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 Transiting Exoplanet Survey Satellite (TESS) will embark in 2018 on a\n2-year wide-field survey mission, discovering over a thousand terrestrial,\nsuper-Earth and sub-Neptune-sized exoplanets potentially suitable for follow-up\nobservations using the James Webb Space Telescope (JWST). This work aims to\nunderstand the suitability of anticipated TESS planet discoveries for\natmospheric characterization by JWST's Near InfraRed Imager and Slitless\nSpectrograph (NIRISS) by employing a simulation tool to estimate the\nsignal-to-noise (S/N) achievable in transmission spectroscopy. We applied this\ntool to Monte Carlo predictions of the TESS expected planet yield and then\ncompared the S/N for anticipated TESS discoveries to our estimates of S/N for\n18 known exoplanets. We analyzed the sensitivity of our results to planetary\ncomposition, cloud cover, and presence of an observational noise floor. We\nfound that several hundred anticipated TESS discoveries with radii from 1.5 to\n2.5 times the Earth's radius will produce S/N higher than currently known\nexoplanets in this radius regime, such as K2-3b or K2-3c. In the terrestrial\nplanet regime, we found that only a few anticipated TESS discoveries will\nresult in higher S/N than currently known exoplanets, such as the TRAPPIST-1\nplanets, GJ1132b, and LHS1140b. However, we emphasize that this outcome is\nbased upon Kepler-derived occurrence rates, and that co-planar compact\nmulti-planet systems (e.g., TRAPPIST-1) may be under-represented in the\npredicted TESS planet yield. Finally, we apply our calculations to estimate the\nrequired magnitude of a JWST follow-up program devoted to mapping the\ntransition region between hydrogen-dominated and high molecular weight\natmospheres. We find that a modest observing program of between 60 to 100 hours\nof charged JWST time can define the nature of that transition (e.g., step\nfunction versus a power law).\n",
"title": "Simulated JWST/NIRISS Transit Spectroscopy of Anticipated TESS Planets Compared to Select Discoveries from Space-Based and Ground-Based Surveys"
}
| null | null | null | null | true | null |
5301
| null |
Default
| null | null |
null |
{
"abstract": " We deal with the symmetries of a (2-term) graded vector space or bundle. Our\nfirst theorem shows that they define a (strict) Lie 2-groupoid in a natural\nway. Our second theorem explores the construction of nerves for Lie\n2-categories, showing that it yields simplicial manifolds if the 2-cells are\ninvertible. Finally, our third and main theorem shows that smooth\npseudofunctors into our general linear 2-groupoid classify 2-term\nrepresentations up to homotopy of Lie groupoids.\n",
"title": "The general linear 2-groupoid"
}
| null | null | null | null | true | null |
5302
| null |
Default
| null | null |
null |
{
"abstract": " Traffic flow prediction is an important research issue to avoid traffic\ncongestion in transportation systems. Traffic congestion avoiding can be\nachieved by knowing traffic flow and then conducting transportation planning.\nAchieving traffic flow prediction is challenging as the prediction is affected\nby many complex factors such as inter-region traffic, vehicles' relations, and\nsudden events. However, as the mobile data of vehicles has been widely\ncollected by sensor-embedded devices in transportation systems, it is possible\nto predict the traffic flow by analysing mobile data. This study proposes a\ndeep learning based prediction algorithm, DeepTFP, to collectively predict the\ntraffic flow on each and every traffic road of a city. This algorithm uses\nthree deep residual neural networks to model temporal closeness, period, and\ntrend properties of traffic flow. Each residual neural network consists of a\nbranch of residual convolutional units. DeepTFP aggregates the outputs of the\nthree residual neural networks to optimize the parameters of a time series\nprediction model. Contrast experiments on mobile time series data from the\ntransportation system of England demonstrate that the proposed DeepTFP\noutperforms the Long Short-Term Memory (LSTM) architecture based method in\nprediction accuracy.\n",
"title": "DeepTFP: Mobile Time Series Data Analytics based Traffic Flow Prediction"
}
| null | null | null | null | true | null |
5303
| null |
Default
| null | null |
null |
{
"abstract": " A non-equilibrium theory of optical conductivity of dirty-limit\nsuperconductors and commensurate charge density wave is presented. We discuss\nthe current response to different experimentally relevant light-field probe\npulses and show that a single frequency definition of the optical conductivity\n$\\sigma(\\omega)\\equiv j(\\omega)/E(\\omega)$ is difficult to interpret out of the\nadiabatic limit. We identify characteristic time domain signatures\ndistinguishing between superconducting, normal metal and charge density wave\nstates. We also suggest a route to directly address the instantaneous\nsuperfluid stiffness of a superconductor by shaping the probe light field.\n",
"title": "Non-equilibrium Optical Conductivity: General Theory and Application to Transient Phases"
}
| null | null |
[
"Physics"
] | null | true | null |
5304
| null |
Validated
| null | null |
null |
{
"abstract": " We report high-resolution neutron Compton scattering measurements of liquid\n$^4$He under saturated vapor pressure. There is excellent agreement between the\nobserved scattering and ab initio predictions of its lineshape. Quantum Monte\nCarlo calculations predict that the Bose condensate fraction is zero in the\nnormal fluid, builds up rapidly just below the superfluid transition\ntemperature, and reaches a value of approximately $7.5\\%$ below 1 K. We also\nused model fit functions to obtain from the scattering data empirical estimates\nfor the average atomic kinetic energy and Bose condensate fraction. These\nquantities are also in excellent agreement with ab initio calculations. The\nconvergence between the scattering data and Quantum Monte Carlo calculations is\nstrong evidence for a Bose broken symmetry in superfluid $^4$He.\n",
"title": "The Momentum Distribution of Liquid $^4$He"
}
| null | null |
[
"Physics"
] | null | true | null |
5305
| null |
Validated
| null | null |
null |
{
"abstract": " We study the problem of semantic code repair, which can be broadly defined as\nautomatically fixing non-syntactic bugs in source code. The majority of past\nwork in semantic code repair assumed access to unit tests against which\ncandidate repairs could be validated. In contrast, the goal here is to develop\na strong statistical model to accurately predict both bug locations and exact\nfixes without access to information about the intended correct behavior of the\nprogram. Achieving such a goal requires a robust contextual repair model, which\nwe train on a large corpus of real-world source code that has been augmented\nwith synthetically injected bugs. Our framework adopts a two-stage approach\nwhere first a large set of repair candidates are generated by rule-based\nprocessors, and then these candidates are scored by a statistical model using a\nnovel neural network architecture which we refer to as Share, Specialize, and\nCompete. Specifically, the architecture (1) generates a shared encoding of the\nsource code using an RNN over the abstract syntax tree, (2) scores each\ncandidate repair using specialized network modules, and (3) then normalizes\nthese scores together so they can compete against one another in comparable\nprobability space. We evaluate our model on a real-world test set gathered from\nGitHub containing four common categories of bugs. Our model is able to predict\nthe exact correct repair 41\\% of the time with a single guess, compared to 13\\%\naccuracy for an attentional sequence-to-sequence model.\n",
