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": " The influence of the Dzyaloshinskii-Moriya interaction on the spin-wave\ndispersion in an Fe double layer grown on W(110) is measured for the first\ntime. It is demonstrated that the Dzyaloshinskii-Moriya interaction breaks the\ndegeneracy of spin waves and leads to an asymmetric spin-wave dispersion\nrelation. An extended Heisenberg spin Hamiltonian is employed to obtain the\nlongitudinal component of the Dzyaloshinskii-Moriya vectors from the\nexperimentally measured energy asymmetry.\n",
"title": "Asymmetric Spin-wave Dispersion on Fe(110): Direct Evidence of Dzyaloshinskii--Moriya Interaction"
} | null | null | null | null | true | null | 20001 | null | Default | null | null |
null | {
"abstract": " I analyze Osaka factory worker households in the early 1920s, whether\nidiosyncratic income shocks were shared efficiently, and which consumption\ncategories were robust to shocks. While the null hypothesis of full\nrisk-sharing of total expenditures was rejected, factory workers maintained\ntheir households, in that they paid for essential expenditures (rent,\nutilities, and commutation) during economic hardship. Additionally, children's\neducation expenditures were possibly robust to idiosyncratic income shocks. The\nresults suggest that temporary income is statistically significantly increased\nif disposable income drops due to idiosyncratic shocks. Historical documents\nsuggest microfinancial lending and saving institutions helped mitigate\nrisk-based vulnerabilities.\n",
"title": "Consumption smoothing in the working-class households of interwar Japan"
} | null | null | null | null | true | null | 20002 | null | Default | null | null |
null | {
"abstract": " On the 29 March 2014 NOAA active region (AR) 12017 produced an X1 flare which\nwas simultaneously observed by an unprecedented number of observatories. We\nhave investigated the pre-flare period of this flare from 14:00 UT until 19:00\nUT using joint observations made by the Interface Region Imaging Spectrometer\n(IRIS) and the Hinode Extreme Ultraviolet Imaging Spectrometer (EIS). Spectral\nlines providing coverage of the solar atmosphere from chromosphere to the\ncorona were analysed to investigate pre-flare activity within the AR. The\nresults of the investigation have revealed evidence of strongly blue-shifted\nplasma flows, with velocities up to 200 km/s, being observed 40 minutes prior\nto flaring. These flows are located along the filament present in the active\nregion and are both spatially discrete and transient. In order to constrain the\npossible explanations for this activity, we undertake non-potential magnetic\nfield modelling of the active region. This modelling indicates the existence of\na weakly twisted flux rope along the polarity inversion line in the region\nwhere a filament and the strong pre-flare flows are observed. We then discuss\nhow these observations relate to the current models of flare triggering. We\nconclude that the most likely drivers of the observed activity are internal\nreconnection in the flux rope, early onset of the flare reconnection, or tether\ncutting reconnection along the filament.\n",
"title": "Observations and Modelling of the Pre-Flare Period of the 29 March 2014 X1 Flare"
} | null | null | null | null | true | null | 20003 | null | Default | null | null |
null | {
"abstract": " We prove that every connected affine scheme of positive characteristic is a\nK(pi, 1) space for the etale topology. The main ingredient is the special case\nof the affine space over a field k. This is dealt with by induction on n, using\na key \"Bertini-type\"' statement regarding the wild ramification of l-adic local\nsystems on affine spaces, which might be of independent interest. Its proof\nuses in an essential way recent advances in higher ramification theory due to\nT. Saito. We also give rigid analytic and mixed characteristic versions of the\nmain result.\n",
"title": "Wild ramification and K(pi, 1) spaces"
} | null | null | null | null | true | null | 20004 | null | Default | null | null |
null | {
"abstract": " Deep neural network models have been proven to be very successful in image\nclassification tasks, also for medical diagnosis, but their main concern is its\nlack of interpretability. They use to work as intuition machines with high\nstatistical confidence but unable to give interpretable explanations about the\nreported results. The vast amount of parameters of these models make difficult\nto infer a rationale interpretation from them. In this paper we present a\ndiabetic retinopathy interpretable classifier able to classify retine images\ninto the different levels of disease severity and of explaining its results by\nassigning a score for every point in the hidden and input space, evaluating its\ncontribution to the final classification in a linear way. The generated visual\nmaps can be interpreted by an expert in order to compare its own knowledge with\nthe interpretation given by the model.\n",
"title": "A Deep Learning Interpretable Classifier for Diabetic Retinopathy Disease Grading"
} | null | null | null | null | true | null | 20005 | null | Default | null | null |
null | {
"abstract": " Arabic word segmentation is essential for a variety of NLP applications such\nas machine translation and information retrieval. Segmentation entails breaking\nwords into their constituent stems, affixes and clitics. In this paper, we\ncompare two approaches for segmenting four major Arabic dialects using only\nseveral thousand training examples for each dialect. The two approaches involve\nposing the problem as a ranking problem, where an SVM ranker picks the best\nsegmentation, and as a sequence labeling problem, where a bi-LSTM RNN coupled\nwith CRF determines where best to segment words. We are able to achieve solid\nsegmentation results for all dialects using rather limited training data. We\nalso show that employing Modern Standard Arabic data for domain adaptation and\nassuming context independence improve overall results.\n",
"title": "Arabic Multi-Dialect Segmentation: bi-LSTM-CRF vs. SVM"
} | null | null | [
"Computer Science"
]
| null | true | null | 20006 | null | Validated | null | null |
null | {
"abstract": " This paper studies an electricity market consisting of an independent system\noperator (ISO) and a group of generators. The goal is to solve the DC optimal\npower flow (DC-OPF) problem: have the generators collectively meet the power\ndemand while minimizing the aggregate generation cost and respecting line flow\nlimits in the network. The ISO by itself cannot solve the DC-OPF problem as\ngenerators are strategic and do not share their cost functions. Instead, each\ngenerator submits to the ISO a bid, consisting of the price per unit of\nelectricity at which it is willing to provide power. Based on the bids, the ISO\ndecides how much production to allocate to each generator to minimize the total\npayment while meeting the load and satisfying the line limits. We provide a\nprovably correct, decentralized iterative scheme, termed BID ADJUSTMENT\nALGORITHM, for the resulting Bertrand competition game. Regarding convergence,\nwe show that the algorithm takes the generators' bids to any desired\nneighborhood of the efficient Nash equilibrium at a linear convergence rate. As\na consequence, the optimal production of the generators converges to the\noptimizer of the DC-OPF problem. Regarding robustness, we show that the\nalgorithm is robust to affine perturbations in the bid adjustment scheme and\nthat there is no incentive for any individual generator to deviate from the\nalgorithm by using an alternative bid update scheme. We also establish the\nalgorithm robustness to collusion, i.e., we show that, as long as each bus with\ngeneration has a generator following the strategy, there is no incentive for\nany group of generators to share information with the intent of tricking the\nsystem to obtain a higher payoff. Simulations illustrate our results.\n",
"title": "Iterative bidding in electricity markets: rationality and robustness"
} | null | null | null | null | true | null | 20007 | null | Default | null | null |
null | {
"abstract": " To investigate the existence of sterile neutrino, we propose a new neutrino\nproduction method using $^{13}$C beams and a $^{9}$Be target for short-baseline\nelectron antineutrino (${\\bar{\\nu}}_{e}$) disappearance study. The production\nof secondary unstable isotopes which can emit neutrinos from the $^{13}$C +\n$^{9}$Be reaction is calculated with three different nucleus-nucleus (AA)\nreaction models. Different isotope yields are obtained using these models, but\nthe results of the neutrino flux are found to have unanimous similarities. This\nfeature gives an opportunity to study neutrino oscillation through shape\nanalysis. In this work, expected neutrino flux and event rates are discussed in\ndetail through intensive simulation of the light ion collision reaction and the\nneutrino flux from the beta decay of unstable isotopes followed by this\ncollision. Together with the reactor and accelerator anomalies, the present\nproposed ${\\bar{\\nu}}_{e}$ source is shown to be a practically alternative test\nof the existence of the $\\Delta m^{2}$ $\\sim$ 1 eV$^{2}$ scale sterile\nneutrino.\n",
"title": "Short-baseline electron antineutrino disappearance study by using neutrino sources from $^{13}$C + $^{9}$Be reaction"
} | null | null | null | null | true | null | 20008 | null | Default | null | null |
null | {
"abstract": " Protein crystal production is a major bottleneck for the structural\ncharacterisation of proteins. To advance beyond large-scale screening, rational\nstrategies for protein crystallization are crucial. Understanding how chemical\nanisotropy (or patchiness) of the protein surface due to the variety of amino\nacid side chains in contact with solvent, contributes to protein protein\ncontact formation in the crystal lattice is a major obstacle to predicting and\noptimising crystallization. The relative scarcity of sophisticated theoretical\nmodels that include sufficient detail to link collective behaviour, captured in\nprotein phase diagrams, and molecular level details, determined from\nhigh-resolution structural information is a further barrier. Here we present\ntwo crystals structures for the P23TR36S mutant of gamma D-crystallin, each\nwith opposite solubility behaviour, one melts when heated, the other when\ncooled. When combined with the protein phase diagram and a tailored patchy\nparticle model we show that a single temperature dependent interaction is\nsufficient to stabilise the inverted solubility crystal. This contact, at the\nP23T substitution site, relates to a genetic cataract and reveals at a\nmolecular level, the origin of the lowered and retrograde solubility of the\nprotein. Our results show that the approach employed here may present an\nalternative strategy for the rationalization of protein crystallization.\n",
"title": "Temperature-dependent non-covalent protein-protein interactions explain normal and inverted solubility in a mutant of human gamma D-crystallin"
} | null | null | null | null | true | null | 20009 | null | Default | null | null |
null | {
"abstract": " We present the Kepler Object of Interest (KOI) catalog of transiting\nexoplanets based on searching four years of Kepler time series photometry (Data\nRelease 25, Q1-Q17). The catalog contains 8054 KOIs of which 4034 are planet\ncandidates with periods between 0.25 and 632 days. Of these candidates, 219 are\nnew and include two in multi-planet systems (KOI-82.06 and KOI-2926.05), and\nten high-reliability, terrestrial-size, habitable zone candidates. This catalog\nwas created using a tool called the Robovetter which automatically vets the\nDR25 Threshold Crossing Events (TCEs, Twicken et al. 2016). The Robovetter also\nvetted simulated data sets and measured how well it was able to separate TCEs\ncaused by noise from those caused by low signal-to-noise transits. We discusses\nthe Robovetter and the metrics it uses to sort TCEs. For orbital periods less\nthan 100 days the Robovetter completeness (the fraction of simulated transits\nthat are determined to be planet candidates) across all observed stars is\ngreater than 85%. For the same period range, the catalog reliability (the\nfraction of candidates that are not due to instrumental or stellar noise) is\ngreater than 98%. However, for low signal-to-noise candidates between 200 and\n500 days around FGK dwarf stars, the Robovetter is 76.7% complete and the\ncatalog is 50.5% reliable. The KOI catalog, the transit fits and all of the\nsimulated data used to characterize this catalog are available at the NASA\nExoplanet Archive.\n",
"title": "Planetary Candidates Observed by Kepler. VIII. A Fully Automated Catalog With Measured Completeness and Reliability Based on Data Release 25"
} | null | null | null | null | true | null | 20010 | null | Default | null | null |
null | {
"abstract": " The dark sector may contain a dark photon that kinetically mixes with the\nStandard Model photon, allowing dark matter to interact weakly with normal\nmatter. In previous work we analyzed the implications of this scenario for dark\nmatter capture by the Sun. Dark matter will gather in the core of the Sun and\nannihilate to dark photons. These dark photons travel outwards from the center\nof the Sun and may decay to produce positrons that can be detected by the Alpha\nMagnetic Spectrometer (AMS-02) on the ISS. We found that the dark photon\nparameter space accessible to this analysis is largely constrained by strong\nlimits on the spin-independent WIMP-nucleon cross section from direct detection\nexperiments. In this paper we build upon previous work by considering the case\nwhere the dark sector contains two species of Dirac fermion that are nearly\ndegenerate in mass and couple inelastically to the dark photon. We find that\nfor small values of the mass splitting $\\Delta \\sim 100 ~\\text{keV}$, the\npredicted positron signal at AMS-02 remains largely unchanged from the\npreviously considered elastic case while constraints from direct detection are\nrelaxed, leaving a region of parameter space with dark matter mass $100\n~\\text{GeV} \\lesssim m_X \\lesssim 10 ~\\text{TeV}$, dark photon mass $1\n~\\text{MeV} \\lesssim m_{A'} \\lesssim 100 ~\\text{MeV}$, and kinetic mixing\nparameter $10^{-9} \\lesssim \\varepsilon \\lesssim 10^{-8}$ that is untouched by\nsupernova observations and fixed target experiments but where an inelastic dark\nsector may still be discovered using existing AMS-02 data.\n",
