $o(x)$, as $x\\to\\infty$. In particular, the EGPS conjecture\nholds for infinite sets with counting function $O(x^{\\frac12 + \\epsilon(x)})$.\nWe also disprove a hypothesis from the same paper of EGPS by showing that for\nany positive numbers $\\alpha$ and $\\epsilon$, there are integers $n$ with\narbitrarily many $s$-preimages lying between $\\alpha(1-\\epsilon)n$ and\n$\\alpha(1+\\epsilon)n$. Finally, we make some remarks on solutions $n$ to\ncongruences of the form $\\sigma(n) \\equiv a\\pmod{n}$, proposing a modification\nof a conjecture appearing in recent work of the first two authors. We also\nimprove a previous upper bound for the number of solutions $n \\leq x$, making\nit uniform in $a$.\n",
"title": "Divisor-sum fibers"
}
| null | null | null | null | true | null |
5632
| null |
Default
| null | null |
null |
{
"abstract": " The formation of precipitated zirconium (Zr) hydrides is closely related to\nthe hydrogen embrittlement problem for the cladding materials of pressured\nwater reactors (PWR). In this work, we systematically investigated the crystal\nstructures of zirconium hydride (ZrHx) with different hydrogen concentrations\n(x = 0~2, atomic ratio) by combining the basin hopping algorithm with first\nprinciples calculations. We conclude that the P3m1 {\\zeta}-ZrH0.5 is\ndynamically unstable, while a novel dynamically stable P3m1 ZrH0.5 structure\nwas discovered in the structure search. The stability of bistable P42/nnm\nZrH1.5 structures and I4/mmm ZrH2 structures are also revisited. We find that\nthe P42/nnm (c/a > 1) ZrH1.5 is dynamically unstable, while the I4/mmm (c/a =\n1.57) ZrH2 is dynamically stable.The P42/nnm (c/a < 1) ZrH1.5 might be a key\nintermediate phase for the transition of {\\gamma}->{\\delta}->{\\epsilon} phases.\nAdditionally, by using the thermal dynamic simulations, we find that\n{\\delta}-ZrH1.5 is the most stable structure at high temperature while ZrH2 is\nthe most stable hydride at low temperature. Slow cooling process will promote\nthe formation of {\\delta}-ZrH1.5, and fast cooling process will promote the\nformation of {\\gamma}-ZrH. These results may help to understand the phase\ntransitions of zirconium hydrides.\n",
"title": "New zirconium hydrides predicted by structure search method based on first principles calculations"
}
| null | null | null | null | true | null |
5633
| null |
Default
| null | null |
null |
{
"abstract": " CaOFeS is a semiconducting oxysulfide with polar layered triangular\nstructure. Here a comprehensive theoretical study has been performed to reveal\nits physical properties, including magnetism, electronic structure, phase\ntransition, magnetodielectric effect, as well as optical absorption. Our\ncalculations confirm the Ising-like G-type antiferromagnetic ground state\ndriven by the next-nearest neighbor exchanges, which breaks the trigonal\nsymmetry and is responsible for the magnetodielectric effect driven by exchange\nstriction. In addition, a large coefficient of visible light absorption is\npredicted, which leads to promising photovoltaic effect with the maximum\nlight-to-electricity energy conversion efficiency up to 24.2%.\n",
"title": "Exchange striction driven magnetodielectric effect and potential photovoltaic effect in polar CaOFeS"
}
| null | null | null | null | true | null |
5634
| null |
Default
| null | null |
null |
{
"abstract": " Low-frequency polarisation observations of pulsars, facilitated by\nnext-generation radio telescopes, provide powerful probes of astrophysical\nplasmas that span many orders of magnitude in magnetic field strength and\nscale: from pulsar magnetospheres to intervening magneto-ionic plasmas\nincluding the ISM and the ionosphere. Pulsar magnetospheres with teragauss\nfield strengths can be explored through their numerous emission phenomena\nacross multiple frequencies, the mechanism behind which remains elusive.\nPrecise dispersion and Faraday rotation measurements towards a large number of\npulsars probe the three-dimensional large-scale (and eventually small-scale)\nstructure of the Galactic magnetic field, which plays a role in many\nastrophysical processes, but is not yet well understood, especially towards the\nGalactic halo. We describe some results and ongoing work from the Low Frequency\nArray (LOFAR) and the Murchison Widefield Array (MWA) radio telescopes in these\nareas. These and other pathfinder and precursor telescopes have reinvigorated\nlow-frequency science and build towards the Square Kilometre Array (SKA), which\nwill make significant advancements in studies of astrophysical magnetic fields\nin the next 50 years.\n",
"title": "Studying Magnetic Fields using Low-frequency Pulsar Observations"
}
| null | null |
[
"Physics"
] | null | true | null |
5635
| null |
Validated
| null | null |
null |
{
"abstract": " Recent years have witnessed the rise of many successful e-commerce\nmarketplace platforms like the Amazon marketplace, AirBnB, Uber/Lyft, and\nUpwork, where a central platform mediates economic transactions between buyers\nand sellers. Motivated by these platforms, we formulate a set of facility\nlocation problems that we term Two-sided Facility location. In our model,\nagents arrive at nodes in an underlying metric space, where the metric distance\nbetween any buyer and seller captures the quality of the corresponding match.\nThe platform posts prices and wages at the nodes, and opens a set of facilities\nto route the agents to. The agents at any facility are assumed to be matched.\nThe platform ensures high match quality by imposing a distance constraint\nbetween a node and the facilities it is routed to. It ensures high service\navailability by ensuring flow to the facility is at least a pre-specified lower\nbound. Subject to these constraints, the goal of the platform is to maximize\nthe social surplus (or gains from trade) subject to weak budget balance, i.e.,\nprofit being non-negative.\nWe present an approximation algorithm for this problem that yields a $(1 +\n\\epsilon)$ approximation to surplus for any constant $\\epsilon > 0$, while\nrelaxing the match quality (i.e., maximum distance of any match) by a constant\nfactor. We use an LP rounding framework that easily extends to other objectives\nsuch as maximizing volume of trade or profit.\nWe justify our models by considering a dynamic marketplace setting where\nagents arrive according to a stochastic process and have finite patience (or\ndeadlines) for being matched. We perform queueing analysis to show that for\npolicies that route agents to facilities and match them, ensuring a low\nabandonment probability of agents reduces to ensuring sufficient flow arrives\nat each facility.\n",
"title": "Two-sided Facility Location"
}
| null | null | null | null | true | null |
5636
| null |
Default
| null | null |
null |
{
"abstract": " In a market with a rough or Markovian mean-reverting stochastic volatility\nthere is no perfect hedge. Here it is shown how various delta-type hedging\nstrategies perform and can be evaluated in such markets. A precise\ncharacterization of the hedging cost, the replication cost caused by the\nvolatility fluctuations, is presented in an asymptotic regime of rapid mean\nreversion for the volatility fluctuations. The optimal dynamic asset based\nhedging strategy in the considered regime is identified as the so-called\n`practitioners' delta hedging scheme. It is moreover shown that the\nperformances of the delta-type hedging schemes are essentially independent of\nthe regularity of the volatility paths in the considered regime and that the\nhedging costs are related to a vega risk martingale whose magnitude is\nproportional to a new market risk parameter.\n",
"title": "Optimal hedging under fast-varying stochastic volatility"
}
| null | null | null | null | true | null |
5637
| null |
Default
| null | null |
null |
{
"abstract": " Due to their interdisciplinary nature, devices for controlled-environment\nagriculture have the possibility to turn into ideal tools not only to conduct\nresearch on plant phenology but also to create curricula in a wide range of\ndisciplines. Controlled-environment devices are increasing their\nfunctionalities as well as improving their accessibility. Traditionally,\nbuilding one of these devices from scratch implies knowledge in fields such as\nmechanical engineering, digital electronics, programming, and energy\nmanagement. However, the requirements of an effective controlled environment\ndevice for personal use brings new constraints and challenges. This paper\npresents the OpenAg Personal Food Computer (PFC); a low cost desktop size\nplatform, which not only targets plant phenology researchers but also\nhobbyists, makers, and teachers from elementary to high-school levels (K-12).\nThe PFC is completely open-source and it is intended to become a tool that can\nbe used for collective data sharing and plant growth analysis. Thanks to its\nmodular design, the PFC can be used in a large spectrum of activities.\n",
"title": "Personal Food Computer: A new device for controlled-environment agriculture"
}
| null | null | null | null | true | null |
5638
| null |
Default
| null | null |
null |
{
"abstract": " Nonlinear oscillators are a key modelling tool in many applications. The\ninfluence of annealed noise on nonlinear oscillators has been studied\nintensively. It can induce effects in nonlinear oscillators not present in the\ndeterministic setting. Yet, there is no theory regarding the quenched noise\nscenario of random parameters sampled on fixed time intervals, although this\nsituation is often a lot more natural. Here we study a paradigmatic nonlinear\noscillator of van-der-Pol/FitzHugh-Nagumo type under quenched noise as a\npiecewise-deterministic Markov process. There are several interesting effects\nsuch as period shifts and new different trapped types of small-amplitude\noscillations, which can be captured analytically. Furthermore, we numerically\ndiscover quenched resonance and show that it differs significantly from\nprevious finite-noise optimality resonance effects. This demonstrates that\nquenched oscillatorscan be viewed as a new building block of nonlinear\ndynamics.\n",