"title": "Semantic Code Repair using Neuro-Symbolic Transformation Networks"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5306
| null |
Validated
| null | null |
null |
{
"abstract": " Several authors have claimed that the less luminous active galactic nuclei\n(AGN) are not capable of sustaining the dusty torus structure. Thus, a gradual\nre-sizing of the torus is expected when the AGN luminosity decreases. Our aim\nis to confront mid-infrared observations of local AGN of different luminosities\nwith this scenario. We decomposed about ~100 IRS/Spitzer spectra of LLAGN and\npowerful Seyferts in order to decontaminate the torus component from other\ncontributors. We have used the affinity propagation (AP) method to cluster the\ndata into five groups within the sample according to torus contribution to the\n5-15 um range (Ctorus) and bolometric luminosity. The AP groups show a\nprogressively higher torus contribution and an increase of the bolometric\nluminosity, from Group 1 (Ctorus~ 0% and logLbol ~ 41) and up to Group 5\n(Ctorus ~80% and log(Lbol) ~44). We have fitted the average spectra of each of\nthe AP groups to clumpy models. The torus is no longer present in Group 1,\nsupporting the disappearance at low-luminosities. We were able to fit the\naverage spectra for the torus component in Groups 3 (Ctorus~ 40% and log(Lbol)~\n42.6), 4 (Ctorus~ 60% and log(Lbol)~ 43.7), and 5 to Clumpy torus models. We\ndid not find a good fitting to Clumpy torus models for Group 2 (Ctorus~ 18% and\nlog(Lbol)~ 42). This might suggest a different configuration and/or composition\nof the clouds for Group 2, which is consistent with a different gas content\nseen in Groups 1, 2, and 3, according to the detections of H2 molecular lines.\nGroups 3, 4, and 5 show a trend to decrease of the width of the torus (which\nyields to a likely decrease of the geometrical covering factor), although we\ncannot confirm it with the present data. Finally, Groups 3, 4, and 5 show an\nincrease on the outer radius of the torus for higher luminosities, consistent\nwith a re-sizing of the torus according to the AGN luminosity.\n",
"title": "Hints on the gradual re-sizing of the torus in AGN by decomposing IRS/Spitzer spectra"
}
| null | null | null | null | true | null |
5307
| null |
Default
| null | null |
null |
{
"abstract": " We solve here completely an irrigation problem from a Dirac mass to the\nLebesgue measure. The functional we consider is a two dimensional analog of a\nfunctional previously derived in the study of branched patterns in type-I\nsuperconductors. The minimizer we obtain is a self-similar tree.\n",
"title": "Self-similar minimizers of a branched transport functional"
}
| null | null | null | null | true | null |
5308
| null |
Default
| null | null |
null |
{
"abstract": " One of the major challenges in object detection is to propose detectors with\nhighly accurate localization of objects. The online sampling of high-loss\nregion proposals (hard examples) uses the multitask loss with equal weight\nsettings across all loss types (e.g, classification and localization, rigid and\nnon-rigid categories) and ignores the influence of different loss distributions\nthroughout the training process, which we find essential to the training\nefficacy. In this paper, we present the Stratified Online Hard Example Mining\n(S-OHEM) algorithm for training higher efficiency and accuracy detectors.\nS-OHEM exploits OHEM with stratified sampling, a widely-adopted sampling\ntechnique, to choose the training examples according to this influence during\nhard example mining, and thus enhance the performance of object detectors. We\nshow through systematic experiments that S-OHEM yields an average precision\n(AP) improvement of 0.5% on rigid categories of PASCAL VOC 2007 for both the\nIoU threshold of 0.6 and 0.7. For KITTI 2012, both results of the same metric\nare 1.6%. Regarding the mean average precision (mAP), a relative increase of\n0.3% and 0.5% (1% and 0.5%) is observed for VOC07 (KITTI12) using the same set\nof IoU threshold. Also, S-OHEM is easy to integrate with existing region-based\ndetectors and is capable of acting with post-recognition level regressors.\n",
"title": "S-OHEM: Stratified Online Hard Example Mining for Object Detection"
}
| null | null | null | null | true | null |
5309
| null |
Default
| null | null |
null |
{
"abstract": " Dynamic topic modeling facilitates the identification of topical trends over\ntime in temporal collections of unstructured documents. We introduce a novel\nunsupervised neural dynamic topic model named as Recurrent Neural\nNetwork-Replicated Softmax Model (RNNRSM), where the discovered topics at each\ntime influence the topic discovery in the subsequent time steps. We account for\nthe temporal ordering of documents by explicitly modeling a joint distribution\nof latent topical dependencies over time, using distributional estimators with\ntemporal recurrent connections. Applying RNN-RSM to 19 years of articles on NLP\nresearch, we demonstrate that compared to state-of-the art topic models, RNNRSM\nshows better generalization, topic interpretation, evolution and trends. We\nalso introduce a metric (named as SPAN) to quantify the capability of dynamic\ntopic model to capture word evolution in topics over time.\n",
"title": "Deep Temporal-Recurrent-Replicated-Softmax for Topical Trends over Time"
}
| null | null | null | null | true | null |
5310
| null |
Default
| null | null |
null |
{
"abstract": " Embedding complex objects as vectors in low dimensional spaces is a\nlongstanding problem in machine learning. We propose in this work an extension\nof that approach, which consists in embedding objects as elliptical probability\ndistributions, namely distributions whose densities have elliptical level sets.\nWe endow these measures with the 2-Wasserstein metric, with two important\nbenefits: (i) For such measures, the squared 2-Wasserstein metric has a closed\nform, equal to a weighted sum of the squared Euclidean distance between means\nand the squared Bures metric between covariance matrices. The latter is a\nRiemannian metric between positive semi-definite matrices, which turns out to\nbe Euclidean on a suitable factor representation of such matrices, which is\nvalid on the entire geodesic between these matrices. (ii) The 2-Wasserstein\ndistance boils down to the usual Euclidean metric when comparing Diracs, and\ntherefore provides a natural framework to extend point embeddings. We show that\nfor these reasons Wasserstein elliptical embeddings are more intuitive and\nyield tools that are better behaved numerically than the alternative choice of\nGaussian embeddings with the Kullback-Leibler divergence. In particular, and\nunlike previous work based on the KL geometry, we learn elliptical\ndistributions that are not necessarily diagonal. We demonstrate the advantages\nof elliptical embeddings by using them for visualization, to compute embeddings\nof words, and to reflect entailment or hypernymy.\n",
"title": "Generalizing Point Embeddings using the Wasserstein Space of Elliptical Distributions"
}
| null | null | null | null | true | null |
5311
| null |
Default
| null | null |
null |
{
"abstract": " This paper describes our participation in Task 5 track 2 of SemEval 2017 to\npredict the sentiment of financial news headlines for a specific company on a\ncontinuous scale between -1 and 1. We tackled the problem using a number of\napproaches, utilising a Support Vector Regression (SVR) and a Bidirectional\nLong Short-Term Memory (BLSTM). We found an improvement of 4-6% using the LSTM\nmodel over the SVR and came fourth in the track. We report a number of\ndifferent evaluations using a finance specific word embedding model and reflect\non the effects of using different evaluation metrics.\n",