"title": "Dark Photons from Captured Inelastic Dark Matter Annihilation: Charged Particle Signatures"
} | null | null | [
"Physics"
]
| null | true | null | 20011 | null | Validated | null | null |
null | {
"abstract": " A network epidemics model based on the classical Polya urn scheme is\ninvestigated. Temporal contagion processes are generated on the network nodes\nusing a modified Polya sampling scheme that accounts for spatial infection\namong neighbouring nodes. The stochastic properties and the asymptotic\nbehaviour of the resulting network contagion process are analyzed. Unlike the\nclassical Polya process, the network process is noted to be non-stationary in\ngeneral, although it is shown to be time-invariant in its first and some of its\nsecond-order statistics and to satisfy martingale convergence properties under\ncertain conditions. Three classical Polya processes, one computational and two\nanalytical, are proposed to statistically approximate the contagion process of\neach node, showing a good fit for a range of system parameters. Finally,\nempirical results compare and contrast our model with the well-known discrete\ntime SIS model.\n",
"title": "A Polya Contagion Model for Networks"
} | null | null | null | null | true | null | 20012 | null | Default | null | null |
null | {
"abstract": " We describe a general framework --compressive statistical learning-- for\nresource-efficient large-scale learning: the training collection is compressed\nin one pass into a low-dimensional sketch (a vector of random empirical\ngeneralized moments) that captures the information relevant to the considered\nlearning task. A near-minimizer of the risk is computed from the sketch through\nthe solution of a nonlinear least squares problem. We investigate sufficient\nsketch sizes to control the generalization error of this procedure. The\nframework is illustrated on compressive clustering, compressive Gaussian\nmixture Modeling with fixed known variance, and compressive PCA.\n",
"title": "Compressive Statistical Learning with Random Feature Moments"
} | null | null | null | null | true | null | 20013 | null | Default | null | null |
null | {
"abstract": " Bandit structured prediction describes a stochastic optimization framework\nwhere learning is performed from partial feedback. This feedback is received in\nthe form of a task loss evaluation to a predicted output structure, without\nhaving access to gold standard structures. We advance this framework by lifting\nlinear bandit learning to neural sequence-to-sequence learning problems using\nattention-based recurrent neural networks. Furthermore, we show how to\nincorporate control variates into our learning algorithms for variance\nreduction and improved generalization. We present an evaluation on a neural\nmachine translation task that shows improvements of up to 5.89 BLEU points for\ndomain adaptation from simulated bandit feedback.\n",
"title": "Bandit Structured Prediction for Neural Sequence-to-Sequence Learning"
} | null | null | null | null | true | null | 20014 | null | Default | null | null |
null | {
"abstract": " The ability to learn from a small number of examples has been a difficult\nproblem in machine learning since its inception. While methods have succeeded\nwith large amounts of training data, research has been underway in how to\naccomplish similar performance with fewer examples, known as one-shot or more\ngenerally few-shot learning. This technique has been shown to have promising\nperformance, but in practice requires fixed-size inputs making it impractical\nfor production systems where class sizes can vary. This impedes training and\nthe final utility of few-shot learning systems. This paper describes an\napproach to constructing and training a network that can handle arbitrary\nexample sizes dynamically as the system is used.\n",
"title": "Dynamic Input Structure and Network Assembly for Few-Shot Learning"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 20015 | null | Validated | null | null |
null | {
"abstract": " Faster R-CNN is one of the most representative and successful methods for\nobject detection, and has been becoming increasingly popular in various\nobjection detection applications. In this report, we propose a robust deep face\ndetection approach based on Faster R-CNN. In our approach, we exploit several\nnew techniques including new multi-task loss function design, online hard\nexample mining, and multi-scale training strategy to improve Faster R-CNN in\nmultiple aspects. The proposed approach is well suited for face detection, so\nwe call it Face R-CNN. Extensive experiments are conducted on two most popular\nand challenging face detection benchmarks, FDDB and WIDER FACE, to demonstrate\nthe superiority of the proposed approach over state-of-the-arts.\n",
"title": "Face R-CNN"
} | null | null | null | null | true | null | 20016 | null | Default | null | null |
null | {
"abstract": " Data-target pairing is an important step towards multi-target localization\nfor the intelligent operation of unmanned systems. Target localization plays a\ncrucial role in numerous applications, such as search, and rescue missions,\ntraffic management and surveillance. The objective of this paper is to present\nan innovative target location learning approach, where numerous machine\nlearning approaches, including K-means clustering and supported vector machines\n(SVM), are used to learn the data pattern across a list of spatially\ndistributed sensors. To enable the accurate data association from different\nsensors for accurate target localization, appropriate data pre-processing is\nessential, which is then followed by the application of different machine\nlearning algorithms to appropriately group data from different sensors for the\naccurate localization of multiple targets. Through simulation examples, the\nperformance of these machine learning algorithms is quantified and compared.\n",
"title": "Multi-Sensor Data Pattern Recognition for Multi-Target Localization: A Machine Learning Approach"
} | null | null | null | null | true | null | 20017 | null | Default | null | null |
null | {
"abstract": " We present Calipso, an interactive method for editing images and videos in a\nphysically-coherent manner. Our main idea is to realize physics-based\nmanipulations by running a full physics simulation on proxy geometries given by\nnon-rigidly aligned CAD models. Running these simulations allows us to apply\nnew, unseen forces to move or deform selected objects, change physical\nparameters such as mass or elasticity, or even add entire new objects that\ninteract with the rest of the underlying scene. In Calipso, the user makes\nedits directly in 3D; these edits are processed by the simulation and then\ntransfered to the target 2D content using shape-to-image correspondences in a\nphoto-realistic rendering process. To align the CAD models, we introduce an\nefficient CAD-to-image alignment procedure that jointly minimizes for rigid and\nnon-rigid alignment while preserving the high-level structure of the input\nshape. Moreover, the user can choose to exploit image flow to estimate scene\nmotion, producing coherent physical behavior with ambient dynamics. We\ndemonstrate Calipso's physics-based editing on a wide range of examples\nproducing myriad physical behavior while preserving geometric and visual\nconsistency.\n",
"title": "Calipso: Physics-based Image and Video Editing through CAD Model Proxies"
} | null | null | [
"Computer Science"
]
| null | true | null | 20018 | null | Validated | null | null |
null | {
"abstract": " The resolutions and maximal sets of compatible resolutions of all 2-(120,8,1)\ndesigns arising frommaximal (120,8)-arcs in the known projective planes of\norder 16 are computed. It is shown that each of these designs is embeddable in\na unique way in a projective plane of order 16.\n",
"title": "Maximal (120,8)-arcs in projective planes of order 16 and related designs"
} | null | null | null | null | true | null | 20019 | null | Default | null | null |
null | {
"abstract": " Spectrum of a first order sentence is the set of all $\\alpha$ such that $G(n,\nn^{-\\alpha})$ does not obey zero-one law w.r.t. this sentence. We have proved\nthat the minimal number of quantifier alternations of a first order sentence\nwith an infinite spectrum equals 3. We have also proved that the spectrum of a\nfirst order sentence with a quantifier depth 4 has no limit points except\npossibly the points 1/2 and 3/5.\n",
"title": "First order sentences about random graphs: small number of alternations"
} | null | null | null | null | true | null | 20020 | null | Default | null | null |
null | {
"abstract": " Within the past few decades we have witnessed digital revolution, which moved\nscholarly communication to electronic media and also resulted in a substantial\nincrease in its volume. Nowadays keeping track with the latest scientific\nachievements poses a major challenge for the researchers. Scientific\ninformation overload is a severe problem that slows down scholarly\ncommunication and knowledge propagation across the academia. Modern research\ninfrastructures facilitate studying scientific literature by providing\nintelligent search tools, proposing similar and related documents, visualizing\ncitation and author networks, assessing the quality and impact of the articles,\nand so on. In order to provide such high quality services the system requires\nthe access not only to the text content of stored documents, but also to their\nmachine-readable metadata. Since in practice good quality metadata is not\nalways available, there is a strong demand for a reliable automatic method of\nextracting machine-readable metadata directly from source documents. This\nresearch addresses these problems by proposing an automatic, accurate and\nflexible algorithm for extracting wide range of metadata directly from\nscientific articles in born-digital form. Extracted information includes basic\ndocument metadata, structured full text and bibliography section. Designed as a\nuniversal solution, proposed algorithm is able to handle a vast variety of\npublication layouts with high precision and thus is well-suited for analyzing\nheterogeneous document collections. This was achieved by employing supervised\nand unsupervised machine-learning algorithms trained on large, diverse\ndatasets. The evaluation we conducted showed good performance of proposed\nmetadata extraction algorithm. The comparison with other similar solutions also\nproved our algorithm performs better than competition for most metadata types.\n",
"title": "New Methods for Metadata Extraction from Scientific Literature"
} | null | null | null | null | true | null | 20021 | null | Default | null | null |
null | {
"abstract": " We introduce a refinement of the classical Liouville function to primes in\narithmetic progressions. Using this, we discover new biases in the appearances\nof primes in a given arithmetic progression in the prime factorizations of\nintegers. For example, we observe that the primes of the form $4k+1$ tend to\nappear an even number of times in the prime factorization of a given integer,\nmore so than for primes of the form $4k+3$. We are led to consider variants of\nPólya's conjecture, supported by extensive numerical evidence, and its\nrelation to other conjectures.\n",
"title": "Unexpected biases in prime factorizations and Liouville functions for arithmetic progressions"
} | null | null | null | null | true | null | 20022 | null | Default | null | null |
null | {
"abstract": " In the presence of a certain class of functions we show that there exists a\nsmooth solution to Navier-Stokes equation. This solution entertains the\nproperty of being nonconvective. We introduce a definition for any possible\nsolution to the problem with minimum assumptions on the existence and the\nregularity of such solution. Then we prove that the proposed class of functions\nrepresents the unique solution to the problem and consequently we conclude that\nthere exists no convective solutions to the problem in the sense of the given\ndefinition.\n",
"title": "Nonconvection and uniqueness in Navier-Stokes equation"
} | null | null | null | null | true | null | 20023 | null | Default | null | null |
null | {
"abstract": " Conditional specification of distributions is a developing area with\nincreasing applications. In the finite discrete case, a variety of compatible\nconditions can be derived. In this paper, we propose an alternative approach to\nstudy the compatibility of two conditional probability distributions under the\nfinite discrete setup. A technique based on rank-based criterion is shown to be\nparticularly convenient for identifying compatible distributions corresponding\nto complete conditional specification including the case with zeros.The\nproposed methods are illustrated with several examples.\n",
"title": "An alternative approach for compatibility of two discrete conditional distributions"
} | null | null | null | null | true | null | 20024 | null | Default | null | null |
null | {
"abstract": " Let $\\mathcal{B}$ denote a set of bicolorings of $[n]$, where each bicoloring\nis a mapping of the points in $[n]$ to $\\{-1,+1\\}$.\nFor each $B \\in \\mathcal{B}$, let $Y_B=(B(1),\\ldots,B(n))$.\nFor each $A \\subseteq [n]$, let $X_A \\in \\{0,1\\}^n$ denote the incidence\nvector of $A$.\nA non-empty set $A$ is said to be an `unbiased representative' for a\nbicoloring $B \\in \\mathcal{B}$ if $\\left\\langle X_A,Y_B\\right\\rangle =0$.\nGiven a set $\\mathcal{B}$ of bicolorings, we study the minimum cardinality of\na family $\\mathcal{A}$ consisting of subsets of $[n]$ such that every\nbicoloring in $\\mathcal{B}$ has an unbiased representative in $\\mathcal{A}$.\n",