"title": "Quenched Noise and Nonlinear Oscillations in Bistable Multiscale Systems"
}
| null | null | null | null | true | null |
5639
| null |
Default
| null | null |
null |
{
"abstract": " The recently proposed Multi-Layer Convolutional Sparse Coding (ML-CSC) model,\nconsisting of a cascade of convolutional sparse layers, provides a new\ninterpretation of Convolutional Neural Networks (CNNs). Under this framework,\nthe computation of the forward pass in a CNN is equivalent to a pursuit\nalgorithm aiming to estimate the nested sparse representation vectors -- or\nfeature maps -- from a given input signal. Despite having served as a pivotal\nconnection between CNNs and sparse modeling, a deeper understanding of the\nML-CSC is still lacking: there are no pursuit algorithms that can serve this\nmodel exactly, nor are there conditions to guarantee a non-empty model. While\none can easily obtain signals that approximately satisfy the ML-CSC\nconstraints, it remains unclear how to simply sample from the model and, more\nimportantly, how one can train the convolutional filters from real data.\nIn this work, we propose a sound pursuit algorithm for the ML-CSC model by\nadopting a projection approach. We provide new and improved bounds on the\nstability of the solution of such pursuit and we analyze different practical\nalternatives to implement this in practice. We show that the training of the\nfilters is essential to allow for non-trivial signals in the model, and we\nderive an online algorithm to learn the dictionaries from real data,\neffectively resulting in cascaded sparse convolutional layers. Last, but not\nleast, we demonstrate the applicability of the ML-CSC model for several\napplications in an unsupervised setting, providing competitive results. Our\nwork represents a bridge between matrix factorization, sparse dictionary\nlearning and sparse auto-encoders, and we analyze these connections in detail.\n",
"title": "Multi-Layer Convolutional Sparse Modeling: Pursuit and Dictionary Learning"
}
| null | null | null | null | true | null |
5640
| null |
Default
| null | null |
null |
{
"abstract": " This paper considers the planar figure of a combinatorial polytope or\ntessellation identified by the Coxeter symbol $k_{i,j}$ , inscribed in a conic,\nsatisfying the geometric constraint that each octahedral cell has a centre.\nThis realisation exists, and is movable, on account of some constraints being\nsatisfied as a consequence of the others. A close connection to the birational\ngroup found originally by Coble in the different context of invariants for sets\nof points in projective space, allows to specify precisely a determining subset\nof vertices that may be freely chosen. This gives a unified geometric view of\ncertain integrable discrete systems in one, two and three dimensions. Making\ncontact with previous geometric accounts in the case of three dimensions, it is\nshown how the figure also manifests as a configuration of circles generalising\nthe Clifford lattices, and how it can be applied to construct the spatial\npoint-line configurations called the Desargues maps.\n",
"title": "Coble's group and the integrability of the Gosset-Elte polytopes and tessellations"
}
| null | null | null | null | true | null |
5641
| null |
Default
| null | null |
null |
{
"abstract": " Deep Learning (DL) methods show very good performance when trained on large,\nbalanced data sets. However, many practical problems involve imbalanced data\nsets, or/and classes with a small number of training samples. The performance\nof DL methods as well as more traditional classifiers drops significantly in\nsuch settings. Most of the existing solutions for imbalanced problems focus on\ncustomizing the data for training. A more principled solution is to use mixed\nHinge-Minimax risk [19] specifically designed to solve binary problems with\nimbalanced training sets. Here we propose a Latent Hinge Minimax (LHM) risk and\na training algorithm that generalizes this paradigm to an ensemble of\nhyperplanes that can form arbitrary complex, piecewise linear boundaries. To\nextract good features, we combine LHM model with CNN via transfer learning. To\nsolve multi-class problem we map pre-trained category-specific LHM classifiers\nto a multi-class neural network and adjust the weights with very fast tuning.\nLHM classifier enables the use of unlabeled data in its training and the\nmapping allows for multi-class inference, resulting in a classifier that\nperforms better than alternatives when trained on a small number of training\nsamples.\n",
"title": "Latent Hinge-Minimax Risk Minimization for Inference from a Small Number of Training Samples"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5642
| null |
Validated
| null | null |
null |
{
"abstract": " Both humans and the sensors on an autonomous vehicle have limited sensing\ncapabilities. When these limitations coincide with scenarios involving\nvulnerable road users, it becomes important to account for these limitations in\nthe motion planner. For the scenario of an occluded pedestrian crosswalk, the\nspeed of the approaching vehicle should be a function of the amount of\nuncertainty on the roadway. In this work, the longitudinal controller is\nformulated as a partially observable Markov decision process and dynamic\nprogramming is used to compute the control policy. The control policy scales\nthe speed profile to be used by a model predictive steering controller.\n",
"title": "Autonomous Vehicle Speed Control for Safe Navigation of Occluded Pedestrian Crosswalk"
}
| null | null | null | null | true | null |
5643
| null |
Default
| null | null |
null |
{
"abstract": " Competing risks data arise frequently in clinical trials. When the\nproportional subdistribution hazard assumption is violated or two cumulative\nincidence function (CIF) curves cross, rather than comparing the overall\ntreatment effects, researchers may be interested in focusing on a comparison of\nclinical utility at some fixed time points. This paper extend a series of tests\nthat are constructed based on a pseudo-value regression technique or different\ntransformation functions for CIFs and their variances based on Gaynor's or\nAalen's work, and the differences among CIFs at a given time point are\ncompared.\n",
"title": "Statistical inference methods for cumulative incidence function curves at a fixed point in time"
}
| null | null | null | null | true | null |
5644
| null |
Default
| null | null |
null |
{
"abstract": " We compare and contrast the statistical physics and quantum physics inspired\napproaches for unsupervised generative modeling of classical data. The two\napproaches represent probabilities of observed data using energy-based models\nand quantum states respectively.Classical and quantum information patterns of\nthe target datasets therefore provide principled guidelines for structural\ndesign and learning in these two approaches. Taking the restricted Boltzmann\nmachines (RBM) as an example, we analyze the information theoretical bounds of\nthe two approaches. We verify our reasonings by comparing the performance of\nRBMs of various architectures on the standard MNIST datasets.\n",
"title": "Information Perspective to Probabilistic Modeling: Boltzmann Machines versus Born Machines"
}
| null | null | null | null | true | null |
5645
| null |
Default
| null | null |
null |
{
"abstract": " We construct a cofibration category structure on the category of closure\nspaces $\\mathbf{Cl}$, the category whose objects are sets endowed with a\nČech closure operator and whose morphisms are the continuous maps between\nthem. We then study various closure structures on metric spaces, graphs, and\nsimplicial complexes, showing how each case gives rise to an interesting\nhomotopy theory. In particular, we show that there exists a natural family of\nclosure structures on metric spaces which produces a non-trivial homotopy\ntheory for finite metric spaces, i.e. point clouds, the spaces of interest in\ntopological data analysis. We then give a closure structure to graphs and\nsimplicial complexes which may be used to construct a new combinatorial (as\nopposed to topological) homotopy theory for each skeleton of those spaces. We\nshow that there is a Seifert-van Kampen theorem for closure spaces, a\nwell-defined notion of persistent homotopy and an associated interleaving\ndistance, and, as an illustration of the difference with the topological\nsetting, we calculate the fundamental group for the circle and the wedge of\ncircles endowed with different closure structures.\n",
"title": "A Cofibration Category on Closure Spaces"
}
| null | null | null | null | true | null |
5646
| null |
Default
| null | null |
null |
{
"abstract": " Open questions with respect to the computational complexity of linear CNF\nformulas in connection with regularity and uniformity are addressed. In\nparticular it is proven that any l-regular monotone CNF formula is\nXSAT-unsatisfiable if its number of clauses m is not a multiple of l. For exact\nlinear formulas one finds surprisingly that l-regularity implies k-uniformity,\nwith m = 1 + k(l-1)) and allowed k-values obey k(k-1) = 0 (mod l). Then the\ncomputational complexity of the class of monotone exact linear and l-regular\nCNF formulas with respect to XSAT can be determined: XSAT-satisfiability is\neither trivial, if m is not a multiple of l, or it can be decided in\nsub-exponential time, namely O(exp(n^^1/2)). Sub-exponential time behaviour for\nthe wider class of regular and uniform linear CNF formulas can be shown for\ncertain subclasses.\n",
"title": "XSAT of Linear CNF Formulas"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
5647
| null |
Validated
| null | null |
null |
{