"title": "Lancaster A at SemEval-2017 Task 5: Evaluation metrics matter: predicting sentiment from financial news headlines"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5312
| null |
Validated
| null | null |
null |
{
"abstract": " The multivariate linear regression model is an important tool for\ninvestigating relationships between several response variables and several\npredictor variables. The primary interest is in inference about the unknown\nregression coefficient matrix. We propose multivariate bootstrap techniques as\na means for making inferences about the unknown regression coefficient matrix.\nThese bootstrapping techniques are extensions of those developed in Freedman\n(1981), which are only appropriate for univariate responses. Extensions to the\nmultivariate linear regression model are made without proof. We formalize this\nextension and prove its validity. A real data example and two simulated data\nexamples which offer some finite sample verification of our theoretical results\nare provided.\n",
"title": "Bootstrapping for multivariate linear regression models"
}
| null | null | null | null | true | null |
5313
| null |
Default
| null | null |
null |
{
"abstract": " We show that in certain one-dimensional spin chains with open boundary\nconditions, the edge spins retain memory of their initial state for very long\ntimes. The long coherence times do not require disorder, only an ordered phase.\nIn the integrable Ising and XYZ chains, the presence of a strong zero mode\nmeans the coherence time is infinite, even at infinite temperature. When Ising\nis perturbed by interactions breaking the integrability, the coherence time\nremains exponentially long in the perturbing couplings. We show that this is a\nconsequence of an edge \"almost\" strong zero mode that almost commutes with the\nHamiltonian. We compute this operator explicitly, allowing us to estimate\naccurately the plateau value of edge spin autocorrelator.\n",
"title": "Long coherence times for edge spins"
}
| null | null |
[
"Physics",
"Mathematics"
] | null | true | null |
5314
| null |
Validated
| null | null |
null |
{
"abstract": " We explore the potential of future cryogenic direct detection experiments to\ndetermine the properties of the mediator that communicates the interactions\nbetween dark matter and nuclei. Due to their low thresholds and large\nexposures, experiments like CRESST-III, SuperCDMS SNOLAB and EDELWEISS-III will\nhave excellent capability to reconstruct mediator masses in the MeV range for a\nlarge class of models. Combining the information from several experiments\nfurther improves the parameter reconstruction, even when taking into account\nadditional nuisance parameters related to background uncertainties and the dark\nmatter velocity distribution. These observations may offer the intriguing\npossibility of studying dark matter self-interactions with direct detection\nexperiments.\n",
"title": "Exploring light mediators with low-threshold direct detection experiments"
}
| null | null | null | null | true | null |
5315
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we present a family of conjectural relations in the\ntautological ring of the moduli spaces of stable curves which implies the\nstrong double ramification/Dubrovin-Zhang equivalence conjecture. Our\ntautological relations have the form of an equality between two different\nfamilies of tautological classes, only one of which involves the double\nramification cycle. We prove that both families behave the same way upon\npullback and pushforward with respect to forgetting a marked point. We also\nprove that our conjectural relations are true in genus $0$ and $1$ and also\nwhen first pushed forward from $\\overline{\\mathcal{M}}_{g,n+m}$ to\n$\\overline{\\mathcal{M}}_{g,n}$ and then restricted to $\\mathcal{M}_{g,n}$, for\nany $g,n,m\\geq 0$. Finally we show that, for semisimple CohFTs, the DR/DZ\nequivalence only depends on a subset of our relations, finite in each genus,\nwhich we prove for $g\\leq 2$. As an application we find a new formula for the\nclass $\\lambda_g$ as a linear combination of dual trees intersected with kappa\nand psi classes, and we check it for $g \\leq 3$.\n",
"title": "DR/DZ equivalence conjecture and tautological relations"
}
| null | null | null | null | true | null |
5316
| null |
Default
| null | null |
null |
{
"abstract": " We study data-driven representations for three-dimensional triangle meshes,\nwhich are one of the prevalent objects used to represent 3D geometry. Recent\nworks have developed models that exploit the intrinsic geometry of manifolds\nand graphs, namely the Graph Neural Networks (GNNs) and its spectral variants,\nwhich learn from the local metric tensor via the Laplacian operator. Despite\noffering excellent sample complexity and built-in invariances, intrinsic\ngeometry alone is invariant to isometric deformations, making it unsuitable for\nmany applications. To overcome this limitation, we propose several upgrades to\nGNNs to leverage extrinsic differential geometry properties of\nthree-dimensional surfaces, increasing its modeling power.\nIn particular, we propose to exploit the Dirac operator, whose spectrum\ndetects principal curvature directions --- this is in stark contrast with the\nclassical Laplace operator, which directly measures mean curvature. We coin the\nresulting models \\emph{Surface Networks (SN)}. We prove that these models\ndefine shape representations that are stable to deformation and to\ndiscretization, and we demonstrate the efficiency and versatility of SNs on two\nchallenging tasks: temporal prediction of mesh deformations under non-linear\ndynamics and generative models using a variational autoencoder framework with\nencoders/decoders given by SNs.\n",
"title": "Surface Networks"
}
| null | null | null | null | true | null |
5317
| null |
Default
| null | null |
null |
{
"abstract": " We used molecular dynamics simulations and the path sampling technique known\nas forward flux sampling to study homogeneous nucleation of NaCl crystals from\nsupersaturated aqueous solutions at 298 K and 1 bar. Nucleation rates were\nobtained for a range of salt concentrations for the Joung-Cheatham NaCl force\nfield combined with the SPC/E water model. The calculated nucleation rates are\nsignificantly lower than available experimental measurements. The estimates for\nthe nucleation rates in this work do not rely on classical nucleation theory,\nbut the pathways observed in the simulations suggest that the nucleation\nprocess is better described by classical nucleation theory than an alternative\ninterpretation based on Ostwald's step rule, in contrast to some prior\nsimulations of related models. In addition to the size of NaCl nucleus, we find\nthat the crystallinity of a nascent cluster plays an important role in the\nnucleation process. Nuclei with high crystallinity were found to have higher\ngrowth probability and longer lifetimes, possibly because they are less exposed\nto hydration water.\n",
"title": "Forward Flux Sampling Calculation of Homogeneous Nucleation Rates from Aqueous NaCl Solutions"
}
| null | null | null | null | true | null |
5318
| null |
Default
| null | null |
null |
{