"title": "System of unbiased representatives for a collection of bicolorings"
} | null | null | null | null | true | null | 20025 | null | Default | null | null |
null | {
"abstract": " In part I we considered the problem of convergence to a saddle point of a\nconcave-convex function via gradient dynamics and an exact characterization was\ngiven to their asymptotic behaviour. In part II we consider a general class of\nsubgradient dynamics that provide a restriction in an arbitrary convex domain.\nWe show that despite the nonlinear and non-smooth character of these dynamics\ntheir $\\omega$-limit set is comprised of solutions to only linear ODEs. In\nparticular, we show that the latter are solutions to subgradient dynamics on\naffine subspaces which is a smooth class of dynamics the asymptotic properties\nof which have been exactly characterized in part I. Various convergence\ncriteria are formulated using these results and several examples and\napplications are also discussed throughout the manuscript.\n",
"title": "Stability and instability in saddle point dynamics Part II: The subgradient method"
} | null | null | null | null | true | null | 20026 | null | Default | null | null |
null | {
"abstract": " JavaScript systems are becoming increasingly complex and large. To tackle the\nchallenges involved in implementing these systems, the language is evolving to\ninclude several constructions for programming- in-the-large. For example,\nalthough the language is prototype-based, the latest JavaScript standard, named\nECMAScript 6 (ES6), provides native support for implementing classes. Even\nthough most modern web browsers support ES6, only a very few applications use\nthe class syntax. In this paper, we analyze the process of migrating structures\nthat emulate classes in legacy JavaScript code to adopt the new syntax for\nclasses introduced by ES6. We apply a set of migration rules on eight legacy\nJavaScript systems. In our study, we document: (a) cases that are\nstraightforward to migrate (the good parts); (b) cases that require manual and\nad-hoc migration (the bad parts); and (c) cases that cannot be migrated due to\nlimitations and restrictions of ES6 (the ugly parts). Six out of eight systems\n(75%) contain instances of bad and/or ugly cases. We also collect the\nperceptions of JavaScript developers about migrating their code to use the new\nsyntax for classes.\n",
"title": "Refactoring Legacy JavaScript Code to Use Classes: The Good, The Bad and The Ugly"
} | null | null | null | null | true | null | 20027 | null | Default | null | null |
null | {
"abstract": " Phononic bandgaps of Parylene-C microfibrous thin films (muFTFs) were\ncomputationally determined by treating them as phononic crystals comprising\nidentical microfibers arranged either on a square or a hexagonal lattice. The\nmicrofibers could be columnar,chevronic, or helical in shape, and the host\nmedium could be either water or air. All bandgaps were observed to lie in the\n0.01-to-162.9-MHz regime, for microfibers of realistically chosen dimensions.\nThe upper limit of the frequency of bandgaps was the highest for the columnar\nmuFTF and the lowest for the chiral muFTF. More bandgaps exist when the host\nmedium is water than air. Complete bandgaps were observed for the columnar\nmuFTF with microfibers arranged on a hexagonal lattice in air, the chevronic\nmuFTF with microfibers arranged on a square lattice in water, and the chiral\nmuFTF with microfibers arranged on a hexagonal lattice in either air or water.\nThe softness of the Parylene-C muFTFs makes them mechanically tunable, and\ntheir bandgaps can be exploited in multiband ultrasonic filters.\n",
"title": "Parylene-C microfibrous thin films as phononic crystals"
} | null | null | null | null | true | null | 20028 | null | Default | null | null |
null | {
"abstract": " Though the deep learning is pushing the machine learning to a new stage,\nbasic theories of machine learning are still limited. The principle of\nlearning, the role of the a prior knowledge, the role of neuron bias, and the\nbasis for choosing neural transfer function and cost function, etc., are still\nfar from clear. In this paper, we present a general theoretical framework for\nmachine learning. We classify the prior knowledge into common and\nproblem-dependent parts, and consider that the aim of learning is to maximally\nincorporate them. The principle we suggested for maximizing the former is the\ndesign risk minimization principle, while the neural transfer function, the\ncost function, as well as pretreatment of samples, are endowed with the role\nfor maximizing the latter. The role of the neuron bias is explained from a\ndifferent angle. We develop a Monte Carlo algorithm to establish the\ninput-output responses, and we control the input-output sensitivity of a\nlearning machine by controlling that of individual neurons. Applications of\nfunction approaching and smoothing, pattern recognition and classification, are\nprovided to illustrate how to train general learning machines based on our\ntheory and algorithm. Our method may in addition induce new applications, such\nas the transductive inference.\n",
"title": "A General Theory for Training Learning Machine"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 20029 | null | Validated | null | null |
null | {
"abstract": " We present semiparametric spectral modeling of the complete larval Drosophila\nmushroom body connectome. Motivated by a thorough exploratory data analysis of\nthe network via Gaussian mixture modeling (GMM) in the adjacency spectral\nembedding (ASE) representation space, we introduce the latent structure model\n(LSM) for network modeling and inference. LSM is a generalization of the\nstochastic block model (SBM) and a special case of the random dot product graph\n(RDPG) latent position model, and is amenable to semiparametric GMM in the ASE\nrepresentation space. The resulting connectome code derived via semiparametric\nGMM composed with ASE captures latent connectome structure and elucidates\nbiologically relevant neuronal properties.\n",
"title": "Semiparametric spectral modeling of the Drosophila connectome"
} | null | null | null | null | true | null | 20030 | null | Default | null | null |
null | {
"abstract": " We present a term rewrite system that formally models the Message\nAuthenticator Algorithm (MAA), which was one of the first cryptographic\nfunctions for computing a Message Authentication Code and was adopted, between\n1987 and 2001, in international standards (ISO 8730 and ISO 8731-2) to ensure\nthe authenticity and integrity of banking transactions. Our term rewrite system\nis large (13 sorts, 18 constructors, 644 non-constructors, and 684 rewrite\nrules), confluent, and terminating. Implementations in thirteen different\nlanguages have been automatically derived from this model and used to validate\n200 official test vectors for the MAA.\n",
"title": "A Large Term Rewrite System Modelling a Pioneering Cryptographic Algorithm"
} | null | null | null | null | true | null | 20031 | null | Default | null | null |
null | {
"abstract": " Most complex networks are not static, but evolve along time. Given a specific\nconfiguration of one such changing network, it becomes a particularly\ninteresting issue to quantify the diversity of possible unfoldings of its\ntopology. In this work, we suggest the concept of malleability of a network,\nwhich is defined as the exponential of the entropy of the probabilities of each\npossible unfolding with respect to a given configuration. We calculate the\nmalleability with respect to specific measurements of the involved topologies.\nMore specifically, we identify the possible topologies derivable from a given\nconfiguration and calculate some topological measurement of them (e.g.\nclustering coefficient, shortest path length, assortativity, etc.), leading to\nrespective probabilities being associated to each possible measurement value.\nThough this approach implies some level of degeneracy in the mapping from\ntopology to measurement space, it still paves the way to inferring the\nmalleability of specific network types with respect to given topological\nmeasurements. We report that the malleability, in general, depends on each\nspecific measurement, with the average shortest path length and degree\nassortativity typically leading to large malleability values. The maximum\nmalleability was observed for the Wikipedia network and the minimum for the\nWatts-Strogatz model.\n",
"title": "Malleability of complex networks"
} | null | null | null | null | true | null | 20032 | null | Default | null | null |
null | {
"abstract": " Potential critical risks of cascading failures in power systems can be\nidentified by exposing those critical electrical elements on which certain\ninitial disturbances may cause maximum disruption to power transmission\nnetworks. In this work, we investigate cascading failures in power systems\ndescribed by the direct current (DC) power flow equations, while initial\ndisturbances take the form of altering admittance of elements. The disruption\nis quantified with the remaining transmission power at the end of cascading\nprocess. In particular, identifying the critical elements and the corresponding\ninitial disturbances causing the worst-case cascading blackout is formulated as\na dynamic optimization problem (DOP) in the framework of optimal control\ntheory, where the entire propagation process of cascading failures is put under\nconsideration. An Identifying Critical Risk Algorithm (ICRA) based on the\nmaximum principle is proposed to solve the DOP. Simulation results on the IEEE\n9-Bus and the IEEE 14-Bus test systems are presented to demonstrate the\neffectiveness of the algorithm.\n",
"title": "Identifying Critical Risks of Cascading Failures in Power Systems"
} | null | null | [
"Computer Science"
]
| null | true | null | 20033 | null | Validated | null | null |
null | {
"abstract": " We present a comprehensive account of the proton radiation hardness of Eljen\nTechnology's EJ-500 optical cement used in the construction of experiment\ndetectors. The cement was embedded into five plastic scintillator tiles which\nwere each exposed to one of five different levels of radiation by a 50 MeV\nproton beam produced at the 88-Inch Cyclotron at Lawrence Berkeley National\nLaboratory. A cosmic ray telescope setup was used to measure signal amplitudes\nbefore and after irradiation. Another post-radiation measurement was taken four\nmonths after the experiment to investigate whether the radiation damage to the\ncement recovers after a short amount of time. We verified that the radiation\ndamage to the tiles increased with increasing dose but showed significant\nimprovement after the four months time interval.\n",
"title": "Radiation Hardness Test of Eljen EJ-500 Optical Cement"
} | null | null | null | null | true | null | 20034 | null | Default | null | null |
null | {
"abstract": " We consider the one-dimensional model of a spin glass with independent\nGaussian-distributed random interactions, that have mean zero and variance\n$1/|i-j|^{2\\sigma}$, between the spins at sites $i$ and $j$ for all $i\\neq j$.\nIt is known that, for $\\sigma>1$, there is no phase transition at any non-zero\ntemperature in this model. We prove rigorously that, for $\\sigma>3/2$, any\nNewman-Stein metastate for the ground states (i.e.\\ the frequencies with which\ndistinct ground states are observed in finite size samples in the limit of\ninfinite size, for given disorder) is trivial and unique. In other words, for\ngiven disorder and asymptotically at large sizes, the same ground state, or its\nglobal spin flip, is obtained (almost) always. The proof consists of two parts:\none is a theorem (based on one by Newman and Stein for short-range\ntwo-dimensional models), valid for all $\\sigma>1$, that establishes triviality\nunder a convergence hypothesis on something similar to the energies of domain\nwalls, and the other (based on older results for the one-dimensional model)\nestablishes that the hypothesis is true for $\\sigma>3/2$. In addition, we\nderive heuristic scaling arguments and rigorous exponent inequalities which\ntend to support the validity of the hypothesis under broader conditions. The\nconstructions of various metastates are extended to all values $\\sigma>1/2$.\nTriviality of the metastate in bond-diluted power-law models for $\\sigma>1$ is\nproved directly.\n",
"title": "Triviality of the ground-state metastate in long-range Ising spin glasses in one dimension"
} | null | null | null | null | true | null | 20035 | null | Default | null | null |
null | {
"abstract": " In this paper, we deform the thermodynamics of a BTZ black hole from rainbow\nfunctions in gravity's rainbow. The rainbow functions will be motivated from\nresults in loop quantum gravity and Noncommutative geometry. It will be\nobserved that the thermodynamics gets deformed due to these rainbow functions,\nindicating the existence of a remnant. However, the Gibbs free energy does not\nget deformed due to these rainbow functions, and so the critical behaviour from\nGibbs does not change by this deformation.This is because the deformation in\nthe entropy cancel's out the temperature deformation.\n",
"title": "Thermodynamics of BTZ Black Holes in Gravity's Rainbow"
} | null | null | null | null | true | null | 20036 | null | Default | null | null |
null | {
"abstract": " In this paper we study rotationally symmetric solutions of the Cahn-Hilliard\nequation in $\\mathbb R^3$ constructed by the authors. These solutions form a\none parameter family analog to the family of Delaunay surfaces and in fact the\nzero level sets of their blowdowns approach these surfaces. Presently we go a\nstep further and show that their stability properties are inherited from the\nstability properties of the Delaunay surfaces. Our main result states that the\nrotationally symmetric solutions are non degenerate and that they have exactly\n$6$ Jacobi fields of temperate growth coming from the natural invariances of\nthe problem (3 translations and 2 rotations) and the variation of the Delaunay\nparameter.\n",