"abstract": " This paper proposes Drone Squadron Optimization, a new self-adaptive\nmetaheuristic for global numerical optimization which is updated online by a\nhyper-heuristic. DSO is an artifact-inspired technique, as opposed to many\nalgorithms used nowadays, which are nature-inspired. DSO is very flexible\nbecause it is not related to behaviors or natural phenomena. DSO has two core\nparts: the semi-autonomous drones that fly over a landscape to explore, and the\nCommand Center that processes the retrieved data and updates the drones'\nfirmware whenever necessary. The self-adaptive aspect of DSO in this work is\nthe perturbation/movement scheme, which is the procedure used to generate\ntarget coordinates. This procedure is evolved by the Command Center during the\nglobal optimization process in order to adapt DSO to the search landscape. DSO\nwas evaluated on a set of widely employed benchmark functions. The statistical\nanalysis of the results shows that the proposed method is competitive with the\nother methods in the comparison, the performance is promising, but several\nfuture improvements are planned.\n",
"title": "Drone Squadron Optimization: a Self-adaptive Algorithm for Global Numerical Optimization"
}
| null | null | null | null | true | null |
5648
| null |
Default
| null | null |
null |
{
"abstract": " The original ImageNet dataset is a popular large-scale benchmark for training\nDeep Neural Networks. Since the cost of performing experiments (e.g, algorithm\ndesign, architecture search, and hyperparameter tuning) on the original dataset\nmight be prohibitive, we propose to consider a downsampled version of ImageNet.\nIn contrast to the CIFAR datasets and earlier downsampled versions of ImageNet,\nour proposed ImageNet32$\\times$32 (and its variants ImageNet64$\\times$64 and\nImageNet16$\\times$16) contains exactly the same number of classes and images as\nImageNet, with the only difference that the images are downsampled to\n32$\\times$32 pixels per image (64$\\times$64 and 16$\\times$16 pixels for the\nvariants, respectively). Experiments on these downsampled variants are\ndramatically faster than on the original ImageNet and the characteristics of\nthe downsampled datasets with respect to optimal hyperparameters appear to\nremain similar. The proposed datasets and scripts to reproduce our results are\navailable at this http URL and\nthis https URL\n",
"title": "A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets"
}
| null | null | null | null | true | null |
5649
| null |
Default
| null | null |
null |
{
"abstract": " The particle Gibbs (PG) sampler is a Markov Chain Monte Carlo (MCMC)\nalgorithm, which uses an interacting particle system to perform the Gibbs\nsteps. Each Gibbs step consists of simulating a particle system conditioned on\none particle path. It relies on a conditional Sequential Monte Carlo (cSMC)\nmethod to create the particle system. We propose a novel interpretation of the\ncSMC algorithm as a perturbed Sequential Monte Carlo (SMC) method and apply\ntelescopic decompositions developed for the analysis of SMC algorithms\n\\cite{delmoral2004} to derive a bound for the distance between the expected\nsampled path from cSMC and the target distribution of the MCMC algorithm. This\ncan be used to get a uniform ergodicity result. In particular, we can show that\nthe mixing rate of cSMC can be kept constant by increasing the number of\nparticles linearly with the number of observations. Based on our decomposition,\nwe also prove a central limit theorem for the cSMC Algorithm, which cannot be\ndone using the approaches in \\cite{Andrieu2013} and \\cite{Lindsten2014}.\n",
"title": "Stability of Conditional Sequential Monte Carlo"
}
| null | null | null | null | true | null |
5650
| null |
Default
| null | null |
null |
{
"abstract": " The present paper is part of a series of articles dedicated to the existence\nof Arnold diffusion for cusp-residual perturbations of Tonelli Hamiltonians on\n$\\mathbb{A}^3$. Our goal here is to construct an abstract geometric framework\nthat can be used to prove the existence of diffusing orbits in the so-called a\npriori stable setting, once the preliminary geometric reductions are preformed.\nOur framework also applies, rather directly, to the a priori unstable setting.\nThe main geometric objects of interest are $3$-dimensional normally\nhyperbolic invariant cylinders with boundary, which in particular admit\nwell-defined stable and unstable manifolds. These enable us to define, in our\nsetting, chains of cylinders, i.e., finite, ordered families of cylinders in\nwhich each cylinder admits homoclinic connections, and any two consecutive\nelements in the family admit heteroclinic connections.\nOur main result is the existence of diffusing orbits drifting along such\nchains, under precise conditions on the dynamics on the cylinders, and on their\nhomoclinic and heteroclinic structure.\n",
"title": "Diffusion along chains of normally hyperbolic cylinders"
}
| null | null | null | null | true | null |
5651
| null |
Default
| null | null |
null |
{
"abstract": " Ellerman bombs (EBs) are a kind of solar activities that is suggested to\noccur in the lower atmosphere. Recent observations using the Interface Region\nImaging Spectrograph (IRIS) show connections of EBs and IRIS bombs (IBs),\nimplying that EBs might be heated to a much higher temperature ($8\\times10^{4}$\nK) than previous results. Here we perform a spectral analysis of the EBs\nsimultaneously observed by the Fast Imaging Solar Spectrograph (FISS) and IRIS.\nThe observational results show clear evidence of heating in the lower\natmosphere, indicated by the wing enhancement in H$\\alpha$, Ca II 8542 Å\nand Mg II triplet lines, and also by brightenings in the images of 1700 Å\nand 2832 Å ultraviolet continuum channels. Additionally, the Mg II triplet\nline intensity is correlated with that of H$\\alpha$ when the EB occurs,\nindicating the possibility to use the triplet as an alternative way to identify\nEBs. However, we do not find any signal in IRIS hotter lines (C II and Si IV).\nFor further analysis, we employ a two-cloud model to fit the two chromospheric\nlines (H$\\alpha$ and Ca II 8542 Å) simultaneously, and obtain a temperature\nenhancement of 2300 K for a strong EB. This temperature is among the highest of\nprevious modeling results while still insufficient to produce IB signatures at\nultraviolet wavelengths.\n",
"title": "Multi-wavelength Spectral Analysis of Ellerman Bombs Observed by FISS and IRIS"
}
| null | null | null | null | true | null |
5652
| null |
Default
| null | null |
null |
{
"abstract": " A detailed numerical study of the long time behaviour of dispersive shock\nwaves in solutions to the Kadomtsev-Petviashvili (KP) I equation is presented.\nIt is shown that modulated lump solutions emerge from the dispersive shock\nwaves. For the description of dispersive shock waves, Whitham modulation\nequations for KP are obtained. It is shown that the modulation equations near\nthe soliton line are hyperbolic for the KPII equation while they are elliptic\nfor the KPI equation leading to a focusing effect and the formation of lumps.\nSuch a behaviour is similar to the appearance of breathers for the focusing\nnonlinear Schrodinger equation in the semiclassical limit.\n",
"title": "Numerical study of the Kadomtsev--Petviashvili equation and dispersive shock waves"
}
| null | null | null | null | true | null |
5653
| null |
Default
| null | null |
null |
{
"abstract": " Let ${\\bf R}$ be the Pearson correlation matrix of $m$ normal random\nvariables. The Rao's score test for the independence hypothesis $H_0 : {\\bf R}\n= {\\bf I}_m$, where ${\\bf I}_m$ is the identity matrix of dimension $m$, was\nfirst considered by Schott (2005) in the high dimensional setting. In this\npaper, we study the asymptotic minimax power function of this test, under an\nasymptotic regime in which both $m$ and the sample size $n$ tend to infinity\nwith the ratio $m/n$ upper bounded by a constant. In particular, our result\nimplies that the Rao's score test is rate-optimal for detecting the dependency\nsignal $\\|{\\bf R} - {\\bf I}_m\\|_F$ of order $\\sqrt{m/n}$, where $\\|\\cdot\\|_F$\nis the matrix Frobenius norm.\n",
"title": "Asymptotic power of Rao's score test for independence in high dimensions"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
5654
| null |
Validated
| null | null |
null |
{
"abstract": " In the past years we have witnessed the emergence of the new discipline of\ncomputational social science, which promotes a new data-driven and\ncomputation-based approach to social sciences. In this article we discuss how\nthe availability of new technologies such as online social media and mobile\nsmartphones has allowed researchers to passively collect human behavioral data\nat a scale and a level of granularity that were just unthinkable some years\nago. We also discuss how these digital traces can then be used to prove (or\ndisprove) existing theories and develop new models of human behavior.\n",
"title": "Sensing and Modeling Human Behavior Using Social Media and Mobile Data"
}
| null | null | null | null | true | null |
5655
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we consider an interior transmission eigenvalue (ITE) problem\non some compact $C^{\\infty }$-Riemannian manifolds with a common smooth\nboundary. In particular, these manifolds may have different topologies, but we\nimpose some conditions of Riemannian metrics, indices of refraction and\nboundary conductivity parameters on the boundary. Then we prove the\ndiscreteness of the set of ITEs, the existence of infinitely many ITEs, and its\nWeyl type lower bound. For our settings, we can adopt the argument by\nLakshtanov and Vainberg, considering the Dirichlet-to-Neumann map. As an\napplication, we derive the existence of non-scattering energies for\ntime-harmonic acoustic equations. For the sake of simplicity, we consider the\nscattering theory on the Euclidean space. However, the argument is applicable\nfor certain kinds of non-compact manifolds with ends on which we can define the\nscattering matrix.\n",
"title": "Interior transmission eigenvalue problems on compact manifolds with boundary conductivity parameters"