"abstract": " First-passage time (FPT) of an Ornstein-Uhlenbeck (OU) process is of immense\ninterest in a variety of contexts. This paper considers an OU process with two\nboundaries, one of which is absorbing while the other one could be either\nreflecting or absorbing, and studies the control strategies that can lead to\ndesired FPT moments. Our analysis shows that the FPT distribution of an OU\nprocess is scale invariant with respect to the drift parameter, i.e., the drift\nparameter just controls the mean FPT and doesn't affect the shape of the\ndistribution. This allows to independently control the mean and coefficient of\nvariation (CV) of the FPT. We show that that increasing the threshold may\nincrease or decrease CV of the FPT, depending upon whether or not one of the\nthreshold is reflecting. We also explore the effect of control parameters on\nthe FPT distribution, and find parameters that minimize the distance between\nthe FPT distribution and a desired distribution.\n",
"title": "Driving an Ornstein--Uhlenbeck Process to Desired First-Passage Time Statistics"
}
| null | null | null | null | true | null |
5319
| null |
Default
| null | null |
null |
{
"abstract": " Perturbation theory using self-consistent Green's functions is one of the\nmost widely used approaches to study many-body effects in condensed matter. On\nthe basis of general considerations and by performing analytical calculations\nfor the specific example of the Hubbard atom, we discuss some key features of\nthis approach. We show that when the domain of the functionals that are used to\nrealize the map between the non-interacting and the interacting Green's\nfunctions is properly defined, there exists a class of self-energy functionals\nfor which the self-consistent Dyson equation has only one solution, which is\nthe physical one. We also show that manipulation of the perturbative expansion\nof the interacting Green's function may lead to a wrong self-energy as\nfunctional of the interacting Green's function, at least for some regions of\nthe parameter space. These findings confirm and explain numerical results of\nKozik et al. for the widely used skeleton series of Luttinger and Ward [Phys.\nRev. Lett. 114, 156402]. Our study shows that it is important to distinguish\nbetween the maps between sets of functions and the functionals that realize\nthose maps. We demonstrate that the self-consistent Green's functions approach\nitself is not problematic, whereas the functionals that are widely used may\nhave a limited range of validity.\n",
"title": "The self-consistent Dyson equation and self-energy functionals: failure or new opportunities?"
}
| null | null | null | null | true | null |
5320
| null |
Default
| null | null |
null |
{
"abstract": " The Affordable Care Act (ACA) includes a permanent revenue transfer\nmethodology which provides financial incentives to health insurance plans that\nhave higher than average actuarial risk. In this paper, we derive some\nstatistical implications of the revenue transfer methodology in the ACA. We\ntreat as random variables the revenue transfers between individual insurance\nplans in a given marketplace, where each plan's revenue transfer amount is\nmeasured as a percentage of the plan's total premium. We analyze the means and\nvariances of those random variables, and deduce from the zero sum nature of the\nrevenue transfers that there is no limit to the magnitude of revenue transfer\npayments relative to plans' total premiums. Using data provided by the American\nAcademy of Actuaries and by the Centers for Medicare and Medicaid Services, we\nobtain an explanation for empirical phenomena that revenue transfers were more\nvariable and can be substantially greater for insurance plans with smaller\nmarket shares. We show that it is often the case that an insurer which has\ndecreasing market share will also have increased volatility in its revenue\ntransfers.\n",
"title": "Statistical Implications of the Revenue Transfer Methodology in the Affordable Care Act"
}
| null | null | null | null | true | null |
5321
| null |
Default
| null | null |
null |
{
"abstract": " We investigate a class of chance-constrained combinatorial optimization\nproblems. Given a pre-specified risk level $\\epsilon \\in [0,1]$, the\nchance-constrained program aims to find the minimum cost selection of a vector\nof binary decisions $x$ such that a desirable event $\\mathcal{B}(x)$ occurs\nwith probability at least $ 1-\\epsilon$. In this paper, we assume that we have\nan oracle that computes $\\mathbb P( \\mathcal{B}(x))$ exactly. Using this\noracle, we propose a general exact method for solving the chance-constrained\nproblem. In addition, we show that if the chance-constrained program is solved\napproximately by a sampling-based approach, then the oracle can be used as a\ntool for checking and fixing the feasibility of the optimal solution given by\nthis approach. We demonstrate the effectiveness of our proposed methods on a\nvariant of the probabilistic set covering problem (PSC), which admits an\nefficient probability oracle. We give a compact mixed-integer program that\nsolves PSC optimally (without sampling) for a special case. For large-scale\ninstances for which the exact methods exhibit slow convergence, we propose a\nsampling-based approach that exploits the special structure of PSC. In\nparticular, we introduce a new class of facet-defining inequalities for a\nsubmodular substructure of PSC, and show that a sampling-based algorithm\ncoupled with the probability oracle solves the large-scale test instances\neffectively.\n",
"title": "Chance-Constrained Combinatorial Optimization with a Probability Oracle and Its Application to Probabilistic Partial Set Covering"
}
| null | null | null | null | true | null |
5322
| null |
Default
| null | null |
null |
{
"abstract": " This paper considers the optimal design of input signals for the purpose of\ndiscriminating among a finite number of affine models with uncontrolled inputs\nand noise. Each affine model represents a different system operating mode,\ncorresponding to unobserved intents of other drivers or robots, or to fault\ntypes or attack strategies, etc. The input design problem aims to find optimal\nseparating/discriminating (controlled) inputs such that the output trajectories\nof all the affine models are guaranteed to be distinguishable from each other,\ndespite uncertainty in the initial condition and uncontrolled inputs as well as\nthe presence of process and measurement noise. We propose a novel formulation\nto solve this problem, with an emphasis on guarantees for model discrimination\nand optimality, in contrast to a previously proposed conservative formulation\nusing robust optimization. This new formulation can be recast as a bilevel\noptimization problem and further reformulated as a mixed-integer linear program\n(MILP). Moreover, our fairly general problem setting allows the incorporation\nof objectives and/or responsibilities among rational agents. For instance, each\ndriver has to obey traffic rules, while simultaneously optimizing for safety,\ncomfort and energy efficiency. Finally, we demonstrate the effectiveness of our\napproach for identifying the intention of other vehicles in several driving\nscenarios.\n",
"title": "Optimal Input Design for Affine Model Discrimination with Applications in Intention-Aware Vehicles"
}
| null | null | null | null | true | null |
5323
| null |
Default
| null | null |
null |
{
"abstract": " The technique of continuous unitary transformations has recently been used to\nprovide physical insight into a diverse array of quantum mechanical systems.\nHowever, the question of how to best numerically implement the flow equations\nhas received little attention. The most immediately apparent approach, using\nstandard Runge-Kutta numerical integration algorithms, suffers from both severe\ninefficiency due to stiffness and the loss of unitarity. After reviewing the\nformalism of continuous unitary transformations and Wegner's original choice\nfor the infinitesimal generator of the flow, we present a number of approaches\nto resolving these issues including a choice of generator which induces what we\ncall the \"uniform tangent decay flow\" and three numerical integrators\nspecifically designed to perform continuous unitary transformations efficiently\nwhile preserving the unitarity of flow. We conclude by applying one of the flow\nalgorithms to a simple calculation that visually demonstrates the many-body\nlocalization transition.\n",