"title": "Nondegeneracy and the Jacobi fields of rotationally symmetric solutions to the Cahn-Hillard equation"
} | null | null | null | null | true | null | 20037 | null | Default | null | null |
null | {
"abstract": " The rank of a hierarchically hyperbolic space is the maximal number of\nunbounded factors of standard product regions; this coincides with the maximal\ndimension of a quasiflat for hierarchically hyperbolic groups. Noteworthy\nexamples where the rank coincides with familiar quantities include: the\ndimension of maximal Dehn twist flats for mapping class groups, the maximal\nrank of a free abelian subgroup for right-angled Coxeter groups and\nright-angled Artin groups (in the latter this coincides with the clique number\nof the defining graph), and, for the Weil-Petersson metric the rank is half the\ncomplex dimension of Teichmuller space.\nWe prove that in a HHS, any quasiflat of dimension equal to the rank lies\nwithin finite distance of a union of standard orthants (under a very mild\ncondition satisfied by all natural examples). This resolves outstanding\nconjectures when applied to a number of different groups and spaces. The\nmapping class group case resolves a conjecture of Farb, in Teichmuller space\nthis resolves a question of Brock, and in the context of CAT(0) cubical groups\nit strengthens previous results (so as to handle, for example, the right-angled\nCoxeter case).\nAn important ingredient, is our proof that the hull of any finite set in an\nHHS is quasi-isometric to a cube complex of dimension equal to the rank.\nWe deduce a number of applications; for instance we show that any\nquasi-isometry between HHS induces a quasi-isometry between certain simpler\nHHS. This allows one, for example, to distinguish quasi-isometry classes of\nright-angled Artin/Coxeter groups.\nAnother application is that our tools, in many cases, allow one to reduce the\nproblem of quasi-isometric rigidity for a given HHG to a combinatorial problem.\nAs a template, we give a new proof of quasi-isometric rigidity of mapping class\ngroups, using simpler combinatorial arguments than in previous proofs.\n",
"title": "Quasiflats in hierarchically hyperbolic spaces"
} | null | null | null | null | true | null | 20038 | null | Default | null | null |
null | {
"abstract": " Recurrent Neural Networks are showing much promise in many sub-areas of\nnatural language processing, ranging from document classification to machine\ntranslation to automatic question answering. Despite their promise, many\nrecurrent models have to read the whole text word by word, making it slow to\nhandle long documents. For example, it is difficult to use a recurrent network\nto read a book and answer questions about it. In this paper, we present an\napproach of reading text while skipping irrelevant information if needed. The\nunderlying model is a recurrent network that learns how far to jump after\nreading a few words of the input text. We employ a standard policy gradient\nmethod to train the model to make discrete jumping decisions. In our benchmarks\non four different tasks, including number prediction, sentiment analysis, news\narticle classification and automatic Q\\&A, our proposed model, a modified LSTM\nwith jumping, is up to 6 times faster than the standard sequential LSTM, while\nmaintaining the same or even better accuracy.\n",
"title": "Learning to Skim Text"
} | null | null | [
"Computer Science"
]
| null | true | null | 20039 | null | Validated | null | null |
null | {
"abstract": " Statistical analysis (SA) is a complex process to deduce population\nproperties from analysis of data. It usually takes a well-trained analyst to\nsuccessfully perform SA, and it becomes extremely challenging to apply SA to\nbig data applications. We propose to use deep neural networks to automate the\nSA process. In particular, we propose to construct convolutional neural\nnetworks (CNNs) to perform automatic model selection and parameter estimation,\ntwo most important SA tasks. We refer to the resulting CNNs as the neural model\nselector and the neural model estimator, respectively, which can be properly\ntrained using labeled data systematically generated from candidate models.\nSimulation study shows that both the selector and estimator demonstrate\nexcellent performances. The idea and proposed framework can be further extended\nto automate the entire SA process and have the potential to revolutionize how\nSA is performed in big data analytics.\n",
"title": "Using Deep Neural Networks to Automate Large Scale Statistical Analysis for Big Data Applications"
} | null | null | null | null | true | null | 20040 | null | Default | null | null |
null | {
"abstract": " We firstly suggest new cache policy applying the duty to delete invalid cache\ndata on Non-volatile Memory (NVM). This cache policy includes generating random\ndata and overwriting the random data into invalid cache data. Proposed cache\npolicy is more economical and effective regarding perfect deletion of data. It\nis ensure that the invalid cache data in NVM is secure against malicious\nhackers.\n",
"title": "Duty to Delete on Non-Volatile Memory"
} | null | null | null | null | true | null | 20041 | null | Default | null | null |
null | {
"abstract": " Shape-constrained density estimation is an important topic in mathematical\nstatistics. We focus on densities on $\\mathbb{R}^d$ that are log-concave, and\nwe study geometric properties of the maximum likelihood estimator (MLE) for\nweighted samples. Cule, Samworth, and Stewart showed that the logarithm of the\noptimal log-concave density is piecewise linear and supported on a regular\nsubdivision of the samples. This defines a map from the space of weights to the\nset of regular subdivisions of the samples, i.e. the face poset of their\nsecondary polytope. We prove that this map is surjective. In fact, every\nregular subdivision arises in the MLE for some set of weights with positive\nprobability, but coarser subdivisions appear to be more likely to arise than\nfiner ones. To quantify these results, we introduce a continuous version of the\nsecondary polytope, whose dual we name the Samworth body. This article\nestablishes a new link between geometric combinatorics and nonparametric\nstatistics, and it suggests numerous open problems.\n",
"title": "Geometry of Log-Concave Density Estimation"
} | null | null | null | null | true | null | 20042 | null | Default | null | null |
null | {
"abstract": " The aim of this paper is to analyze the array synthesis for 5 G massive MIMO\nsystems in the line-of-sight working condition. The main result of the\nnumerical investigation performed is that non-uniform arrays are the natural\nchoice in this kind of application. In particular, by using non-equispaced\narrays, we show that it is possible to achieve a better average condition\nnumber of the channel matrix and a significantly higher spectral efficiency.\nFurthermore, we verify that increasing the array size is beneficial also for\ncircular arrays, and we provide some useful rules-of-thumb for antenna array\ndesign for massive MIMO applications. These results are in contrast to the\nwidely-accepted idea in the 5 G massive MIMO literature, in which the\nhalf-wavelength linear uniform array is universally adopted.\n",
"title": "Antenna Arrays for Line-of-Sight Massive MIMO: Half Wavelength is not Enough"
} | null | null | null | null | true | null | 20043 | null | Default | null | null |
null | {
"abstract": " In this study, the gravitational octree code originally optimized for the\nFermi, Kepler, and Maxwell GPU architectures is adapted to the Volta\narchitecture. The Volta architecture introduces independent thread scheduling\nrequiring either the insertion of the explicit synchronizations at appropriate\nlocations or the enforcement of the same implicit synchronizations as do the\nPascal or earlier architectures by specifying \\texttt{-gencode\narch=compute\\_60,code=sm\\_70}. The performance measurements on Tesla V100, the\ncurrent flagship GPU by NVIDIA, revealed that the $N$-body simulations of the\nAndromeda galaxy model with $2^{23} = 8388608$ particles took $3.8 \\times\n10^{-2}$~s or $3.3 \\times 10^{-2}$~s per step for each case. Tesla V100\nachieves a 1.4 to 2.2-fold acceleration in comparison with Tesla P100, the\nflagship GPU in the previous generation. The observed speed-up of 2.2 is\ngreater than 1.5, which is the ratio of the theoretical peak performance of the\ntwo GPUs. The independence of the units for integer operations from those for\nfloating-point number operations enables the overlapped execution of integer\nand floating-point number operations. It hides the execution time of the\ninteger operations leading to the speed-up rate above the theoretical peak\nperformance ratio. Tesla V100 can execute $N$-body simulation with up to $25\n\\times 2^{20} = 26214400$ particles, and it took $2.0 \\times 10^{-1}$~s per\nstep. It corresponds to $3.5$~TFlop/s, which is 22\\% of the single-precision\ntheoretical peak performance.\n",
"title": "Gravitational octree code performance evaluation on Volta GPU"
} | null | null | null | null | true | null | 20044 | null | Default | null | null |
null | {
"abstract": " Instanton bundles on $\\mathbb{P}^3$ have been at the core of the research in\nAlgebraic Geometry during the last thirty years. Motivated by the recent\nextension of their definition to other Fano threefolds of Picard number one, we\ndevelop the theory of instanton bundles on the complete flag variety\n$F:=F(0,1,2)$ of point-lines on $\\mathbb{P}^2$. After giving for them two\ndifferent monadic presentations, we use it to show that the moduli space\n$MI_F(k)$ of instanton bundles of charge $k$ is a geometric GIT quotient with a\ngenerically smooth component of dim $8k-3$. Finally we study their locus of\njumping conics.\n",
"title": "Instanton bundles on the flag variety F(0,1,2)"
} | null | null | null | null | true | null | 20045 | null | Default | null | null |
null | {
"abstract": " Learning weights in a spiking neural network with hidden neurons, using\nlocal, stable and online rules, to control non-linear body dynamics is an open\nproblem. Here, we employ a supervised scheme, Feedback-based Online Local\nLearning Of Weights (FOLLOW), to train a network of heterogeneous spiking\nneurons with hidden layers, to control a two-link arm so as to reproduce a\ndesired state trajectory. The network first learns an inverse model of the\nnon-linear dynamics, i.e. from state trajectory as input to the network, it\nlearns to infer the continuous-time command that produced the trajectory.\nConnection weights are adjusted via a local plasticity rule that involves\npre-synaptic firing and post-synaptic feedback of the error in the inferred\ncommand. We choose a network architecture, termed differential feedforward,\nthat gives the lowest test error from different feedforward and recurrent\narchitectures. The learned inverse model is then used to generate a\ncontinuous-time motor command to control the arm, given a desired trajectory.\n",
"title": "Non-linear motor control by local learning in spiking neural networks"
} | null | null | null | null | true | null | 20046 | null | Default | null | null |
null | {
"abstract": " In this work, we propose to train a deep neural network by distributed\noptimization over a graph. Two nonlinear functions are considered: the\nrectified linear unit (ReLU) and a linear unit with both lower and upper\ncutoffs (DCutLU). The problem reformulation over a graph is realized by\nexplicitly representing ReLU or DCutLU using a set of slack variables. We then\napply the alternating direction method of multipliers (ADMM) to update the\nweights of the network layerwise by solving subproblems of the reformulated\nproblem. Empirical results suggest that the ADMM-based method is less sensitive\nto overfitting than the stochastic gradient descent (SGD) and Adam methods.\n",
"title": "Training Deep Neural Networks via Optimization Over Graphs"
} | null | null | null | null | true | null | 20047 | null | Default | null | null |
null | {
"abstract": " An analytical expression is received for the effective interaction potential\nof a fast charged particle with the ionic crystal CsCl near the direction of\n<100> axis as a function of the temperature of the medium. By numerical\nanalysis it is shown that the effective potential of axial channeling of\npositrons along the axis <100> of negatively charged ions practically does not\ndepend on temperature of the media\n",
"title": "Calculation of Effective Interaction Potential During Positron Channeling in Ionic Crystals"
} | null | null | null | null | true | null | 20048 | null | Default | null | null |
null | {
"abstract": " Online Social Networks (OSN) are increasingly being used as platform for an\neffective communication, to engage with other users, and to create a social\nworth via number of likes, followers and shares. Such metrics and crowd-sourced\nratings give the OSN user a sense of social reputation which she tries to\nmaintain and boost to be more influential. Users artificially bolster their\nsocial reputation via black-market web services. In this work, we identify\nusers which manipulate their projected follower count using an unsupervised\nlocal neighborhood detection method. We identify a neighborhood of the user\nbased on a robust set of features which reflect user similarity in terms of the\nexpected follower count. We show that follower count estimation using our\nmethod has 84.2% accuracy with a low error rate. In addition, we estimate the\nfollower count of the user under suspicion by finding its neighborhood drawn\nfrom a large random sample of Twitter. We show that our method is highly\ntolerant to synthetic manipulation of followers. Using the deviation of\npredicted follower count from the displayed count, we are also able to detect\ncustomers with a high precision of 98.62%\n",
"title": "The Follower Count Fallacy: Detecting Twitter Users with Manipulated Follower Count"
} | null | null | null | null | true | null | 20049 | null | Default | null | null |