}
| null | null | null | null | true | null |
5656
| null |
Default
| null | null |
null |
{
"abstract": " Unambiguous non-deterministic finite automata have intermediate expressive\npower and succinctness between deterministic and non-deterministic automata. It\nhas been conjectured that every unambiguous non-deterministic one-way finite\nautomaton (1UFA) recognizing some language L can be converted into a 1UFA\nrecognizing the complement of the original language L with polynomial increase\nin the number of states. We disprove this conjecture by presenting a family of\n1UFAs on a single-letter alphabet such that recognizing the complements of the\ncorresponding languages requires superpolynomial increase in the number of\nstates even for generic non-deterministic one-way finite automata. We also note\nthat both the languages and their complements can be recognized by sweeping\ndeterministic automata with a linear increase in the number of states.\n",
"title": "A superpolynomial lower bound for the size of non-deterministic complement of an unambiguous automaton"
}
| null | null | null | null | true | null |
5657
| null |
Default
| null | null |
null |
{
"abstract": " We propose a novel, semi-supervised approach towards domain taxonomy\ninduction from an input vocabulary of seed terms. Unlike all previous\napproaches, which typically extract direct hypernym edges for terms, our\napproach utilizes a novel probabilistic framework to extract hypernym\nsubsequences. Taxonomy induction from extracted subsequences is cast as an\ninstance of the minimumcost flow problem on a carefully designed directed\ngraph. Through experiments, we demonstrate that our approach outperforms\nstateof- the-art taxonomy induction approaches across four languages.\nImportantly, we also show that our approach is robust to the presence of noise\nin the input vocabulary. To the best of our knowledge, no previous approaches\nhave been empirically proven to manifest noise-robustness in the input\nvocabulary.\n",
"title": "Taxonomy Induction using Hypernym Subsequences"
}
| null | null | null | null | true | null |
5658
| null |
Default
| null | null |
null |
{
"abstract": " The purpose of a clickbait is to make a link so appealing that people click\non it. However, the content of such articles is often not related to the title,\nshows poor quality, and at the end leaves the reader unsatisfied.\nTo help the readers, the organizers of the clickbait challenge\n(this http URL) asked the participants to build a machine\nlearning model for scoring articles with respect to their \"clickbaitness\".\nIn this paper we propose to solve the clickbait problem with an ensemble of\nLinear SVM models, and our approach was tested successfully in the challenge:\nit showed great performance of 0.036 MSE and ranked 3rd among all the solutions\nto the contest.\n",
"title": "Identifying Clickbait Posts on Social Media with an Ensemble of Linear Models"
}
| null | null | null | null | true | null |
5659
| null |
Default
| null | null |
null |
{
"abstract": " Secure multi-party computation (MPC) enables a set of mutually distrusting\nparties to cooperatively compute, using a cryptographic protocol, a function\nover their private data. This paper presents Wys*, a new domain-specific\nlanguage (DSL) for writing mixed-mode MPCs. Wys* is an embedded DSL hosted in\nF*, a verification-oriented, effectful programming language. Wys* source\nprograms are essentially F* programs written in a custom MPC effect, meaning\nthat the programmers can use F*'s logic to verify the correctness and security\nproperties of their programs. To reason about the distributed runtime semantics\nof these programs, we formalize a deep embedding of Wys*, also in F*. We\nmechanize the necessary metatheory to prove that the properties verified for\nthe Wys* source programs carry over to the distributed, multi-party semantics.\nFinally, we use F*'s extraction to extract an interpreter that we have proved\nmatches this semantics, yielding a partially verified implementation. Wys* is\nthe first DSL to enable formal verification of MPC programs. We have\nimplemented several MPC protocols in Wys*, including private set intersection,\njoint median, and an MPC card dealing application, and have verified their\ncorrectness and security.\n",
"title": "WYS*: A DSL for Verified Secure Multi-party Computations"
}
| null | null | null | null | true | null |
5660
| null |
Default
| null | null |
null |
{
"abstract": " Multiple changes in Earth's climate system have been observed over the past\ndecades. Determining how likely each of these changes are to have been caused\nby human influence, is important for decision making on mitigation and\nadaptation policy. Here we describe an approach for deriving the probability\nthat anthropogenic forcings have caused a given observed change. The proposed\napproach is anchored into causal counterfactual theory (Pearl 2009) which has\nbeen introduced recently, and was in fact partly used already, in the context\nof extreme weather event attribution (EA). We argue that these concepts are\nalso relevant, and can be straightforwardly extended to, the context of\ndetection and attribution of long term trends associated to climate change\n(D&A). For this purpose, and in agreement with the principle of\n\"fingerprinting\" applied in the conventional D&A framework, a trajectory of\nchange is converted into an event occurrence defined by maximizing the causal\nevidence associated to the forcing under scrutiny. Other key assumptions used\nin the conventional D&A framework, in particular those related to numerical\nmodels error, can also be adapted conveniently to this approach. Our proposal\nthus allows to bridge the conventional framework with the standard causal\ntheory, in an attempt to improve the quantification of causal probabilities. An\nillustration suggests that our approach is prone to yield a significantly\nhigher estimate of the probability that anthropogenic forcings have caused the\nobserved temperature change, thus supporting more assertive causal claims.\n",
"title": "Probabilities of causation of climate changes"
}
| null | null | null | null | true | null |
5661
| null |
Default
| null | null |
null |
{
"abstract": " The field of Distributed Constraint Optimization has gained momentum in\nrecent years thanks to its ability to address various applications related to\nmulti-agent cooperation. While techniques to solve Distributed Constraint\nOptimization Problems (DCOPs) are abundant and have matured substantially since\nthe field inception, the number of DCOP realistic applications and benchmark\nused to asses the performance of DCOP algorithms is lagging behind. To contrast\nthis background we (i) introduce the Smart Home Device Scheduling (SHDS)\nproblem, which describe the problem of coordinating smart devices schedules\nacross multiple homes as a multi-agent system, (ii) detail the physical models\nadopted to simulate smart sensors, smart actuators, and homes environments, and\n(iii) introduce a DCOP realistic benchmark for SHDS problems.\n",
"title": "A Realistic Dataset for the Smart Home Device Scheduling Problem for DCOPs"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5662
| null |
Validated
| null | null |
null |
{
"abstract": " In principle a minimal extension of the standard model of Particle Physics,\nthe two Higgs doublet model, can be invoked to explain the scalar field\nresponsible of dark energy. The two doublets are in general mixed. After\ndiagonalization, the lightest CP-even Higgs and CP-odd Higgs are jointly taken\nto be the dark energy candidate. The dark energy obtained from Higgs fields in\nthis case is indistinguishable from the cosmological constant.\n",
"title": "The extra scalar degrees of freedom from the two Higgs doublet model for dark energy"
}
| null | null | null | null | true | null |
5663
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of transferring policies to the real world by\ntraining on a distribution of simulated scenarios. Rather than manually tuning\nthe randomization of simulations, we adapt the simulation parameter\ndistribution using a few real world roll-outs interleaved with policy training.\nIn doing so, we are able to change the distribution of simulations to improve\nthe policy transfer by matching the policy behavior in simulation and the real\nworld. We show that policies trained with our method are able to reliably\ntransfer to different robots in two real world tasks: swing-peg-in-hole and\nopening a cabinet drawer. The video of our experiments can be found at\nthis https URL\n",
"title": "Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience"
}
| null | null | null | null | true | null |
5664
| null |
Default
| null | null |
null |
{
"abstract": " We address the problem of learning vector representations for entities and\nrelations in Knowledge Graphs (KGs) for Knowledge Base Completion (KBC). This\nproblem has received significant attention in the past few years and multiple\nmethods have been proposed. Most of the existing methods in the literature use\na predefined characteristic scoring function for evaluating the correctness of\nKG triples. These scoring functions distinguish correct triples (high score)\nfrom incorrect ones (low score). However, their performance vary across\ndifferent datasets. In this work, we demonstrate that a simple neural network\nbased score function can consistently achieve near start-of-the-art performance\non multiple datasets. We also quantitatively demonstrate biases in standard\nbenchmark datasets, and highlight the need to perform evaluation spanning\nvarious datasets.\n",
"title": "Revisiting Simple Neural Networks for Learning Representations of Knowledge Graphs"
}
| null | null | null | null | true | null |
5665
| null |
Default
| null | null |
null |
{