"title": "Stable Unitary Integrators for the Numerical Implementation of Continuous Unitary Transformations"
}
| null | null | null | null | true | null |
5324
| null |
Default
| null | null |
null |
{
"abstract": " Sparsity of the solution of a linear regression model is a common\nrequirement, and many prior distributions have been designed for this purpose.\nA combination of the sparsity requirement with smoothness of the solution is\nalso common in application, however, with considerably fewer existing prior\nmodels. In this paper, we compare two prior structures, the Bayesian fused\nlasso (BFL) and least-squares with adaptive prior covariance matrix (LS-APC).\nSince only variational solution was published for the latter, we derive a Gibbs\nsampling algorithm for its inference and Bayesian model selection. The method\nis designed for high dimensional problems, therefore, we discuss numerical\nissues associated with evaluation of the posterior. In simulation, we show that\nthe LS-APC prior achieves results comparable to that of the Bayesian Fused\nLasso for piecewise constant parameter and outperforms the BFL for parameters\nof more general shapes. Another advantage of the LS-APC priors is revealed in\nreal application to estimation of the release profile of the European Tracer\nExperiment (ETEX). Specifically, the LS-APC model provides more conservative\nuncertainty bounds when the regressor matrix is not informative.\n",
"title": "Sparse and Smooth Prior for Bayesian Linear Regression with Application to ETEX Data"
}
| null | null | null | null | true | null |
5325
| null |
Default
| null | null |
null |
{
"abstract": " Generalization error defines the discriminability and the representation\npower of a deep model. In this work, we claim that feature space design using\ndeep compositional function plays a significant role in generalization along\nwith explicit and implicit regularizations. Our claims are being established\nwith several image classification experiments. We show that the information\nloss due to convolution and max pooling can be marginalized with the\ncompositional design, improving generalization performance. Also, we will show\nthat learning rate decay acts as an implicit regularizer in deep model\ntraining.\n",
"title": "Deep Learning: Generalization Requires Deep Compositional Feature Space Design"
}
| null | null | null | null | true | null |
5326
| null |
Default
| null | null |
null |
{
"abstract": " We report on the detection of linear polarization of the forbidden [O i]\n630.03 nm spectral line. The observations were carried out in the broader\ncontext of the determination of the solar oxygen abundance, an important\nproblem in astrophysics that still remains unresolved. We obtained\nspectro-polarimetric data of the forbidden [O i] line at 630.03 nm as well as\nother neighboring permitted lines with the Solar Optical Telescope of the\nHinode satellite. A novel averaging technique was used, yielding very high\nsignal-to-noise ratios in excess of $10^5$. We confirm that the linear\npolarization is sign-reversed compared to permitted lines as a result of the\nline being dominated by a magnetic dipole transition. Our observations open a\nnew window for solar oxygen abundance studies, offering an alternative method\nto disentangle the Ni i blend from the [O i] line at 630.03 nm that has the\nadvantage of simple LTE formation physics.\n",
"title": "First detection of sign-reversed linear polarization from the forbidden [O I] 630.03 nm line"
}
| null | null | null | null | true | null |
5327
| null |
Default
| null | null |
null |
{
"abstract": " Neural networks are vulnerable to adversarial examples and researchers have\nproposed many heuristic attack and defense mechanisms. We address this problem\nthrough the principled lens of distributionally robust optimization, which\nguarantees performance under adversarial input perturbations. By considering a\nLagrangian penalty formulation of perturbing the underlying data distribution\nin a Wasserstein ball, we provide a training procedure that augments model\nparameter updates with worst-case perturbations of training data. For smooth\nlosses, our procedure provably achieves moderate levels of robustness with\nlittle computational or statistical cost relative to empirical risk\nminimization. Furthermore, our statistical guarantees allow us to efficiently\ncertify robustness for the population loss. For imperceptible perturbations,\nour method matches or outperforms heuristic approaches.\n",
"title": "Certifying Some Distributional Robustness with Principled Adversarial Training"
}
| null | null | null | null | true | null |
5328
| null |
Default
| null | null |
null |
{
"abstract": " Master equations are commonly used to model the dynamics of physical systems.\nSurprisingly, many deterministic maps $x \\rightarrow f(x)$ cannot be\nimplemented by any master equation, even approximately. This raises the\nquestion of how they arise in real-world systems like digital computers. We\nshow that any deterministic map over some \"visible\" states can be implemented\nwith a master equation--but only if additional \"hidden\" states are dynamically\ncoupled to those visible states. We also show that any master equation\nimplementing a given map can be decomposed into a sequence of \"hidden\"\ntimesteps, demarcated by changes in what transitions are allowed under the rate\nmatrix. Often there is a real-world cost for each additional hidden state, and\nfor each additional hidden timestep. We derive the associated \"space/time\"\ntradeoff between the numbers of hidden states and of hidden timesteps needed to\nimplement any given $f(x)$.\n",
"title": "The minimal hidden computer needed to implement a visible computation"
}
| null | null | null | null | true | null |
5329
| null |
Default
| null | null |
null |
{
"abstract": " Mega-city analysis with very high resolution (VHR) satellite images has been\ndrawing increasing interest in the fields of city planning and social\ninvestigation. It is known that accurate land-use, urban density, and\npopulation distribution information is the key to mega-city monitoring and\nenvironmental studies. Therefore, how to generate land-use, urban density, and\npopulation distribution maps at a fine scale using VHR satellite images has\nbecome a hot topic. Previous studies have focused solely on individual tasks\nwith elaborate hand-crafted features and have ignored the relationship between\ndifferent tasks. In this study, we aim to propose a universal framework which\ncan: 1) automatically learn the internal feature representation from the raw\nimage data; and 2) simultaneously produce fine-scale land-use, urban density,\nand population distribution maps. For the first target, a deep convolutional\nneural network (CNN) is applied to learn the hierarchical feature\nrepresentation from the raw image data. For the second target, a novel\nCNN-based universal framework is proposed to process the VHR satellite images\nand generate the land-use, urban density, and population distribution maps. To\nthe best of our knowledge, this is the first CNN-based mega-city analysis\nmethod which can process a VHR remote sensing image with such a large data\nvolume. A VHR satellite image (1.2 m spatial resolution) of the center of Wuhan\ncovering an area of 2606 km2 was used to evaluate the proposed method. The\nexperimental results confirm that the proposed method can achieve a promising\naccuracy for land-use, urban density, and population distribution maps.\n",