null | {
"abstract": " We characterize operator-theoretic properties (boundedness, compactness, and\nSchatten class membership) of Toeplitz operators with positive measure symbols\non Bergman spaces of holomorphic hermitian line bundles over Kähler\nCartan-Hadamard manifolds in terms of geometric or operator-theoretic\nproperties of measures.\n",
"title": "Bergman kernel estimates and Toeplitz operators on holomorphic line bundles"
} | null | null | null | null | true | null | 20050 | null | Default | null | null |
null | {
"abstract": " We present results from the first observations of the Hubble Space Telescope\n(HST) Panchromatic Comparative Exoplanet Treasury (PanCET) program for\nWASP-101b, a highly inflated hot Jupiter and one of the community targets\nproposed for the James Webb Space Telescope (JWST) Early Release Science (ERS)\nprogram. From a single HST Wide Field Camera 3 (WFC3) observation, we find that\nthe near-infrared transmission spectrum of WASP-101b contains no significant\nH$_2$O absorption features and we rule out a clear atmosphere at 13{\\sigma}.\nTherefore, WASP-101b is not an optimum target for a JWST ERS program aimed at\nobserving strong molecular transmission features. We compare WASP-101b to the\nwell studied and nearly identical hot Jupiter WASP-31b. These twin planets show\nsimilar temperature-pressure profiles and atmospheric features in the\nnear-infrared. We suggest exoplanets in the same parameter space as WASP-101b\nand WASP-31b will also exhibit cloudy transmission spectral features. For\nfuture HST exoplanet studies, our analysis also suggests that a lower count\nlimit needs to be exceeded per pixel on the detector in order to avoid unwanted\ninstrumental systematics.\n",
"title": "HST PanCET program: A Cloudy Atmosphere for the promising JWST target WASP-101b"
} | null | null | null | null | true | null | 20051 | null | Default | null | null |
null | {
"abstract": " We derive the gaugeon formalism of the Kalb-Ramond field theory, a reducible\ngauge theory, which discusses the quantum gauge freedom. In gaugeon formalism,\ntheory admits quantum gauge symmetry which leaves the action form-invariant.\nThe BRST symmetric gaugeon formalism is also studied which introduces the\ngaugeon ghost fields and gaugeon ghosts of ghosts fields. To replace the\nYokoyama subsidiary conditions by a single Kugo-Ojima type condition the virtue\nof BRST symmetry is utilized. Under generalized BRST transformations, we show\nthat the gaugeon fields appear naturally in the reducible gauge theory.\n",
"title": "Quantum gauge symmetry of reducible gauge theory"
} | null | null | null | null | true | null | 20052 | null | Default | null | null |
null | {
"abstract": " In this work, the authors give a new method for phase determination, the Tian\npseudo atom method (TPAM) or pseudo atom method (PAM) for short. In this new\nmethod, the figure of merit function, Rtian, replaces Rcf in the charge\nflipping algorithm. The key difference between Rcf and Rtian is the oberved\nstructure factor was replaced by the pseudo structure factor. The test results\nshow that Rtian is more powerful and robust than Rcf to estimate the correct\nstructure especially with low resolution data. Therefore, the pseudo atom\nmethod could overcome the charge flipping method's defeat to some extent. In\ntheory, the pseudo atom method could deal with quite low resolution data but it\nneeds a further test.\n",
"title": "The Tian Pseudo-Atom Method"
} | null | null | null | null | true | null | 20053 | null | Default | null | null |
null | {
"abstract": " Complex distribution networks are pervasive in biology. Examples include\nnutrient transport in the slime mold $Physarum$ $polycephalum$ as well as\nmammalian and plant venation. Adaptive rules are believed to guide development\nof these networks and lead to a reticulate, hierarchically nested topology that\nis both efficient and resilient against perturbations. However, as of yet no\nmechanism is known that can generate such networks on all scales. We show how\nhierarchically organized reticulation can be generated and maintained through\nspatially collective load fluctuations on a particular length scale. We\ndemonstrate that the resulting network topologies represent a trade-off between\noptimizing power dissipation, construction cost, and damage robustness and\nidentify the Pareto-efficient front that evolution is expected to favor and\nselect for. We show that the typical fluctuation length scale controls the\nposition of the networks on the Pareto front and thus on the spectrum of\nvenation phenotypes. We compare the Pareto archetypes predicted by our model\nwith examples of real leaf networks.\n",
"title": "The Phenotypes of Fluctuating Flow: Development of Distribution Networks in Biology and the Trade-off between Efficiency, Cost, and Resilience"
} | null | null | null | null | true | null | 20054 | null | Default | null | null |
null | {
"abstract": " Deep learning methods employ multiple processing layers to learn hierarchical\nrepresentations of data and have produced state-of-the-art results in many\ndomains. Recently, a variety of model designs and methods have blossomed in the\ncontext of natural language processing (NLP). In this paper, we review\nsignificant deep learning related models and methods that have been employed\nfor numerous NLP tasks and provide a walk-through of their evolution. We also\nsummarize, compare and contrast the various models and put forward a detailed\nunderstanding of the past, present and future of deep learning in NLP.\n",
"title": "Recent Trends in Deep Learning Based Natural Language Processing"
} | null | null | null | null | true | null | 20055 | null | Default | null | null |
null | {
"abstract": " The increasing impact of robotics on industry and on society will unavoidably\nlead to the involvement of robots in incidents and mishaps. In such cases,\nforensic analyses are key techniques to provide useful evidence on what\nhappened, and try to prevent future incidents. This article discusses volatile\nmemory forensics for the Robot Operating System (ROS). The authors start by\nproviding a general overview of forensic techniques in robotics and then\npresent a robotics-specific Volatility plugin named linux_rosnode, packaged\nwithin the ros_volatility project and aimed to extract evidence from robot's\nvolatile memory. They demonstrate how this plugin can be used to detect a\nspecific attack pattern on ROS, where a publisher node is unregistered\nexternally, leading to denial of service and disruption of robotic behaviors.\nStep-by-step, common practices are introduced for performing forensic analysis\nand several techniques to capture memory are described. The authors finalize by\nintroducing some future remarks while providing references to reproduce their\nwork.\n",
"title": "Volatile memory forensics for the Robot Operating System"
} | null | null | null | null | true | null | 20056 | null | Default | null | null |
null | {
"abstract": " Remote Attestation (RA) allows a trusted entity (verifier) to securely\nmeasure internal state of a remote untrusted hardware platform (prover). RA can\nbe used to establish a static or dynamic root of trust in embedded and\ncyber-physical systems. It can also be used as a building block for other\nsecurity services and primitives, such as software updates and patches,\nverifiable deletion and memory resetting. There are three major classes of RA\ndesigns: hardware-based, software-based, and hybrid, each with its own set of\nbenefits and drawbacks. This paper presents the first hybrid RA design, called\nHYDRA, that builds upon formally verified software components that ensure\nmemory isolation and protection, as well as enforce access control to memory\nand other resources. HYDRA obtains these properties by using the formally\nverified seL4 microkernel. (Until now, this was only attainable with purely\nhardware-based designs.) Using seL4 requires fewer hardware modifications to\nthe underlying microprocessor. Building upon a formally verified software\ncomponent increases confidence in security of the overall design of HYDRA and\nits implementation. We instantiate HYDRA on two commodity hardware platforms\nand assess the performance and overhead of performing RA on such platforms via\nexperimentation; we show that HYDRA can attest 10MB of memory in less than\n500msec when using a Speck-based message authentication code (MAC) to compute a\ncryptographic checksum over the memory to be attested.\n",
"title": "HYDRA: HYbrid Design for Remote Attestation (Using a Formally Verified Microkernel)"
} | null | null | null | null | true | null | 20057 | null | Default | null | null |
null | {
"abstract": " Social robots, also known as service or assistant robots, have been developed\nto improve the quality of human life in recent years. The design of socially\ncapable and intelligent robots can vary, depending on the target user groups.\nIn this work, we assess the effect of social robots' roles, functions, and\ncommunication approaches in the context of a social agent providing service or\nentertainment to users with developmental disabilities. In this paper, we\ndescribe an exploratory study of interface design for a social robot that\nassists people suffering from developmental disabilities. We developed series\nof prototypes and tested one in a user study that included three residents with\nvarious function levels. This entire study had been recorded for the following\nqualitative data analysis. Results show that each design factor played a\ndifferent role in delivering information and in increasing engagement. We also\nnote that some of the fundamental design principles that would work for\nordinary users did not apply to our target user group. We conclude that social\nrobots could benefit our target users, and acknowledge that these robots were\nnot suitable for certain scenarios based on the feedback from our users.\n",
"title": "Social Robots for People with Developmental Disabilities: A User Study on Design Features of a Graphical User Interface"
} | null | null | null | null | true | null | 20058 | null | Default | null | null |
null | {
"abstract": " In this paper we solve the inverse problem for a class of mean field models\n(Curie-Weiss model and its multi-species version) when multiple thermodynamic\nstates are present, as in the low temperature phase where the phase space is\nclustered. The inverse problem consists in reconstructing the model parameters\nstarting from configuration data generated according to the distribution of the\nmodel. We show that the application of the inversion procedure without taking\ninto account the presence of many states produces very poor inference results.\nThis problem is overcomed using the clustering algorithm. When the system has\ntwo symmetric states of positive and negative magnetization, the parameter\nreconstruction can be also obtained with smaller computational effort simply by\nflipping the sign of the magnetizations from positive to negative (or\nviceversa). The parameter reconstruction fails when the system is critical: in\nthis case we give the correct inversion formulas for the Curie-Weiss model and\nwe show that they can be used to measuring how much the system is close to\ncriticality.\n",
"title": "Inverse problem for multi-species mean field models in the low temperature phase"
} | null | null | null | null | true | null | 20059 | null | Default | null | null |
null | {
"abstract": " Building a complete inertial navigation system using the limited quality data\nprovided by current smartphones has been regarded challenging, if not\nimpossible. This paper shows that by careful crafting and accounting for the\nweak information in the sensor samples, smartphones are capable of pure\ninertial navigation. We present a probabilistic approach for orientation and\nuse-case free inertial odometry, which is based on double-integrating rotated\naccelerations. The strength of the model is in learning additive and\nmultiplicative IMU biases online. We are able to track the phone position,\nvelocity, and pose in real-time and in a computationally lightweight fashion by\nsolving the inference with an extended Kalman filter. The information fusion is\ncompleted with zero-velocity updates (if the phone remains stationary),\naltitude correction from barometric pressure readings (if available), and\npseudo-updates constraining the momentary speed. We demonstrate our approach\nusing an iPad and iPhone in several indoor dead-reckoning applications and in a\nmeasurement tool setup.\n",
"title": "Inertial Odometry on Handheld Smartphones"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 20060 | null | Validated | null | null |
null | {
"abstract": " We demonstrate that photonic and phononic crystals consisting of closely\nspaced inclusions constitute a versatile class of subwavelength metamaterials.\nIntuitively, the voids and narrow gaps that characterise the crystal form an\ninterconnected network of Helmholtz-like resonators. We use this intuition to\nargue that these continuous photonic (phononic) crystals are in fact\nasymptotically equivalent, at low frequencies, to discrete capacitor-inductor\n(mass-spring) networks whose lumped parameters we derive explicitly. The\ncrystals are tantamount to metamaterials as their entire acoustic branch, or\nbranches when the discrete analogue is polyatomic, is squeezed into a\nsubwavelength regime where the ratio of wavelength to period scales like the\nratio of period to gap width raised to the power 1/4; at yet larger wavelengths\nwe accordingly find a comparably large effective refractive index. The fully\nanalytical dispersion relations predicted by the discrete models yield\ndispersion curves that agree with those from finite-element simulations of the\ncontinuous crystals. The insight gained from the network approach is used to\nshow that, surprisingly, the continuum created by a closely packed hexagonal\nlattice of cylinders is represented by a discrete honeycomb lattice. The\nanalogy is utilised to show that the hexagonal continuum lattice has a\nDirac-point degeneracy that is lifted in a controlled manner by specifying the\narea of a symmetry-breaking defect.\n",