"abstract": " Conditions for geometric ergodicity of multivariate autoregressive\nconditional heteroskedasticity (ARCH) processes, with the so-called BEKK (Baba,\nEngle, Kraft, and Kroner) parametrization, are considered. We show for a class\nof BEKK-ARCH processes that the invariant distribution is regularly varying. In\norder to account for the possibility of different tail indices of the\nmarginals, we consider the notion of vector scaling regular variation, in the\nspirit of Perfekt (1997, Advances in Applied Probability, 29, pp. 138-164). The\ncharacterization of the tail behavior of the processes is used for deriving the\nasymptotic properties of the sample covariance matrices.\n",
"title": "On the tail behavior of a class of multivariate conditionally heteroskedastic processes"
}
| null | null | null | null | true | null |
5666
| null |
Default
| null | null |
null |
{
"abstract": " It has recently been shown that the problem of testing global convexity of\npolynomials of degree four is {strongly} NP-hard, answering an open question of\nN.Z. Shor. This result is minimal in the degree of the polynomial when global\nconvexity is of concern. In a number of applications however, one is interested\nin testing convexity only over a compact region, most commonly a box (i.e.,\nhyper-rectangle). In this paper, we show that this problem is also strongly\nNP-hard, in fact for polynomials of degree as low as three. This result is\nminimal in the degree of the polynomial and in some sense justifies why\nconvexity detection in nonlinear optimization solvers is limited to quadratic\nfunctions or functions with special structure. As a byproduct, our proof shows\nthat the problem of testing whether all matrices in an interval family are\npositive semidefinite is strongly NP-hard. This problem, which was previously\nshown to be (weakly) NP-hard by Nemirovski, is of independent interest in the\ntheory of robust control.\n",
"title": "On the Complexity of Detecting Convexity over a Box"
}
| null | null | null | null | true | null |
5667
| null |
Default
| null | null |
null |
{
"abstract": " Several useful variance-reduced stochastic gradient algorithms, such as SVRG,\nSAGA, Finito, and SAG, have been proposed to minimize empirical risks with\nlinear convergence properties to the exact minimizer. The existing convergence\nresults assume uniform data sampling with replacement. However, it has been\nobserved in related works that random reshuffling can deliver superior\nperformance over uniform sampling and, yet, no formal proofs or guarantees of\nexact convergence exist for variance-reduced algorithms under random\nreshuffling. This paper makes two contributions. First, it resolves this open\nissue and provides the first theoretical guarantee of linear convergence under\nrandom reshuffling for SAGA; the argument is also adaptable to other\nvariance-reduced algorithms. Second, under random reshuffling, the paper\nproposes a new amortized variance-reduced gradient (AVRG) algorithm with\nconstant storage requirements compared to SAGA and with balanced gradient\ncomputations compared to SVRG. AVRG is also shown analytically to converge\nlinearly.\n",
"title": "Variance-Reduced Stochastic Learning under Random Reshuffling"
}
| null | null | null | null | true | null |
5668
| null |
Default
| null | null |
null |
{
"abstract": " Developing applications for interactive space is different from developing\ncross-platform applications for personal computing. Input, output, and\narchitectural variations in each interactive space introduce big overhead in\nterms of cost and time for developing, deploying and maintaining applications\nfor interactive spaces. Often, these applications become on-off experience tied\nto the deployed spaces. To alleviate this problem and enable rapid responsive\nspace design applications similar to responsive web design, we present CELIO\napplication development framework for interactive spaces. The framework is\nmicro services based and neatly decouples application and design specifications\nfrom hardware and architecture specifications of an interactive space. In this\npaper, we describe this framework and its implementation details. Also, we\nbriefly discuss the use cases developed using this framework.\n",
"title": "CELIO: An application development framework for interactive spaces"
}
| null | null | null | null | true | null |
5669
| null |
Default
| null | null |
null |
{
"abstract": " The increasing complexity of distribution network calls for advancement in\ndistribution system state estimation (DSSE) to monitor the operating conditions\nmore accurately. Sufficient number of measurements is imperative for a reliable\nand accurate state estimation. The limitation on the measurement devices is\ngenerally tackled with using the so-called pseudo measured data. However, the\nerrors in pseudo data by cur-rent techniques are quite high leading to a poor\nDSSE. As customer loads in distribution networks show high cross-correlation in\nvarious locations and over successive time steps, it is plausible that\ndeploying the spatial-temporal dependencies can improve the pseudo data\naccuracy and estimation. Although, the role of spatial dependency in DSSE has\nbeen addressed in the literature, one can hardly find an efficient DSSE\nframework capable of incorporating temporal dependencies present in customer\nloads. Consequently, to obtain a more efficient and accurate state estimation,\nwe propose a new non-iterative DSSE framework to involve spatial-temporal\ndependencies together. The spatial-temporal dependencies are modeled by\nconditional multivariate complex Gaussian distributions and are studied for\nboth static and real-time state estimations, where information at preceding\ntime steps are employed to increase the accuracy of DSSE. The efficiency of the\nproposed approach is verified based on quality and accuracy indices, standard\ndeviation and computational time. Two balanced medium voltage (MV) and one\nunbalanced low voltage (LV) distribution case studies are used for evaluations.\n",
"title": "Single Iteration Conditional Based DSE Considering Spatial and Temporal Correlation"
}
| null | null | null | null | true | null |
5670
| null |
Default
| null | null |
null |
{
"abstract": " Network embedding methods aim at learning low-dimensional latent\nrepresentation of nodes in a network. While achieving competitive performance\non a variety of network inference tasks such as node classification and link\nprediction, these methods treat the relations between nodes as a binary\nvariable and ignore the rich semantics of edges. In this work, we attempt to\nlearn network embeddings which simultaneously preserve network structure and\nrelations between nodes. Experiments on several real-world networks illustrate\nthat by considering different relations between different node pairs, our\nmethod is capable of producing node embeddings of higher quality than a number\nof state-of-the-art network embedding methods, as evaluated on a challenging\nmulti-label node classification task.\n",
"title": "Enhanced Network Embeddings via Exploiting Edge Labels"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5671
| null |
Validated
| null | null |
null |
{
"abstract": " Underactuation is ubiquitous in human locomotion and should be ubiquitous in\nbipedal robotic locomotion as well. This chapter presents a coherent theory for\nthe design of feedback controllers that achieve stable walking gaits in\nunderactuated bipedal robots. Two fundamental tools are introduced, virtual\nconstraints and hybrid zero dynamics. Virtual constraints are relations on the\nstate variables of a mechanical model that are imposed through a time-invariant\nfeedback controller. One of their roles is to synchronize the robot's joints to\nan internal gait phasing variable. A second role is to induce a low dimensional\nsystem, the zero dynamics, that captures the underactuated aspects of a robot's\nmodel, without any approximations. To enhance intuition, the relation between\nphysical constraints and virtual constraints is first established. From here,\nthe hybrid zero dynamics of an underactuated bipedal model is developed, and\nits fundamental role in the design of asymptotically stable walking motions is\nestablished. The chapter includes numerous references to robots on which the\nhighlighted techniques have been implemented.\n",
"title": "Virtual Constraints and Hybrid Zero Dynamics for Realizing Underactuated Bipedal Locomotion"
}
| null | null | null | null | true | null |
5672
| null |
Default
| null | null |
null |
{
"abstract": " We study a three-component fermionic fluid in an optical lattice in a regime\nof intermediate-to- strong interactions allowing for Raman processes connecting\nthe different components, similar to those used to create artificial gauge\nfields (AGF). Using Dynamical Mean-Field Theory we show that the combined\neffect of interactions and AGFs induces a variety of anomalous phases in which\ndifferent components of the fermionic fluid display qualitative differences,\ni.e., the physics is flavor-selective. Remarkably, the different components can\ndisplay huge differences in the correlation effects, measured by their\neffective masses and non-monotonic behavior of their occupation number as a\nfunction of the chemical potential, signaling a sort of selective instability\nof the overall stable quantum fluid.\n",
"title": "Selective insulators and anomalous responses in correlated fermions with synthetic extra dimensions"
}
| null | null | null | null | true | null |
5673
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a widely applicable approach to solving (multi-marginal,\nmartingale) optimal transport and related problems via neural networks. The\ncore idea is to penalize the optimization problem in its dual formulation and\nreduce it to a finite dimensional one which corresponds to optimizing a neural\nnetwork with smooth objective function. We present numerical examples from\noptimal transport, martingale optimal transport, portfolio optimization under\nuncertainty and generative adversarial networks that showcase the generality\nand effectiveness of the approach.\n",