"title": "A multi-task convolutional neural network for mega-city analysis using very high resolution satellite imagery and geospatial data"
}
| null | null | null | null | true | null |
5330
| null |
Default
| null | null |
null |
{
"abstract": " Multi-attributed graph matching is a problem of finding correspondences\nbetween two sets of data while considering their complex properties described\nin multiple attributes. However, the information of multiple attributes is\nlikely to be oversimplified during a process that makes an integrated\nattribute, and this degrades the matching accuracy. For that reason, a\nmulti-layer graph structure-based algorithm has been proposed recently. It can\neffectively avoid the problem by separating attributes into multiple layers.\nNonetheless, there are several remaining issues such as a scalability problem\ncaused by the huge matrix to describe the multi-layer structure and a\nback-projection problem caused by the continuous relaxation of the quadratic\nassignment problem. In this work, we propose a novel multi-attributed graph\nmatching algorithm based on the multi-layer graph factorization. We reformulate\nthe problem to be solved with several small matrices that are obtained by\nfactorizing the multi-layer structure. Then, we solve the problem using a\nconvex-concave relaxation procedure for the multi-layer structure. The proposed\nalgorithm exhibits better performance than state-of-the-art algorithms based on\nthe single-layer structure.\n",
"title": "Exploiting Multi-layer Graph Factorization for Multi-attributed Graph Matching"
}
| null | null | null | null | true | null |
5331
| null |
Default
| null | null |
null |
{
"abstract": " \\emph{Secure Search} is the problem of retrieving from a database table (or\nany unsorted array) the records matching specified attributes, as in SQL SELECT\nqueries, but where the database and the query are encrypted. Secure search has\nbeen the leading example for practical applications of Fully Homomorphic\nEncryption (FHE) starting in Gentry's seminal work; however, to the best of our\nknowledge all state-of-the-art secure search algorithms to date are realized by\na polynomial of degree $\\Omega(m)$ for $m$ the number of records, which is\ntypically too slow in practice even for moderate size $m$.\nIn this work we present the first algorithm for secure search that is\nrealized by a polynomial of degree polynomial in $\\log m$. We implemented our\nalgorithm in an open source library based on HELib implementation for the\nBrakerski-Gentry-Vaikuntanthan's FHE scheme, and ran experiments on Amazon's\nEC2 cloud. Our experiments show that we can retrieve the first match in a\ndatabase of millions of entries in less than an hour using a single machine;\nthe time reduced almost linearly with the number of machines.\nOur result utilizes a new paradigm of employing coresets and sketches, which\nare modern data summarization techniques common in computational geometry and\nmachine learning, for efficiency enhancement for homomorphic encryption. As a\ncentral tool we design a novel sketch that returns the first positive entry in\na (not necessarily sparse) array; this sketch may be of independent interest.\n",
"title": "Secure Search on the Cloud via Coresets and Sketches"
}
| null | null | null | null | true | null |
5332
| null |
Default
| null | null |
null |
{
"abstract": " The Large Array Telescope for Tracking Energetic Sources (LATTES), is a novel\nconcept for an array of hybrid EAS array detectors, composed of a Resistive\nPlate Counter array coupled to a Water Cherenkov Detector, planned to cover\ngamma rays from less than 100 GeV up to 100 TeVs. This experiment, to be\ninstalled at high altitude in South America, could cover the existing gap in\nsensitivity between satellite and ground arrays.\nThe low energy threshold, large duty cycle and wide field of view of LATTES\nmakes it a powerful tool to detect transient phenomena and perform long term\nobservations of variable sources. Moreover, given its characteristics, it would\nbe fully complementary to the planned Cherenkov Telescope Array (CTA) as it\nwould be able to issue alerts.\nIn this talk, a description of its main features and capabilities, as well as\nresults on its expected performance, and sensitivity, will be presented.\n",
"title": "LATTES: a novel detector concept for a gamma-ray experiment in the Southern hemisphere"
}
| null | null |
[
"Physics"
] | null | true | null |
5333
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we propose a general model for plane-based clustering. The\ngeneral model contains many existing plane-based clustering methods, e.g.,\nk-plane clustering (kPC), proximal plane clustering (PPC), twin support vector\nclustering (TWSVC) and its extensions. Under this general model, one may obtain\nan appropriate clustering method for specific purpose. The general model is a\nprocedure corresponding to an optimization problem, where the optimization\nproblem minimizes the total loss of the samples. Thereinto, the loss of a\nsample derives from both within-cluster and between-cluster. In theory, the\ntermination conditions are discussed, and we prove that the general model\nterminates in a finite number of steps at a local or weak local optimal point.\nFurthermore, based on this general model, we propose a plane-based clustering\nmethod by introducing a new loss function to capture the data distribution\nprecisely. Experimental results on artificial and public available datasets\nverify the effectiveness of the proposed method.\n",
"title": "A general model for plane-based clustering with loss function"
}
| null | null | null | null | true | null |
5334
| null |
Default
| null | null |
null |
{
"abstract": " Doped free carriers can substantially renormalize electronic self-energy and\nquasiparticle band gaps of two-dimensional (2D) materials. However, it is still\nchallenging to quantitatively calculate this many-electron effect, particularly\nat the low doping density that is most relevant to realistic experiments and\ndevices. Here we develop a first-principles-based effective-mass model within\nthe GW approximation and show a dramatic band gap renormalization of a few\nhundred meV for typical 2D semiconductors. Moreover, we reveal the roles of\ndifferent many-electron interactions: The Coulomb-hole contribution is dominant\nfor low doping densities while the screened-exchange contribution is dominant\nfor high doping densities. Three prototypical 2D materials are studied by this\nmethod, h-BN, MoS2, and black phosphorus, covering insulators to\nsemiconductors. Especially, anisotropic black phosphorus exhibits a\nsurprisingly large band gap renormalization because of its smaller\ndensity-of-state that enhances the screened-exchange interactions. Our work\ndemonstrates an efficient way to accurately calculate band gap renormalization\nand provides quantitative understanding of doping-dependent many-electron\nphysics of general 2D semiconductors.\n",
"title": "Renormalization of quasiparticle band gap in doped two-dimensional materials from many-body calculations"
}
| null | null | null | null | true | null |
5335
| null |
Default
| null | null |
null |
{