"title": "Asymptotic network models of subwavelength metamaterials formed by closely packed photonic and phononic crystals"
} | null | null | null | null | true | null | 20061 | null | Default | null | null |
null | {
"abstract": " The experimental efforts to detect the redshifted 21 cm signal from the Epoch\nof Reionization (EoR) are limited predominantly by the chromatic instrumental\nsystematic effect. The delay spectrum methodology for 21 cm power spectrum\nmeasurements brought new attention to the critical impact of an antenna's\nchromaticity on the viability of making this measurement. This methodology\nestablished a straightforward relationship between time-domain response of an\ninstrument and the power spectrum modes accessible to a 21 cm EoR experiment.\nWe examine the performance of a prototype of the Hydrogen Epoch of Reionization\nArray (HERA) array element that is currently observing in Karoo desert, South\nAfrica. We present a mathematical framework to derive the beam integrated\nfrequency response of a HERA prototype element in reception from the return\nloss measurements between 100-200 MHz and determined the extent of additional\nforeground contamination in the delay space. The measurement reveals excess\nspectral structures in comparison to the simulation studies of the HERA\nelement. Combined with the HERA data analysis pipeline that incorporates\ninverse covariance weighting in optimal quadratic estimation of power spectrum,\nwe find that in spite of its departure from the simulated response, HERA\nprototype element satisfies the necessary criteria posed by the foreground\nattenuation limits and potentially can measure the power spectrum at spatial\nmodes as low as $k_{\\parallel} > 0.1h$~Mpc$^{-1}$. The work highlights a\nstraightforward method for directly measuring an instrument response and\nassessing its impact on 21 cm EoR power spectrum measurements for future\nexperiments that will use reflector-type antenna.\n",
"title": "The Hydrogen Epoch of Reionization Array Dish III: Measuring Chromaticity of Prototype Element with Reflectometry"
} | null | null | null | null | true | null | 20062 | null | Default | null | null |
null | {
"abstract": " We unify the feeding and feedback of supermassive black holes with the global\nproperties of galaxies, groups, and clusters, by linking for the first time the\nphysical mechanical efficiency at the horizon and Mpc scale. The macro hot halo\nis tightly constrained by the absence of overheating and overcooling as probed\nby X-ray data and hydrodynamic simulations ($\\varepsilon_{\\rm BH} \\simeq$\n10$^{-3}\\, T_{\\rm x,7.4}$). The micro flow is shaped by general relativistic\neffects tracked by state-of-the-art GR-RMHD simulations ($\\varepsilon_\\bullet\n\\simeq$ 0.03). The SMBH properties are tied to the X-ray halo temperature\n$T_{\\rm x}$, or related cosmic scaling relation (as $L_{\\rm x}$). The model is\nminimally based on first principles, as conservation of energy and mass\nrecycling. The inflow occurs via chaotic cold accretion (CCA), the rain of cold\nclouds condensing out of the quenched cooling flow and recurrently funneled via\ninelastic collisions. Within 100s gravitational radii, the accretion energy is\ntransformed into ultrafast 10$^4$ km s$^{-1}$ outflows (UFOs) ejecting most of\nthe inflowing mass. At larger radii the energy-driven outflow entrains\nprogressively more mass: at roughly kpc scale, the velocities of the\nhot/warm/cold outflows are a few 10$^3$, 1000, 500 km s$^{-1}$, with median\nmass rates ~10, 100, several 100 M$_\\odot$ yr$^{-1}$, respectively. The unified\nCCA model is consistent with the observations of nuclear UFOs, and ionized,\nneutral, and molecular macro outflows. We provide step-by-step implementation\nfor subgrid simulations, (semi)analytic works, or observational interpretations\nwhich require self-regulated AGN feedback at coarse scales, avoiding the\na-posteriori fine-tuning of efficiencies.\n",
"title": "Unifying the micro and macro properties of AGN feeding and feedback"
} | null | null | [
"Physics"
]
| null | true | null | 20063 | null | Validated | null | null |
null | {
"abstract": " The inherent noise in the observed (e.g., scanned) binary document image\ndegrades the image quality and harms the compression ratio through breaking the\npattern repentance and adding entropy to the document images. In this paper, we\ndesign a cost function in Bayesian framework with dictionary learning.\nMinimizing our cost function produces a restored image which has better quality\nthan that of the observed noisy image, and a dictionary for representing and\nencoding the image. After the restoration, we use this dictionary (from the\nsame cost function) to encode the restored image following the\nsymbol-dictionary framework by JBIG2 standard with the lossless mode.\nExperimental results with a variety of document images demonstrate that our\nmethod improves the image quality compared with the observed image, and\nsimultaneously improves the compression ratio. For the test images with\nsynthetic noise, our method reduces the number of flipped pixels by 48.2% and\nimproves the compression ratio by 36.36% as compared with the best encoding\nmethods. For the test images with real noise, our method visually improves the\nimage quality, and outperforms the cutting-edge method by 28.27% in terms of\nthe compression ratio.\n",
"title": "Model-based Iterative Restoration for Binary Document Image Compression with Dictionary Learning"
} | null | null | null | null | true | null | 20064 | null | Default | null | null |
null | {
"abstract": " Online advertisers and analytics services (or trackers), are constantly\ntracking users activities as they access web services either through browsers\nor a mobile apps. Numerous tools such as browser plugins and specialized mobile\napps have been proposed to limit intrusive advertisements and prevent tracking\non desktop computing and mobile phones. For desktop computing, browser plugins\nare heavily studied for their usability and efficiency issues, however, tools\nthat block ads and prevent trackers in mobile platforms, have received the\nleast or no attention.\nIn this paper, we present a first look at 97 Android adblocking apps (or\nadblockers), extracted from more than 1.5 million apps from Google Play, that\npromise to block advertisements and analytics services. With our data\ncollection and analysis pipeline of the Android adblockers, we reveal the\npresences of third-party tracking libraries and sensitive permissions for\ncritical resources on user mobile devices as well as have malware in the source\ncodes. We analyze users' reviews for the in-effectiveness of adblockers in\nterms of not blocking ads and trackers. We found that a significant fraction of\nadblockers are not fulfilling their advertised functionality.\n",
"title": "A First Look at Ad Blocking Apps on Google Play"
} | null | null | null | null | true | null | 20065 | null | Default | null | null |
null | {
"abstract": " We consider the $2+1$ Toda system \\[ \\frac{1}{4}\\Delta\nq_{n}=e^{q_{n-1}-q_{n}}-e^{q_{n}-q_{n+1}}\\text{ in }\\mathbb{R}^{2},\\\nn\\in\\mathbb{Z}. \\] It has a traveling wave type solution $\\left\\{ Q_{n}\\right\\}\n$ satisfying $Q_{n+1}(x,y)=Q_{n}(x+\\frac{1}{2\\sqrt{2}},y)$, and is explicitly\ngiven by \\[ Q_{n}\\left( x,y\\right) =\\ln\\frac{\\frac{1}{4}+\\left(\nn-1+2\\sqrt{2}x\\right) ^{2}+4y^{2}}{\\frac{1}{4}+\\left( n+2\\sqrt{2}x\\right)\n^{2}+4y^{2}}. \\] In this paper we prove that \\{$Q_{n}$\\} is nondegenerate.\n",
"title": "Nondegeneracy of the traveling lump solution to the $2+1$ Toda lattice"
} | null | null | null | null | true | null | 20066 | null | Default | null | null |
null | {
"abstract": " We prove that a meromorphic mapping, which sends a peace of a real analytic\nstrictly pseudoconvex hypersurface in $\\cc^2$ to a compact subset of $\\cc^N$\nwhich doesn't contain germs of non-constant complex curves is continuous from\nthe concave side of the hypersurface. This implies the analytic continuability\nalong CR-paths of germs of holomorphic mappings from real analytic\nhypersurfaces with non-vanishing Levi form to the locally spherical ones in all\ndimensions.\n",
"title": "One side continuity of meromorphic mappings between real analytic hypersurfaces"
} | null | null | null | null | true | null | 20067 | null | Default | null | null |
null | {
"abstract": " Cameras are the most widely exploited sensor in both robotics and computer\nvision communities. Despite their popularity, two dominant attributes (i.e.,\ngain and exposure time) have been determined empirically and images are\ncaptured in very passive manner. In this paper, we present an active and\ngeneric camera attribute control scheme using Bayesian optimization. We extend\nfrom our previous work [1] in two aspects. First, we propose a method that\njointly controls camera gain and exposure time. Secondly, to speed up the\nBayesian optimization process, we introduce image synthesis using the camera\nresponse function (CRF). These synthesized images allowed us to diminish the\nimage acquisition time during the Bayesian optimization phase, substantially\nimproving overall control performance. The proposed method is validated both in\nan indoor and an outdoor environment where light condition rapidly changes.\nSupplementary material is available at this https URL .\n",
"title": "Generic Camera Attribute Control using Bayesian Optimization"
} | null | null | null | null | true | null | 20068 | null | Default | null | null |
null | {
"abstract": " The time evolution of the energy transport triggered in a strongly coupled\nsystem by a temperature gradient is holographically related to the evolution of\nan asymptotically AdS black brane. We study the far-from-equilibrium properties\nof such a system by using the AdS/CFT correspondence. In particular, we\ndescribe the appearance of a steady state, and study the information flow by\ncomputing the time evolution of the holographic entanglement entropy. Some\nuniversal properties of the quenching process are presented.\n",
"title": "Far-from-equilibrium energy flow and entanglement entropy"
} | null | null | null | null | true | null | 20069 | null | Default | null | null |
null | {
"abstract": " This chapter of the forthcoming Handbook of Graphical Models contains an\noverview of basic theorems and techniques from algebraic geometry and how they\ncan be applied to the study of conditional independence and graphical models.\nIt also introduces binomial ideals and some ideas from real algebraic geometry.\nWhen random variables are discrete or Gaussian, tools from computational\nalgebraic geometry can be used to understand implications between conditional\nindependence statements. This is accomplished by computing primary\ndecompositions of conditional independence ideals. As examples the chapter\npresents in detail the graphical model of a four cycle and the intersection\naxiom, a certain implication of conditional independence statements. Another\nimportant problem in the area is to determine all constraints on a graphical\nmodel, for example, equations determined by trek separation. The full set of\nequality constraints can be determined by computing the model's vanishing\nideal. The chapter illustrates these techniques and ideas with examples from\nthe literature and provides references for further reading.\n",
"title": "Algebraic Aspects of Conditional Independence and Graphical Models"
} | null | null | null | null | true | null | 20070 | null | Default | null | null |
null | {
"abstract": " This paper presents a survey of some new applications of algebraic Riccati\nequations. In particular, the paper surveys some recent results on the use of\nalgebraic Riccati equations in testing whether a system is negative imaginary\nand in synthesizing state feedback controllers which make the closed loop\nsystem negative imaginary. The paper also surveys the use of Riccati equation\nmethods in the control of quantum linear systems including coherent $H^\\infty$\ncontrol.\n",
"title": "A Survey of Riccati Equation Results in Negative Imaginary Systems Theory and Quantum Control Theory"
} | null | null | [
"Computer Science",
"Mathematics"
]
| null | true | null | 20071 | null | Validated | null | null |
null | {
"abstract": " HD$\\,$169142 is an excellent target to investigate signs of planet-disk\ninteraction due to the previous evidence of gap structures. We performed J-band\n(~1.2{\\mu}m) polarized intensity imaging of HD169142 with VLT/SPHERE. We\nobserve polarized scattered light down to 0.16\" (~19 au) and find an inner gap\nwith a significantly reduced scattered light flux. We confirm the previously\ndetected double ring structure peaking at 0.18\" (~21 au) and 0.56\" (~66 au),\nand marginally detect a faint third gap at 0.70\"-0.73\" (~82-85 au). We explore\ndust evolution models in a disk perturbed by two giant planets, as well as\nmodels with a parameterized dust size distribution. The dust evolution model is\nable to reproduce the ring locations and gap widths in polarized intensity, but\nfails to reproduce their depths. It, however, gives a good match with the ALMA\ndust continuum image at 1.3 mm. Models with a parameterized dust size\ndistribution better reproduce the gap depth in scattered light, suggesting that\ndust filtration at the outer edges of the gaps is less effective. The pile-up\nof millimeter grains in a dust trap and the continuous distribution of small\ngrains throughout the gap likely require a more efficient dust fragmentation\nand dust diffusion in the dust trap. Alternatively, turbulence or charging\neffects might lead to a reservoir of small grains at the surface layer that is\nnot affected by the dust growth and fragmentation cycle dominating the dense\ndisk midplane. The exploration of models shows that extracting planet\nproperties such as mass from observed gap profiles is highly degenerate.\n",
"title": "The circumstellar disk HD$\\,$169142: gas, dust and planets acting in concert?"