"title": "Computation of optimal transport and related hedging problems via penalization and neural networks"
}
| null | null | null | null | true | null |
5674
| null |
Default
| null | null |
null |
{
"abstract": " Both natural and artificial small-scale swimmers may often self-propel in\nenvironments subject to complex geometrical constraints. While most past\ntheoretical work on low-Reynolds number locomotion addressed idealised\ngeometrical situations, not much is known on the motion of swimmers in\nheterogeneous environments. As a first theoretical model, we investigate\nnumerically the behaviour of a single spherical micro-swimmer located in an\ninfinite, periodic body-centred cubic lattice consisting of rigid inert spheres\nof the same size as the swimmer. Running a large number of simulations we\nuncover the phase diagram of possible trajectories as a function of the\nstrength of the swimming actuation and the packing density of the lattice. We\nthen use hydrodynamic theory to rationalise our computational results and show\nin particular how the far-field nature of the swimmer (pusher vs. puller)\ngoverns even the behaviour at high volume fractions.\n",
"title": "Active particles in periodic lattices"
}
| null | null | null | null | true | null |
5675
| null |
Default
| null | null |
null |
{
"abstract": " We present the first measurements of tritium beta-decay spectrum in the\nelectron energy range 16-18.6 keV. The goal is to find distortions which may\ncorrespond to the presence of a heavy sterile neutrinos. A possible\ncontribution of this kind would manifest itself as a kink in the spectrum with\na similar shape but with end point shifted by the value of a heavy neutrino\nmass. We set a new upper limits to the neutrino mixing matrix element U^2_{e4}\nwhich improve existing limits by a factor from 2 to 5 in the mass range 0.1-2\nkeV.\n",
"title": "First measeurements in search for keV-sterile neutrino in tritium beta-decay by Troitsk nu-mass experiment"
}
| null | null | null | null | true | null |
5676
| null |
Default
| null | null |
null |
{
"abstract": " These notes constitute chapter 7 from \"l'Ecole de Physique des Houches\"\nSession CIII, August 2014 dedicated to Topological Aspects of Condensed matter\nphysics. The tenfold way in quasi-one-dimensional space is presented. The\nmethod of chiral Abelian bosonization is reviewed. It is then applied to the\nstability analysis for the edge theory in symmetry class AII, and for the\nconstruction of two-dimensional topological phases from coupled wires.\n",
"title": "Fractional Abelian topological phases of matter for fermions in two-dimensional space"
}
| null | null | null | null | true | null |
5677
| null |
Default
| null | null |
null |
{
"abstract": " For each $n$, we construct a separable metric space $\\mathbb{U}_n$ that is\nuniversal in the coarse category of separable metric spaces with asymptotic\ndimension ($\\mathop{asdim}$) at most $n$ and universal in the uniform category\nof separable metric spaces with uniform dimension ($\\mathop{udim}$) at most\n$n$. Thus, $\\mathbb{U}_n$ serves as a universal space for dimension $n$ in both\nthe large-scale and infinitesimal topology. More precisely, we prove:\n\\[\n\\mathop{asdim} \\mathbb{U}_n = \\mathop{udim} \\mathbb{U}_n = n\n\\] and such that for each separable metric space $X$,\na) if $\\mathop{asdim} X \\leq n$, then $X$ is coarsely equivalent to a subset\nof $\\mathbb{U}_n$;\nb) if $\\mathop{udim} X \\leq n$, then $X$ is uniformly homeomorphic to a\nsubset of $\\mathbb{U}_n$.\n",
"title": "A new construction of universal spaces for asymptotic dimension"
}
| null | null |
[
"Mathematics"
] | null | true | null |
5678
| null |
Validated
| null | null |
null |
{
"abstract": " Bike sharing is a vital component of a modern multi-modal transportation\nsystem. However, its implementation can lead to bike supply-demand imbalance\ndue to fluctuating spatial and temporal demands. This study proposes a\ncomprehensive framework to develop optimal dynamic bike rebalancing strategies\nin a large bike sharing network. It consists of three components, including a\nstation-level pick-up/drop-off prediction model, station clustering model, and\ncapacitated location-routing optimization model. For the first component, we\npropose a powerful deep learning model called graph convolution neural network\nmodel (GCNN) with data-driven graph filter (DDGF), which can automatically\nlearn the hidden spatial-temporal correlations among stations to provide more\naccurate predictions; for the second component, we apply a graph clustering\nalgorithm labeled the Community Detection algorithm to cluster stations that\nlocate geographically close to each other and have a small net demand gap;\nlast, a capacitated location-routing problem (CLRP) is solved to deal with the\ncombination of two types of decision variables: the locations of bike\ndistribution centers and the design of distribution routes for each cluster.\n",
"title": "A Comprehensive Framework for Dynamic Bike Rebalancing in a Large Bike Sharing Network"
}
| null | null | null | null | true | null |
5679
| null |
Default
| null | null |
null |
{
"abstract": " To model relaxed memory, we propose confusion-free event structures over an\nalphabet with a justification relation. Executions are modeled by justified\nconfigurations, where every read event has a justifying write event.\nJustification alone is too weak a criterion, since it allows cycles of the kind\nthat result in so-called thin-air reads. Acyclic justification forbids such\ncycles, but also invalidates event reorderings that result from compiler\noptimizations and dynamic instruction scheduling. We propose the notion of\nwell-justification, based on a game-like model, which strikes a middle ground.\nWe show that well-justified configurations satisfy the DRF theorem: in any\ndata-race free program, all well-justified configurations are sequentially\nconsistent. We also show that rely-guarantee reasoning is sound for\nwell-justified configurations, but not for justified configurations. For\nexample, well-justified configurations are type-safe.\nWell-justification allows many, but not all reorderings performed by relaxed\nmemory. In particular, it fails to validate the commutation of independent\nreads. We discuss variations that may address these shortcomings.\n",
"title": "On Thin Air Reads: Towards an Event Structures Model of Relaxed Memory"
}
| null | null | null | null | true | null |
5680
| null |
Default
| null | null |
null |
{
"abstract": " Tensor decompositions such as the canonical format and the tensor train\nformat have been widely utilized to reduce storage costs and operational\ncomplexities for high-dimensional data, achieving linear scaling with the input\ndimension instead of exponential scaling. In this paper, we investigate even\nlower storage-cost representations in the tensor ring format, which is an\nextension of the tensor train format with variable end-ranks. Firstly, we\nintroduce two algorithms for converting a tensor in full format to tensor ring\nformat with low storage cost. Secondly, we detail a rounding operation for\ntensor rings and show how this requires new definitions of common linear\nalgebra operations in the format to obtain storage-cost savings. Lastly, we\nintroduce algorithms for transforming the graph structure of graph-based tensor\nformats, with orders of magnitude lower complexity than existing literature.\nThe efficiency of all algorithms is demonstrated on a number of numerical\nexamples, and we achieve up to more than an order of magnitude higher\ncompression ratios than previous approaches to using the tensor ring format.\n",
"title": "Tensor ring decomposition"
}
| null | null | null | null | true | null |
5681
| null |
Default
| null | null |
null |
{
"abstract": " Even though the forecasting literature agrees that aggregating multiple\npredictions of some future outcome typically outperforms the individual\npredictions, there is no general consensus about the right way to do this. Most\ncommon aggregators are means, defined loosely as aggregators that always remain\nbetween the smallest and largest predictions. Examples include the arithmetic\nmean, trimmed means, median, mid-range, and many other measures of central\ntendency. If the forecasters use different information, the aggregator ideally\ncombines their information into a consensus without losing or distorting any of\nit. An aggregator that achieves this is considered efficient. Unfortunately,\nour results show that if the forecasters use their information accurately, an\naggregator that always remains strictly between the smallest and largest\npredictions is never efficient in practice. A similar result holds even if the\nideal predictions are distorted with random error that is centered at zero. If\nthese noisy predictions are aggregated with a similar notion of centrality,\nthen, under some mild conditions, the aggregator is asymptotically inefficient.\n",
"title": "Combining Information from Multiple Forecasters: Inefficiency of Central Tendency"
}
| null | null | null | null | true | null |
5682
| null |
Default
| null | null |
null |
{
"abstract": " Recent progress in applying machine learning for jet physics has been built\nupon an analogy between calorimeters and images. In this work, we present a\nnovel class of recursive neural networks built instead upon an analogy between\nQCD and natural languages. In the analogy, four-momenta are like words and the\nclustering history of sequential recombination jet algorithms is like the\nparsing of a sentence. Our approach works directly with the four-momenta of a\nvariable-length set of particles, and the jet-based tree structure varies on an\nevent-by-event basis. Our experiments highlight the flexibility of our method\nfor building task-specific jet embeddings and show that recursive architectures\nare significantly more accurate and data efficient than previous image-based\nnetworks. We extend the analogy from individual jets (sentences) to full events\n(paragraphs), and show for the first time an event-level classifier operating\non all the stable particles produced in an LHC event.\n",