"abstract": " Ghys and Sergiescu proved in the $80$s that Thompson's group $T$, and hence\n$F$, admits actions by $C^{\\infty}$ diffeomorphisms of the circle . They proved\nthat the standard actions of these groups are topologically conjugate to a\ngroup of $C^\\infty$ diffeomorphisms. Monod defined a family of groups of\npiecewise projective homeomorphisms, and Lodha-Moore defined finitely\npresentable groups of piecewise projective homeomorphisms. These groups are of\nparticular interest because they are nonamenable and contain no free subgroup.\nIn contrast to the result of Ghys-Sergiescu, we prove that the groups of Monod\nand Lodha-Moore are not topologically conjugate to a group of $C^1$\ndiffeomorphisms.\nFurthermore, we show that the group of Lodha-Moore has no nonabelian $C^1$\naction on the interval. We also show that many Monod's groups $H(A)$, for\ninstance when $A$ is such that $\\mathsf{PSL}(2,A)$ contains a rational\nhomothety $x\\mapsto \\tfrac{p}{q}x$, do not admit a $C^1$ action on the\ninterval. The obstruction comes from the existence of hyperbolic fixed points\nfor $C^1$ actions. With slightly different techniques, we also show that some\ngroups of piecewise affine homeomorphisms of the interval or the circle are not\nsmoothable.\n",
"title": "Hyperbolicity as an obstruction to smoothability for one-dimensional actions"
}
| null | null | null | null | true | null |
5336
| null |
Default
| null | null |
null |
{
"abstract": " Consider the problem of estimating the entries of an unknown mean matrix or\ntensor given a single noisy realization. In the matrix case, this problem can\nbe addressed by decomposing the mean matrix into a component that is additive\nin the rows and columns, i.e.\\ the additive ANOVA decomposition of the mean\nmatrix, plus a matrix of elementwise effects, and assuming that the elementwise\neffects may be sparse. Accordingly, the mean matrix can be estimated by solving\na penalized regression problem, applying a lasso penalty to the elementwise\neffects. Although solving this penalized regression problem is straightforward,\nspecifying appropriate values of the penalty parameters is not. Leveraging the\nposterior mode interpretation of the penalized regression problem, moment-based\nempirical Bayes estimators of the penalty parameters can be defined. Estimation\nof the mean matrix using these these moment-based empirical Bayes estimators\ncan be called LANOVA penalization, and the corresponding estimate of the mean\nmatrix can be called the LANOVA estimate. The empirical Bayes estimators are\nshown to be consistent. Additionally, LANOVA penalization is extended to\naccommodate sparsity of row and column effects and to estimate an unknown mean\ntensor. The behavior of the LANOVA estimate is examined under misspecification\nof the distribution of the elementwise effects, and LANOVA penalization is\napplied to several datasets, including a matrix of microarray data, a three-way\ntensor of fMRI data and a three-way tensor of wheat infection data.\n",
"title": "Lasso ANOVA Decompositions for Matrix and Tensor Data"
}
| null | null |
[
"Statistics"
] | null | true | null |
5337
| null |
Validated
| null | null |
null |
{
"abstract": " This short article presents a summary of the NetSciEd (Network Science and\nEducation) initiative that aims to address the need for curricula, resources,\naccessible materials, and tools for introducing K-12 students and the general\npublic to the concept of networks, a crucial framework in understanding\ncomplexity. NetSciEd activities include (1) the NetSci High educational\noutreach program (since 2010), which connects high school students and their\nteachers with regional university research labs and provides them with the\nopportunity to work on network science research projects; (2) the NetSciEd\nsymposium series (since 2012), which brings network science researchers and\neducators together to discuss how network science can help and be integrated\ninto formal and informal education; and (3) the Network Literacy: Essential\nConcepts and Core Ideas booklet (since 2014), which was created collaboratively\nand subsequently translated into 18 languages by an extensive group of network\nscience researchers and educators worldwide.\n",
"title": "NetSciEd: Network Science and Education for the Interconnected World"
}
| null | null | null | null | true | null |
5338
| null |
Default
| null | null |
null |
{
"abstract": " A version of Gromov's cup product lemma in which one factor is the (1,0)-part\nof the differential of a continuous plurisubharmonic function is obtained. As\nan application, it is shown that a connected noncompact complete Kaehler\nmanifold that has exactly one end and admits a continuous plurisubharmonic\nfunction that is strictly plurisubharmonic along some germ of a 2-dimensional\ncomplex analytic set at some point has the Bochner-Hartogs property; that is,\nthe first compactly supported cohomology with values in the structure sheaf\nvanishes.\n",
"title": "A cup product lemma for continuous plurisubharmonic functions"
}
| null | null | null | null | true | null |
5339
| null |
Default
| null | null |
null |
{
"abstract": " Density functional theory and nonequilibrium Green's function calculations\nhave been used to explore spin-resolved transport through the high-spin state\nof an iron(II)sulfur single molecular magnet. Our results show that this\nmolecule exhibits near-perfect spin filtering, where the spin-filtering\nefficiency is above 99%, as well as significant negative differential\nresistance centered at a low bias voltage. The rise in the spin-up conductivity\nup to the bias voltage of 0.4 V is dominated by a conductive lowest unoccupied\nmolecular orbital, and this is accompanied by a slight increase in the magnetic\nmoment of the Fe atom. The subsequent drop in the spin-up conductivity is\nbecause the conductive channel moves to the highest occupied molecular orbital\nwhich has a lower conductance contribution. This is accompanied by a drop in\nthe magnetic moment of the Fe atom. These two exceptional properties, and the\nfact that the onset of negative differential resistance occurs at low bias\nvoltage, suggests the potential of the molecule in nanoelectronic and\nnanospintronic applications.\n",
"title": "Near-perfect spin filtering and negative differential resistance in an Fe(II)S complex"
}
| null | null | null | null | true | null |
5340
| null |
Default
| null | null |
null |
{
"abstract": " We present the first search for dark matter-induced delayed coincidence\nsignals in a dual-phase xenon time projection chamber, using the 224.6 live\ndays of the XENON100 science run II. This very distinct signature is predicted\nin the framework of magnetic inelastic dark matter which has been proposed to\nreconcile the modulation signal reported by the DAMA/LIBRA collaboration with\nthe null results from other direct detection experiments. No candidate event\nhas been found in the region of interest and upper limits on the WIMP's\nmagnetic dipole moment are derived. The scenarios proposed to explain the\nDAMA/LIBRA modulation signal by magnetic inelastic dark matter interactions of\nWIMPs with masses of 58.0 GeV/c$^2$ and 122.7 GeV/c$^2$ are excluded at 3.3\n$\\sigma$ and 9.3 $\\sigma$, respectively.\n",
"title": "Search for magnetic inelastic dark matter with XENON100"
}
| null | null | null | null | true | null |
5341
| null |
Default
| null | null |
null |
{
"abstract": " Theoretical investigation of structural, elastic, electronic and bonding\nproperties of A-15 Nb-based intermetallic compounds Nb3B (B = Pt, Os) have been\nperformed using first principles calculations based on the density functional\ntheory (DFT). Optimized cell parameters are found to be in good agreement with\navailable experimental and theoretical results. The elastic constants at zero\npressure and temperature are calculated and the anisotropic behaviors of the\ncompounds are studied. Both the compounds are mechanically stable and ductile\nin nature. Other elastic properties such as Pugh's ratio, Cauchy pressure,\nmachinability index are derived for the first time. Nb3Os is expected to have\ngood lubricating properties compared to Nb3Pt. The electronic band structure\nand energy density of states (DOS) have been studied with and without\nspin-orbit coupling (SOC). The band structures of both the compounds are spin\nsymmetric. Electronic band structure and DOS reveal that both the compounds are\nmetallic and the conductivity mainly arise from the Nb 4d states. The Fermi\nsurface features have been studied for the first time. The Fermi surfaces of\nNb3B contain both hole- and electron-like sheets which change as one replaces\nPt with Os. The electronic charge density distribution shows that Nb3Pt and\nNb3Os both have a mixture of ionic and covalent bonding. The charge transfer\nbetween atomic species in these compounds has been explained by the Mulliken\nbond population analysis.\n",