} | null | null | [
"Physics"
]
| null | true | null | 20072 | null | Validated | null | null |
null | {
"abstract": " The mechanism of ion bombardment induced magnetic patterning of exchange bias\nlayer systems for creating engineered magnetic stray field landscapes is still\nunclear. We compare results from vectorial magneto-optic Kerr effect\nmeasurements to a recently proposed model with time dependent rotatable\nmagnetic anisotropy. Results show massive reduction of rotational magnetic\nanisotropy compared to all other magnetic anisotropies. We disprove the\nassumption of comparable weakening of all magnetic anisotropies and show that\nion bombardment mainly influences smaller grains in the antiferromagnet.\n",
"title": "Quantitative analysis of the influence of keV He ion bombardment on exchange bias layer systems"
} | null | null | null | null | true | null | 20073 | null | Default | null | null |
null | {
"abstract": " We implement two algorithms in MATHEMATICA for classifying automorphisms of\nlower-dimensional non-commutative Lie algebras. The first algorithm is a\nbrute-force approach whereas the second is an evolutionary strategy. These\nalgorithms are delivered as the MATHEMATICA package cwsAutoClass. In order to\nfacilitate the application of this package to symmetry Lie algebras of\ndifferential equations, we also provide a package, cwsLieSymTools, for\nmanipulating finite-dimensional Lie algebras of vector fields. In particular,\nthis package allows the computations of Lie brackets, structure constants, and\nthe visualization of commutator tables. Several examples are provided to\nillustrate the pertinence of our approach.\n",
"title": "Automatic classification of automorphisms of lower-dimensional Lie algebras"
} | null | null | null | null | true | null | 20074 | null | Default | null | null |
null | {
"abstract": " We show that model compression can improve the population risk of a\npre-trained model, by studying the tradeoff between the decrease in the\ngeneralization error and the increase in the empirical risk with model\ncompression. We first prove that model compression reduces an\ninformation-theoretic bound on the generalization error; this allows for an\ninterpretation of model compression as a regularization technique to avoid\noverfitting. We then characterize the increase in empirical risk with model\ncompression using rate distortion theory. These results imply that the\npopulation risk could be improved by model compression if the decrease in\ngeneralization error exceeds the increase in empirical risk. We show through a\nlinear regression example that such a decrease in population risk due to model\ncompression is indeed possible. Our theoretical results further suggest that\nthe Hessian-weighted $K$-means clustering compression approach can be improved\nby regularizing the distance between the clustering centers. We provide\nexperiments with neural networks to support our theoretical assertions.\n",
"title": "Information-Theoretic Understanding of Population Risk Improvement with Model Compression"
} | null | null | null | null | true | null | 20075 | null | Default | null | null |
null | {
"abstract": " This paper presents a procedure to retrieve subsets of relevant documents\nfrom large text collections for Content Analysis, e.g. in social sciences.\nDocument retrieval for this purpose needs to take account of the fact that\nanalysts often cannot describe their research objective with a small set of key\nterms, especially when dealing with theoretical or rather abstract research\ninterests. Instead, it is much easier to define a set of paradigmatic documents\nwhich reflect topics of interest as well as targeted manner of speech. Thus, in\ncontrast to classic information retrieval tasks we employ manually compiled\ncollections of reference documents to compose large queries of several hundred\nkey terms, called dictionaries. We extract dictionaries via Topic Models and\nalso use co-occurrence data from reference collections. Evaluations show that\nthe procedure improves retrieval results for this purpose compared to\nalternative methods of key term extraction as well as neglecting co-occurrence\ndata.\n",
"title": "Document Retrieval for Large Scale Content Analysis using Contextualized Dictionaries"
} | null | null | null | null | true | null | 20076 | null | Default | null | null |
null | {
"abstract": " Space photometric missions have been steadily accumulating observations of\nCepheids in recent years, leading to a flow of new discoveries. In this short\nreview we summarize the findings provided by the early missions such as WIRE,\nMOST, and CoRoT, and the recent results of the Kepler and K2 missions. The\nsurprising and fascinating results from the high-precision, quasi-continuous\ndata include the detection of the amplitude increase of Polaris, and exquisite\ndetails about V1154 Cyg within the original Kepler field of view. We also\nbriefly discuss the current opportunities with the K2 mission, and the\nprospects of the TESS space telescope regarding Cepheids.\n",
"title": "Cepheids with the eyes of photometric space telescopes"
} | null | null | null | null | true | null | 20077 | null | Default | null | null |
null | {
"abstract": " We present an explicitly correlated formalism for the second-order\nsingle-particle Green's function method (GF2-F12) that does not assume the\npopular diagonal approximation, and describes the energy dependence of the\nexplicitly correlated terms. For small and medium organic molecules the basis\nset errors of ionization potentials of GF2-F12 are radically improved relative\nto GF2: the performance of GF2-F12/aug- cc-pVDZ is better than that of\nGF2/aug-cc-pVQZ, at a significantly lower cost.\n",
"title": "Explicitly correlated formalism for second-order single-particle Green's function"
} | null | null | null | null | true | null | 20078 | null | Default | null | null |
null | {
"abstract": " We study algebro-geometric consequences of the quantised extremal Kähler\nmetrics, introduced in the previous work of the author. We prove that the\nexistence of quantised extremal metrics implies weak relative Chow\npolystability. As a consequence, we obtain asymptotic weak relative Chow\npolystability and $K$-semistability of extremal manifolds by using quantised\nextremal metrics; this gives an alternative proof of the results of Mabuchi and\nStoppa--Székelyhidi. In proving them, we further provide an explicit local\ndensity formula for the equivariant Riemann--Roch theorem.\n",
"title": "Relative stability associated to quantised extremal Kähler metrics"
} | null | null | null | null | true | null | 20079 | null | Default | null | null |
null | {
"abstract": " Companies and academic researchers may collect, process, and distribute large\nquantities of personal data without the explicit knowledge or consent of the\nindividuals to whom the data pertains. Existing forms of consent often fail to\nbe appropriately readable and ethical oversight of data mining may not be\nsufficient. This raises the question of whether existing consent instruments\nare sufficient, logistically feasible, or even necessary, for data mining. In\nthis chapter, we review the data collection and mining landscape, including\ncommercial and academic activities, and the relevant data protection concerns,\nto determine the types of consent instruments used. Using three case studies,\nwe use the new paradigm of human-data interaction to examine whether these\nexisting approaches are appropriate. We then introduce an approach to consent\nthat has been empirically demonstrated to improve on the state of the art and\ndeliver meaningful consent. Finally, we propose some best practices for data\ncollectors to ensure their data mining activities do not violate the\nexpectations of the people to whom the data relate.\n",
"title": "Beyond the EULA: Improving consent for data mining"
} | null | null | null | null | true | null | 20080 | null | Default | null | null |
null | {
"abstract": " A fundamental problem in neuroscience is to characterize the dynamics of\nspiking from the neurons in a circuit that is involved in learning about a\nstimulus or a contingency. A key limitation of current methods to analyze\nneural spiking data is the need to collapse neural activity over time or\ntrials, which may cause the loss of information pertinent to understanding the\nfunction of a neuron or circuit. We introduce a new method that can determine\nnot only the trial-to-trial dynamics that accompany the learning of a\ncontingency by a neuron, but also the latency of this learning with respect to\nthe onset of a conditioned stimulus. The backbone of the method is a separable\ntwo-dimensional (2D) random field (RF) model of neural spike rasters, in which\nthe joint conditional intensity function of a neuron over time and trials\ndepends on two latent Markovian state sequences that evolve separately but in\nparallel. Classical tools to estimate state-space models cannot be applied\nreadily to our 2D separable RF model. We develop efficient statistical and\ncomputational tools to estimate the parameters of the separable 2D RF model. We\napply these to data collected from neurons in the pre-frontal cortex (PFC) in\nan experiment designed to characterize the neural underpinnings of the\nassociative learning of fear in mice. Overall, the separable 2D RF model\nprovides a detailed, interpretable, characterization of the dynamics of neural\nspiking that accompany the learning of a contingency.\n",
"title": "Estimating a Separably-Markov Random Field (SMuRF) from Binary Observations"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 20081 | null | Validated | null | null |
null | {
"abstract": " Deep reinforcement learning enables algorithms to learn complex behavior,\ndeal with continuous action spaces and find good strategies in environments\nwith high dimensional state spaces. With deep reinforcement learning being an\nactive area of research and many concurrent inventions, we decided to focus on\na relatively simple robotic task to evaluate a set of ideas that might help to\nsolve recent reinforcement learning problems. We test a newly created\ncombination of two commonly used reinforcement learning methods, whether it is\nable to learn more effectively than a baseline. We also compare different ideas\nto preprocess information before it is fed to the reinforcement learning\nalgorithm. The goal of this strategy is to reduce training time and eventually\nhelp the algorithm to converge. The concluding evaluation proves the general\napplicability of the described concepts by testing them using a simulated\nenvironment. These concepts might be reused for future experiments.\n",
"title": "Using Deep Reinforcement Learning for the Continuous Control of Robotic Arms"
} | null | null | null | null | true | null | 20082 | null | Default | null | null |
null | {
"abstract": " This article deals with the first detection of gravitational waves by the\nadvanced Laser Interferometer Gravitational Wave Observatory (LIGO) detectors\non 14 September 2015, where the signal was generated by two stellar mass black\nholes with masses 36 $ M_{\\odot}$ and 29 $ M_{\\odot}$ that merged to form a 62\n$ M_{\\odot}$ black hole, releasing 3 $M_{\\odot}$ energy in gravitational waves,\nalmost 1.3 billion years ago. We begin by providing a brief overview of\ngravitational waves, their sources and the gravitational wave detectors. We\nthen describe in detail the first detection of gravitational waves from a\nbinary black hole merger. We then comment on the electromagnetic follow up of\nthe detection event with various telescopes. Finally, we conclude with the\ndiscussion on the tests of gravity and fundamental physics with the first\ngravitational wave detection event.\n",
"title": "The First Detection of Gravitational Waves"
} | null | null | null | null | true | null | 20083 | null | Default | null | null |
null | {
"abstract": " The remarkably strong chemical adsorption behaviors of nitric oxide on\nmagnesia (001) film deposited on metal substrate have been investigated by\nemploying periodic density functional calculations with Van der Waals\ncorrections. The molybdenum supported magnesia (001) show significantly\nenhanced adsorption properties and the nitric oxide is chemisorbed strongly and\npreferably trapped in flat adsorption configuration on metal supported oxide\nfilm, due to the substantially large adsorption energies and transformation\nbarriers. The analysis of Bader charges, projected density of states,\ndifferential charge densities, electron localization function, highest occupied\norbital and particular orbital with largest Mg-NO-Mg bonding coefficients, are\napplied to reveal the electronic adsorption properties and characteristics of\nbonding between nitric oxide and surface as well as the bonding within the\nhybrid structure. The strong chemical binding of nitric oxide on magnesia\ndeposited on molybdenum slab offers new opportunities for toxic gas detection\nand treatment. We anticipate that hybrid structure promoted remarkable chemical\nadsorption of nitric oxide on magnesia in this study will provide versatile\nstrategy for enhancing chemical reactivity and properties of insulating oxide.\n",
"title": "Remarkably strong chemisorption of nitric oxide on insulating oxide films promoted by hybrid structure"
} | null | null | null | null | true | null | 20084 | null | Default | null | null |
null | {
"abstract": " In this work we propose approaches to effectively transfer knowledge from\nweakly labeled web audio data. We first describe a convolutional neural network\n(CNN) based framework for sound event detection and classification using weakly\nlabeled audio data. Our model trains efficiently from audios of variable\nlengths; hence, it is well suited for transfer learning. We then propose\nmethods to learn representations using this model which can be effectively used\nfor solving the target task. We study both transductive and inductive transfer\nlearning tasks, showing the effectiveness of our methods for both domain and\ntask adaptation. We show that the learned representations using the proposed\nCNN model generalizes well enough to reach human level accuracy on ESC-50 sound\nevents dataset and set state of art results on this dataset. We further use\nthem for acoustic scene classification task and once again show that our\nproposed approaches suit well for this task as well. We also show that our\nmethods are helpful in capturing semantic meanings and relations as well.\nMoreover, in this process we also set state-of-art results on Audioset dataset,\nrelying on balanced training set.\n",
"title": "Knowledge Transfer from Weakly Labeled Audio using Convolutional Neural Network for Sound Events and Scenes"
} | null | null | null | null | true | null | 20085 | null | Default | null | null |
null | {
"abstract": " We derive a priori estimates for the incompressible free-boundary Euler\nequations with surface tension in three spatial dimensions. Working in\nLagrangian coordinates, we provide a priori estimates for the local existence\nwhen the initial velocity, which is rotational, belongs to $H^3$ and the trace\nof initial velocity on the free boundary to $H^{3.5}$, thus lowering the\nrequirement on the regularity of initial data in the Lagrangian setting. Our\nmethods are direct and involve three key elements: estimates for the pressure,\nthe boundary regularity provided by the mean curvature, and the Cauchy\ninvariance.\n",
"title": "A priori estimates for the free-boundary Euler equations with surface tension in three dimensions"
} | null | null | null | null | true | null | 20086 | null | Default | null | null |
null | {