"title": "QCD-Aware Recursive Neural Networks for Jet Physics"
}
| null | null | null | null | true | null |
5683
| null |
Default
| null | null |
null |
{
"abstract": " For a skew-symmetrizable cluster algebra $\\mathcal A_{t_0}$ with principal\ncoefficients at $t_0$, we prove that each seed $\\Sigma_t$ of $\\mathcal A_{t_0}$\nis uniquely determined by its {\\bf C-matrix}, which was proposed by Fomin and\nZelevinsky in \\cite{FZ3} as a conjecture. Our proof is based on the fact that\nthe positivity of cluster variables and sign-coherence of $c$-vectors hold for\n$\\mathcal A_{t_0}$, which was actually verified in \\cite{GHKK}. More discussion\nis given in the sign-skew-symmetric case so as to obtain a conclusion as weak\nversion of the conjecture in this general case.\n",
"title": "A conjecture on $C$-matrices of cluster algebras"
}
| null | null | null | null | true | null |
5684
| null |
Default
| null | null |
null |
{
"abstract": " Link prediction is one of the fundamental problems in computational social\nscience. A particularly common means to predict existence of unobserved links\nis via structural similarity metrics, such as the number of common neighbors;\nnode pairs with higher similarity are thus deemed more likely to be linked.\nHowever, a number of applications of link prediction, such as predicting links\nin gang or terrorist networks, are adversarial, with another party incentivized\nto minimize its effectiveness by manipulating observed information about the\nnetwork. We offer a comprehensive algorithmic investigation of the problem of\nattacking similarity-based link prediction through link deletion, focusing on\ntwo broad classes of such approaches, one which uses only local information\nabout target links, and another which uses global network information. While we\nshow several variations of the general problem to be NP-Hard for both local and\nglobal metrics, we exhibit a number of well-motivated special cases which are\ntractable. Additionally, we provide principled and empirically effective\nalgorithms for the intractable cases, in some cases proving worst-case\napproximation guarantees.\n",
"title": "Attacking Similarity-Based Link Prediction in Social Networks"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5685
| null |
Validated
| null | null |
null |
{
"abstract": " In 1979 Valiant showed that the complexity class VP_e of families with\npolynomially bounded formula size is contained in the class VP_s of families\nthat have algebraic branching programs (ABPs) of polynomially bounded size.\nMotivated by the problem of separating these classes we study the topological\nclosure VP_e-bar, i.e. the class of polynomials that can be approximated\narbitrarily closely by polynomials in VP_e. We describe VP_e-bar with a\nstrikingly simple complete polynomial (in characteristic different from 2)\nwhose recursive definition is similar to the Fibonacci numbers. Further\nunderstanding this polynomial seems to be a promising route to new formula\nlower bounds.\nOur methods are rooted in the study of ABPs of small constant width. In 1992\nBen-Or and Cleve showed that formula size is polynomially equivalent to width-3\nABP size. We extend their result (in characteristic different from 2) by\nshowing that approximate formula size is polynomially equivalent to approximate\nwidth-2 ABP size. This is surprising because in 2011 Allender and Wang gave\nexplicit polynomials that cannot be computed by width-2 ABPs at all! The\ndetails of our construction lead to the aforementioned characterization of\nVP_e-bar.\nAs a natural continuation of this work we prove that the class VNP can be\ndescribed as the class of families that admit a hypercube summation of\npolynomially bounded dimension over a product of polynomially many affine\nlinear forms. This gives the first separations of algebraic complexity classes\nfrom their nondeterministic analogs.\n",
"title": "On algebraic branching programs of small width"
}
| null | null | null | null | true | null |
5686
| null |
Default
| null | null |
null |
{
"abstract": " A previously designed cryogenic thermal heat switch for space applications\nhas been optimized for low mass, high structural stability, and reliability.\nThe heat switch makes use of the large linear thermal expansion coefficient\n(CTE) of the thermoplastic UHMW-PE for actuation. A structure model, which\nincludes the temperature dependent properties of the actuator, is derived to be\nable to predict the contact pressure between the switch parts. This pressure\nwas used in a thermal model in order to predict the switch performance under\ndifferent heat loads and operating temperatures. The two models were used to\noptimize the mass and stability of the switch. Its reliability was proven by\ncyclic actuation of the switch and by shaker tests.\n",
"title": "A lightweight thermal heat switch for redundant cryocooling on satellites"
}
| null | null | null | null | true | null |
5687
| null |
Default
| null | null |
null |
{
"abstract": " Multiple generalized additive models (GAMs) are a type of distributional\nregression wherein parameters of probability distributions depend on predictors\nthrough smooth functions, with selection of the degree of smoothness via $L_2$\nregularization. Multiple GAMs allow finer statistical inference by\nincorporating explanatory information in any or all of the parameters of the\ndistribution. Owing to their nonlinearity, flexibility and interpretability,\nGAMs are widely used, but reliable and fast methods for automatic smoothing in\nlarge datasets are still lacking, despite recent advances. We develop a general\nmethodology for automatically learning the optimal degree of $L_2$\nregularization for multiple GAMs using an empirical Bayes approach. The smooth\nfunctions are penalized by different amounts, which are learned simultaneously\nby maximization of a marginal likelihood through an approximate\nexpectation-maximization algorithm that involves a double Laplace approximation\nat the E-step, and leads to an efficient M-step. Empirical analysis shows that\nthe resulting algorithm is numerically stable, faster than all existing methods\nand achieves state-of-the-art accuracy. For illustration, we apply it to an\nimportant and challenging problem in the analysis of extremal data.\n",
"title": "Fast Automatic Smoothing for Generalized Additive Models"
}
| null | null | null | null | true | null |
5688
| null |
Default
| null | null |
null |
{
"abstract": " We show that elongated magnetic skyrmions can host Majorana bound states in a\nproximity-coupled two-dimensional electron gas sandwiched between a chiral\nmagnet and an $s$-wave superconductor. Our proposal requires stable skyrmions\nwith unit topological charge, which can be realized in a wide range of\nmultilayer magnets, and allows quantum information transfer by using standard\nmethods in spintronics via skyrmion motion. We also show how braiding\noperations can be realized in our proposal.\n",
"title": "Stabilization and control of Majorana bound states with elongated skyrmions"
}
| null | null | null | null | true | null |
5689
| null |
Default
| null | null |
null |
{
"abstract": " We consider the family of all meromorphic functions $f$ of the form $$\nf(z)=\\frac{1}{z}+b_0+b_1z+b_2z^2+\\cdots $$ analytic and locally univalent in\nthe puncture disk $\\mathbb{D}_0:=\\{z\\in\\mathbb{C}:\\,0<|z|<1\\}$. Our first\nobjective in this paper is to find a sufficient condition for $f$ to be\nmeromorphically convex of order $\\alpha$, $0\\le \\alpha<1$, in terms of the fact\nthat the absolute value of the well-known Schwarzian derivative $S_f (z)$ of\n$f$ is bounded above by a smallest positive root of a non-linear equation.\nSecondly, we consider a family of functions $g$ of the form\n$g(z)=z+a_2z^2+a_3z^3+\\cdots$ analytic and locally univalent in the open unit\ndisk $\\mathbb{D}:=\\{z\\in\\mathbb{C}:\\,|z|<1\\}$, and show that $g$ is belonging\nto a family of functions convex in one direction if $|S_g(z)|$ is bounded above\nby a small positive constant depending on the second coefficient $a_2$. In\nparticular, we show that such functions $g$ are also contained in the starlike\nand close-to-convex family.\n",
"title": "Meromorphic functions with small Schwarzian derivative"
}
| null | null | null | null | true | null |
5690
| null |
Default
| null | null |
null |
{
"abstract": " In 2002 Freiberg and Zähle introduced and developed a harmonic calculus for\nmeasure-geometric Laplacians associated to continuous distributions. We show\ntheir theory can be extended to encompass distributions with finite support and\ngive a matrix representation for the resulting operators. In the case of a\nuniform discrete distribution we make use of this matrix representation to\nexplicitly determine the eigenvalues and the eigenfunctions of the associated\nLaplacian.\n",
"title": "Measure-geometric Laplacians for discrete distributions"
}
| null | null | null | null | true | null |
5691
| null |
Default
| null | null |
null |
{
"abstract": " A heat exchanger can be modeled as a closed domain containing an\nincompressible fluid. The moving fluid has a temperature distribution obeying\nthe advection-diffusion equation, with zero temperature boundary conditions at\nthe walls. Starting from a positive initial temperature distribution in the\ninterior, the goal is to flux the heat through the walls as efficiently as\npossible. Here we consider a distinct but closely related problem, that of the\nintegrated mean exit time of Brownian particles starting inside the domain.\nSince flows favorable to rapid heat exchange should lower exit times, we\nminimize a norm of the exit time. This is a time-independent optimization\nproblem that we solve analytically in some limits, and numerically otherwise.\nWe find an (at least locally) optimal velocity field that cools the domain on a\nmechanical time scale, in the sense that the integrated mean exit time is\nindependent on molecular diffusivity in the limit of large-energy flows.\n",