"abstract": " Let $M$ be a Liouville 6-manifold which is the smooth fiber of a Lefschetz\nfibration on $\\mathbb{C}^4$ constructed by suspending a Lefschetz fibration on\n$\\mathbb{C}^3$. We prove that for many examples including stabilizations of\nMilnor fibers of hypersurface cusp singularities, the compact Fukaya category\n$\\mathcal{F}(M)$ and the wrapped Fukaya category $\\mathcal{W}(M)$ are related\nthrough $A_\\infty$-Koszul duality, by identifying them with cyclic and\nCalabi-Yau completions of the same quiver algebra. This implies the\nsplit-generation of the compact Fukaya category $\\mathcal{F}(M)$ by vanishing\ncycles. Moreover, new examples of Liouville manifolds which admit\nquasi-dilations in the sense of Seidel-Solomon are obtained.\n",
"title": "Koszul duality via suspending Lefschetz fibrations"
} | null | null | null | null | true | null | 20087 | null | Default | null | null |
null | {
"abstract": " We construct an iterated function system consisting of strictly increasing\ncontractions $f,g\\colon [0,1]\\to [0,1]$ with $f([0,1])\\cap g([0,1])=\\emptyset$\nand such that its attractor has positive Lebesgue measure.\n",
"title": "Attractor of Cantor Type with Positive Measure"
} | null | null | null | null | true | null | 20088 | null | Default | null | null |
null | {
"abstract": " Recent years have seen tremendous growth of many online social networks such\nas Facebook, LinkedIn and MySpace. People connect to each other through these\nnetworks forming large social communities providing researchers rich datasets\nto understand, model and predict social interactions and behaviors. New\ncontacts in these networks can be formed due to an individual's demographic\nattributes such as age group, gender, geographic location, or due to a\nnetwork's structural dynamics such as triadic closure and preferential\nattachment, or a combination of both demographic and structural\ncharacteristics.\nA number of network generation models have been proposed in the last decade\nto explain the structure, evolution and processes taking place in different\ntypes of networks, and notably social networks. Network generation models\nstudied in the literature primarily consider structural properties, and in some\ncases an individual's demographic profile in the formation of new social\ncontacts. These models do not present a mechanism to combine both structural\nand demographic characteristics for the formation of new links. In this paper,\nwe propose a new network generation algorithm which incorporates both these\ncharacteristics to model network formation. We use different publicly available\nFacebook datasets as benchmarks to demonstrate the correctness of the proposed\nnetwork generation model. The proposed model is flexible and thus can generate\nnetworks with varying demographic and structural properties.\n",
"title": "Generating online social networks based on socio-demographic attributes"
} | null | null | [
"Computer Science",
"Physics"
]
| null | true | null | 20089 | null | Validated | null | null |
null | {
"abstract": " A model for the development of turbulent shear flows, created by non-uniform\nparallel flows in a confining channel, is used to identify the diffuser shape\nthat maximises pressure recovery when the inflow is non-uniform. Wide diffuser\nangles tend to accentuate the non- uniform flow, causing poor pressure\nrecovery. On the other hand, shallow diffuser angles create longer regions with\nlarge wall drag, which is also detrimental to pressure recovery. Thus, optimal\ndiffuser shapes strike a balance between the two effects. We use a simple model\nwhich describes the evolution of an approximate flow profile and pressure in\nthe diffuser. The model equations form the dynamics of an optimal control\nproblem where the control is the diffuser channel shape. A numerical\noptimisation approach is used to solve the optimal control problem and we use\nanalytical results to interpret the numerics in some limiting cases. The\nresults of the optimisation are compared to calculations from computational\nfluid dynamics.\n",
"title": "Optimal control of diffuser shapes for confined turbulent shear flows"
} | null | null | null | null | true | null | 20090 | null | Default | null | null |
null | {
"abstract": " This paper gives a thorough overview of Solar Car Optimized Route Estimation\n(SCORE), novel route optimization scheme for solar vehicles based on solar\nirradiance and target distance. In order to conduct the optimization, both data\ncollection and the optimization algorithm itself have to be performed using\nappropriate hardware. Here we give an insight to both stages, hardware and\nsoftware used and present some results of the SCORE system together with\ncertain improvements of its fusion and optimization criteria. Results and the\nlimited applicability of SCORE are discussed together with an overview of\nfuture research plans and comparison with state-of-the-art solar vehicle\noptimization solutions.\n",
"title": "Criteria for Solar Car Optimized Route Estimation"
} | null | null | null | null | true | null | 20091 | null | Default | null | null |
null | {
"abstract": " The present study investigates a way to design dykes which can filter the\nwavelengths of ocean surface waves. This offers the possibility to achieve a\nstructure that can attenuate waves associated with storm swell, without\naffecting coastline in other conditions. Our approach is based on low frequency\nresonances in metamaterials combined with Bragg frequencies for which waves\ncannot propagate in periodic lattices.\n",
"title": "Dykes for filtering ocean waves using c-shaped vertical cylinders"
} | null | null | null | null | true | null | 20092 | null | Default | null | null |
null | {
"abstract": " A fraction of early-type dwarf galaxies in the Virgo cluster have a disk\ncomponent and even possess disk features such as bar, lens, and spiral arms. In\nthis study, we construct 15 galaxy models that resemble VCC856, considered to\nbe an infalling progenitor of disk dwarf galaxies, within observational error\nranges, and use $N$-body simulations to study their long-term dynamical\nevolution in isolation as well as the formation of bar in them. We find that\ndwarf disk galaxies readily form bars unless they have an excessively\nconcentrated halo or a hot disk. This suggests that infalling dwarf disk\ngalaxies are intrinsically unstable to bar formation, even without any external\nperturbation, accounting for a population of barred dwarf galaxies in the\noutskirts of the Virgo cluster. The bars form earlier and stronger in galaxies\nwith a lower fraction of counter-streaming motions, lower halo concentration,\nlower velocity anisotropy, and thinner disk. Similarly to normal disk galaxies,\ndwarf disk galaxies also undergo recurrent buckling instabilities. The first\nbuckling instability tends to shorten the bar and to thicken the disk, and\ndrives a dynamical transition in the bar pattern speed as well as mass inflow\nrate. In nine models, the bars regrow after the mild first buckling instability\ndue to the efficient transfer of disk angular momentum to the halo, and are\nsubject to recurrent buckling instabilities to turn into X-shaped bulges.\n",
"title": "Origin of Non-axisymmetric Features of Virgo Cluster Early-type Dwarf Galaxies. I. Bar Formation and Recurrent Buckling"
} | null | null | null | null | true | null | 20093 | null | Default | null | null |
null | {
"abstract": " This paper reports the first results of a direct dark matter search with the\nDEAP-3600 single-phase liquid argon (LAr) detector. The experiment was\nperformed 2 km underground at SNOLAB (Sudbury, Canada) utilizing a large target\nmass, with the LAr target contained in a spherical acrylic vessel of 3600 kg\ncapacity. The LAr is viewed by an array of PMTs, which would register\nscintillation light produced by rare nuclear recoil signals induced by dark\nmatter particle scattering. An analysis of 4.44 live days (fiducial exposure of\n9.87 tonne-days) of data taken with the nearly full detector during the initial\nfilling phase demonstrates the detector performance and the best electronic\nrecoil rejection using pulse-shape discrimination in argon, with leakage\n$<1.2\\times 10^{-7}$ (90% C.L.) between 16 and 33 keV$_{ee}$. No candidate\nsignal events are observed, which results in the leading limit on WIMP-nucleon\nspin-independent cross section on argon, $<1.2\\times 10^{-44}$ cm$^2$ for a 100\nGeV/c$^2$ WIMP mass (90% C.L.).\n",
"title": "First results from the DEAP-3600 dark matter search with argon at SNOLAB"
} | null | null | null | null | true | null | 20094 | null | Default | null | null |
null | {
"abstract": " We address the problem of epipolar geometry using the motion of silhouettes.\nSuch methods match epipolar lines or frontier points across views, which are\nthen used as the set of putative correspondences. We introduce an approach that\nimproves by two orders of magnitude the performance over state-of-the-art\nmethods, by significantly reducing the number of outliers in the putative\nmatching. We model the frontier points' correspondence problem as constrained\nflow optimization, requiring small differences between their coordinates over\nconsecutive frames. Our approach is formulated as a Linear Integer Program and\nwe show that due to the nature of our problem, it can be solved efficiently in\nan iterative manner. Our method was validated on four standard datasets\nproviding accurate calibrations across very different viewpoints.\n",
"title": "Camera Calibration by Global Constraints on the Motion of Silhouettes"
} | null | null | null | null | true | null | 20095 | null | Default | null | null |
null | {
"abstract": " This paper describes a new modelling language for the effective design and\nvalidation of Java annotations. Since their inclusion in the 5th edition of\nJava, annotations have grown from a useful tool for the addition of meta-data\nto play a central role in many popular software projects. Usually they are not\nconceived in isolation, but in groups, with dependency and integrity\nconstraints between them. However, the native support provided by Java for\nexpressing this design is very limited.\nTo overcome its deficiencies and make explicit the rich conceptual model\nwhich lies behind a set of annotations, we propose a domain-specific modelling\nlanguage. The proposal has been implemented as an Eclipse plug-in, including an\neditor and an integrated code generator that synthesises annotation processors.\nThe environment also integrates a model finder, able to detect unsatisfiable\nconstraints between different annotations, and to provide examples of correct\nannotation usages for validation. The language has been tested using a real set\nof annotations from the Java Persistence API (JPA). Within this subset we have\nfound enough rich semantics expressible with Ann and omitted nowadays by the\nJava language, which shows the benefits of Ann in a relevant field of\napplication.\n",
"title": "Ann: A domain-specific language for the effective design and validation of Java annotations"
} | null | null | null | null | true | null | 20096 | null | Default | null | null |
null | {
"abstract": " The new wave of successful generative models in machine learning has\nincreased the interest in deep learning driven de novo drug design. However,\nassessing the performance of such generative models is notoriously difficult.\nMetrics that are typically used to assess the performance of such generative\nmodels are the percentage of chemically valid molecules or the similarity to\nreal molecules in terms of particular descriptors, such as the partition\ncoefficient (logP) or druglikeness. However, method comparison is difficult\nbecause of the inconsistent use of evaluation metrics, the necessity for\nmultiple metrics, and the fact that some of these measures can easily be\ntricked by simple rule-based systems. We propose a novel distance measure\nbetween two sets of molecules, called Fréchet ChemNet distance (FCD), that\ncan be used as an evaluation metric for generative models. The FCD is similar\nto a recently established performance metric for comparing image generation\nmethods, the Fréchet Inception Distance (FID). Whereas the FID uses one of\nthe hidden layers of InceptionNet, the FCD utilizes the penultimate layer of a\ndeep neural network called ChemNet, which was trained to predict drug\nactivities. Thus, the FCD metric takes into account chemically and biologically\nrelevant information about molecules, and also measures the diversity of the\nset via the distribution of generated molecules. The FCD's advantage over\nprevious metrics is that it can detect if generated molecules are a) diverse\nand have similar b) chemical and c) biological properties as real molecules. We\nfurther provide an easy-to-use implementation that only requires the SMILES\nrepresentation of the generated molecules as input to calculate the FCD.\nImplementations are available at: this https URL\n",
"title": "Fréchet ChemNet Distance: A metric for generative models for molecules in drug discovery"
} | null | null | null | null | true | null | 20097 | null | Default | null | null |
null | {
"abstract": " We consider the RANSAC algorithm in the context of subspace recovery and\nsubspace clustering. We derive some theory and perform some numerical\nexperiments. We also draw some correspondences with the methods of Hardt and\nMoitra (2013) and Chen and Lerman (2009b).\n",
"title": "RANSAC Algorithms for Subspace Recovery and Subspace Clustering"
} | null | null | null | null | true | null | 20098 | null | Default | null | null |
null | {
"abstract": " Though there is a growing body of literature on fairness for supervised\nlearning, the problem of incorporating fairness into unsupervised learning has\nbeen less well-studied. This paper studies fairness in the context of principal\ncomponent analysis (PCA). We first present a definition of fairness for\ndimensionality reduction, and our definition can be interpreted as saying that\na reduction is fair if information about a protected class (e.g., race or\ngender) cannot be inferred from the dimensionality-reduced data points. Next,\nwe develop convex optimization formulations that can improve the fairness (with\nrespect to our definition) of PCA and kernel PCA. These formulations are\nsemidefinite programs (SDP's), and we demonstrate the effectiveness of our\nformulations using several datasets. We conclude by showing how our approach\ncan be used to perform a fair (with respect to age) clustering of health data\nthat may be used to set health insurance rates.\n",
"title": "Convex Formulations for Fair Principal Component Analysis"
} | null | null | [
"Statistics"
]
| null | true | null | 20099 | null | Validated | null | null |
null | {
"abstract": " In the research of the impact of gestures using by a lecturer, one\nchallenging task is to infer the attention of a group of audiences. Two\nimportant measurements that can help infer the level of attention are eye\nmovement data and Electroencephalography (EEG) data. Under the fundamental\nassumption that a group of people would look at the same place if they all pay\nattention at the same time, we apply a method, \"Time Warp Edit Distance\", to\ncalculate the similarity of their eye movement trajectories. Moreover, we also\ncluster eye movement pattern of audiences based on these pair-wised similarity\nmetrics. Besides, since we don't have a direct metric for the \"attention\"\nground truth, a visual assessment would be beneficial to evaluate the\ngesture-attention relationship. Thus we also implement a visualization tool.\n",
"title": "Report: Dynamic Eye Movement Matching and Visualization Tool in Neuro Gesture"
} | null | null | null | null | true | null | 20100 | null | Default | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.