"title": "Optimal heat transfer and optimal exit times"
}
| null | null | null | null | true | null |
5692
| null |
Default
| null | null |
null |
{
"abstract": " A number of high-level languages and libraries have been proposed that offer\nnovel and simple to use abstractions for concurrent, asynchronous, and\ndistributed programming. The execution models that realise them, however, often\nchange over time---whether to improve performance, or to extend them to new\nlanguage features---potentially affecting behavioural and safety properties of\nexisting programs. This is exemplified by SCOOP, a message-passing approach to\nconcurrent object-oriented programming that has seen multiple changes proposed\nand implemented, with demonstrable consequences for an idiomatic usage of its\ncore abstraction. We propose a semantics comparison workbench for SCOOP with\nfully and semi-automatic tools for analysing and comparing the state spaces of\nprograms with respect to different execution models or semantics. We\ndemonstrate its use in checking the consistency of properties across semantics\nby applying it to a set of representative programs, and highlighting a\ndeadlock-related discrepancy between the principal execution models of SCOOP.\nFurthermore, we demonstrate the extensibility of the workbench by generalising\nthe formalisation of an execution model to support recently proposed extensions\nfor distributed programming. Our workbench is based on a modular and\nparameterisable graph transformation semantics implemented in the GROOVE tool.\nWe discuss how graph transformations are leveraged to atomically model\nintricate language abstractions, how the visual yet algebraic nature of the\nmodel can be used to ascertain soundness, and highlight how the approach could\nbe applied to similar languages.\n",
"title": "A Semantics Comparison Workbench for a Concurrent, Asynchronous, Distributed Programming Language"
}
| null | null | null | null | true | null |
5693
| null |
Default
| null | null |
null |
{
"abstract": " Linear Logic was introduced by Girard as a resource-sensitive refinement of\nclassical logic. It turned out that full propositional Linear Logic is\nundecidable (Lincoln, Mitchell, Scedrov, and Shankar) and, hence, it is more\nexpressive than (modalized) classical or intuitionistic logic. In this paper we\nfocus on the study of the simplest fragments of Linear Logic, such as the\none-literal and constant-only fragments (the latter contains no literals at\nall). Here we demonstrate that all these extremely simple fragments of Linear\nLogic (one-literal, $\\bot$-only, and even unit-only) are exactly of the same\nexpressive power as the corresponding full versions. We present also a complete\ncomputational interpretation (in terms of acyclic programs with stack) for\nbottom-free Intuitionistic Linear Logic. Based on this interpretation, we prove\nthe fairness of our encodings and establish the foregoing complexity results.\n",
"title": "Simulating Linear Logic in 1-Only Linear Logic"
}
| null | null | null | null | true | null |
5694
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we propose two novel physical layer aware adaptive network\ncoding and coded modulation schemes for time variant channels. The proposed\nschemes have been applied to different satellite communications scenarios with\ndifferent Round Trip Times (RTT). Compared to adaptive network coding, and\nclassical non-adaptive network coding schemes for time variant channels, as\nbenchmarks, the proposed schemes demonstrate that adaptation of packet\ntransmission based on the channel variation and corresponding erasures allows\nfor significant gains in terms of throughput, delay and energy efficiency. We\nshed light on the trade-off between energy efficiency and delay-throughput\ngains, demonstrating that conservative adaptive approaches that favors less\ntransmission under high erasures, might cause higher delay and less throughput\ngains in comparison to non-conservative approaches that favor more transmission\nto account for high erasures.\n",
"title": "Adaptive Network Coding Schemes for Satellite Communications"
}
| null | null | null | null | true | null |
5695
| null |
Default
| null | null |
null |
{
"abstract": " The entropy principle in the formulation of Müller and Liu is a common\ntool used in constitutive modelling for the development of restrictions on the\nunknown constitutive functions describing material properties of various\nphysical continua. In the current work, a symbolic software implementation of\nthe Liu algorithm, based on \\verb|Maple| software and the \\verb|GeM| package,\nis presented. The computational framework is used to algorithmically perform\ntechnically demanding symbolic computations related to the entropy principle,\nto simplify and reduce Liu's identities, and ultimately to derive explicit\nformulas describing classes of constitutive functions that do not violate the\nentropy principle. Detailed physical examples are presented and discussed.\n",
"title": "A Symbolic Computation Framework for Constitutive Modelling Based On Entropy Principles"
}
| null | null | null | null | true | null |
5696
| null |
Default
| null | null |
null |
{
"abstract": " Bots, social media accounts controlled by software rather than by humans,\nhave recently been under the spotlight for their association with various forms\nof online manipulation. To date, much work has focused on social bot detection,\nbut little attention has been devoted to the characterization and measurement\nof the behavior and activity of bots, as opposed to humans'. Over the course of\nthe years, bots have become more sophisticated, and capable to reflect some\nshort-term behavior, emulating that of human users. The goal of this paper is\nto study the behavioral dynamics that bots exhibit over the course of one\nactivity session, and highlight if and how these differ from human activity\nsignatures. By using a large Twitter dataset associated with recent political\nevents, we first separate bots and humans, then isolate their activity\nsessions. We compile a list of quantities to be measured, like the propensity\nof users to engage in social interactions or to produce content. Our analysis\nhighlights the presence of short-term behavioral trends in humans, which can be\nassociated with a cognitive origin, that are absent in bots, intuitively due to\ntheir automated activity. These findings are finally codified to create and\nevaluate a machine learning algorithm to detect activity sessions produced by\nbots and humans, to allow for more nuanced bot detection strategies.\n",
"title": "Measuring bot and human behavioral dynamics"
}
| null | null | null | null | true | null |
5697
| null |
Default
| null | null |
null |
{
"abstract": " Optimal control problems without control costs in general do not possess\nsolutions due to the lack of coercivity. However, unilateral constraints\ntogether with the assumption of existence of strictly positive solutions of a\npre-adjoint state equation, are sufficient to obtain existence of optimal\nsolutions in the space of Radon measures. Optimality conditions for these\ngeneralized minimizers can be obtained using Fenchel duality, which requires a\nnon-standard perturbation approach if the control-to-observation mapping is not\ncontinuous (e.g., for Neumann boundary control in three dimensions). Combining\na conforming discretization of the measure space with a semismooth Newton\nmethod allows the numerical solution of the optimal control problem.\n",
"title": "Optimal control of elliptic equations with positive measures"
}
| null | null | null | null | true | null |
5698
| null |
Default
| null | null |
null |
{
"abstract": " Text representations using neural word embeddings have proven effective in\nmany NLP applications. Recent researches adapt the traditional word embedding\nmodels to learn vectors of multiword expressions (concepts/entities). However,\nthese methods are limited to textual knowledge bases (e.g., Wikipedia). In this\npaper, we propose a novel and simple technique for integrating the knowledge\nabout concepts from two large scale knowledge bases of different structure\n(Wikipedia and Probase) in order to learn concept representations. We adapt the\nefficient skip-gram model to seamlessly learn from the knowledge in Wikipedia\ntext and Probase concept graph. We evaluate our concept embedding models on two\ntasks: (1) analogical reasoning, where we achieve a state-of-the-art\nperformance of 91% on semantic analogies, (2) concept categorization, where we\nachieve a state-of-the-art performance on two benchmark datasets achieving\ncategorization accuracy of 100% on one and 98% on the other. Additionally, we\npresent a case study to evaluate our model on unsupervised argument type\nidentification for neural semantic parsing. We demonstrate the competitive\naccuracy of our unsupervised method and its ability to better generalize to out\nof vocabulary entity mentions compared to the tedious and error prone methods\nwhich depend on gazetteers and regular expressions.\n",
"title": "Beyond Word Embeddings: Learning Entity and Concept Representations from Large Scale Knowledge Bases"
}
| null | null | null | null | true | null |
5699
| null |
Default
| null | null |
null |
{
"abstract": " A two-dimensional (2D) mathematical model of quadratically distorted (QD)\ngrating is established with the principles of Fraunhofer diffraction and\nFourier optics. Discrete sampling and bisection algorithm are applied for\nfinding numerical solution of the diffraction pattern of QD grating. This 2D\nmathematical model allows the precise design of QD grating and improves the\noptical performance of simultaneous multiplane imaging system.\n",
"title": "Fraunhofer diffraction at the two-dimensional quadratically distorted (QD) Grating"
}
| null | null | null | null | true | null |
5700
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.