"title": "Structural, elastic, electronic, and bonding properties of intermetallic Nb3Pt and Nb3Os compounds: a DFT study"
}
| null | null | null | null | true | null |
5342
| null |
Default
| null | null |
null |
{
"abstract": " In this study, we consider unsupervised clustering of categorical vectors\nthat can be of different size using mixture. We use likelihood maximization to\nestimate the parameters of the underlying mixture model and a penalization\ntechnique to select the number of mixture components. Regardless of the true\ndistribution that generated the data, we show that an explicit penalty, known\nup to a multiplicative constant, leads to a non-asymptotic oracle inequality\nwith the Kullback-Leibler divergence on the two sides of the inequality. This\ntheoretical result is illustrated by a document clustering application. To this\naim a novel robust expectation-maximization algorithm is proposed to estimate\nthe mixture parameters that best represent the different topics. Slope\nheuristics are used to calibrate the penalty and to select a number of\nclusters.\n",
"title": "Clustering and Model Selection via Penalized Likelihood for Different-sized Categorical Data Vectors"
}
| null | null | null | null | true | null |
5343
| null |
Default
| null | null |
null |
{
"abstract": " The topology of any complex system is key to understanding its structure and\nfunction. Fundamentally, algebraic topology guarantees that any system\nrepresented by a network can be understood through its closed paths. The length\nof each path provides a notion of scale, which is vitally important in\ncharacterizing dominant modes of system behavior. Here, by combining topology\nwith scale, we prove the existence of universal features which reveal the\ndominant scales of any network. We use these features to compare several\ncanonical network types in the context of a social media discussion which\nevolves through the sharing of rumors, leaks and other news. Our analysis\nenables for the first time a universal understanding of the balance between\nloops and tree-like structure across network scales, and an assessment of how\nthis balance interacts with the spreading of information online. Crucially, our\nresults allow networks to be quantified and compared in a purely model-free way\nthat is theoretically sound, fully automated, and inherently scalable.\n",
"title": "Topology reveals universal features for network comparison"
}
| null | null | null | null | true | null |
5344
| null |
Default
| null | null |
null |
{
"abstract": " Recurrent Neural Networks (RNNs) with sophisticated units that implement a\ngating mechanism have emerged as powerful technique for modeling sequential\nsignals such as speech or electroencephalography (EEG). The latter is the focus\non this paper. A significant big data resource, known as the TUH EEG Corpus\n(TUEEG), has recently become available for EEG research, creating a unique\nopportunity to evaluate these recurrent units on the task of seizure detection.\nIn this study, we compare two types of recurrent units: long short-term memory\nunits (LSTM) and gated recurrent units (GRU). These are evaluated using a state\nof the art hybrid architecture that integrates Convolutional Neural Networks\n(CNNs) with RNNs. We also investigate a variety of initialization methods and\nshow that initialization is crucial since poorly initialized networks cannot be\ntrained. Furthermore, we explore regularization of these convolutional gated\nrecurrent networks to address the problem of overfitting. Our experiments\nrevealed that convolutional LSTM networks can achieve significantly better\nperformance than convolutional GRU networks. The convolutional LSTM\narchitecture with proper initialization and regularization delivers 30%\nsensitivity at 6 false alarms per 24 hours.\n",
"title": "Gated Recurrent Networks for Seizure Detection"
}
| null | null |
[
"Statistics"
] | null | true | null |
5345
| null |
Validated
| null | null |
null |
{
"abstract": " We investigate a projection free method, namely conditional gradient sliding\non batched, stochastic and finite-sum non-convex problem. CGS is a smart\ncombination of Nesterov's accelerated gradient method and Frank-Wolfe (FW)\nmethod, and outperforms FW in the convex setting by saving gradient\ncomputations. However, the study of CGS in the non-convex setting is limited.\nIn this paper, we propose the non-convex conditional gradient sliding (NCGS)\nwhich surpasses the non-convex Frank-Wolfe method in batched, stochastic and\nfinite-sum setting.\n",
"title": "Non-convex Conditional Gradient Sliding"
}
| null | null | null | null | true | null |
5346
| null |
Default
| null | null |
null |
{
"abstract": " For a pair of positive integers $n,k$ with $n\\geq 2$, in this paper we prove\nthat $$ \\sum_{r=1}^k\\sum_{|\\bf\\alpha|=k}{k\\choose\\bf\\alpha}\n\\zeta(n\\bf\\alpha)=\\zeta(n)^k =\\sum^k_{r=1}\\sum_{|\\bf\\alpha|=k}\n{k\\choose\\bf\\alpha}(-1)^{k-r}\\zeta^\\star(n\\bf\\alpha), $$ where\n$\\bf\\alpha=(\\alpha_1,\\alpha_2,\\ldots,\\alpha_r)$ is a $r$-tuple of positive\nintegers. Moreover, we give an application to combinatorics and get the\nfollowing identity: $$ \\sum^{2k}_{r=1}r!{2k\\brace\nr}=\\sum^k_{p=1}\\sum^k_{q=1}{k\\brace p}{k\\brace q} p!q!D(p,q), $$ where\n${k\\brace p}$ is the Stirling numbers of the second kind and $D(p,q)$ is the\nDelannoy number.\n",
"title": "Multinomial Sum Formulas of Multiple Zeta Values"
}
| null | null | null | null | true | null |
5347
| null |
Default
| null | null |
null |
{
"abstract": " We introduce a new operation, copolar addition, on unbounded convex subsets\nof the positive orthant of real euclidean space and establish convexity of the\ncovolumes of the corresponding convex combinations. The proof is based on a\ntechnique of geodesics of plurisubharmonic functions. As an application, we\nshow that there are no relative extremal functions inside a non-constant\ngeodesic curve between two toric relative extremal functions.\n",
"title": "Copolar convexity"
}
| null | null |
[
"Mathematics"
] | null | true | null |
5348
| null |
Validated
| null | null |
null |
{
"abstract": " Concurrent separation logics have helped to significantly simplify\ncorrectness proofs for concurrent data structures. However, a recurring problem\nin such proofs is that data structure abstractions that work well in the\nsequential setting are much harder to reason about in a concurrent setting due\nto complex sharing and overlays. To solve this problem, we propose a novel\napproach to abstracting regions in the heap by encoding the data structure\ninvariant into a local condition on each individual node. This condition may\ndepend on a quantity associated with the node that is computed as a fixpoint\nover the entire heap graph. We refer to this quantity as a flow. Flows can\nencode both structural properties of the heap (e.g. the reachable nodes from\nthe root form a tree) as well as data invariants (e.g. sortedness). We then\nintroduce the notion of a flow interface, which expresses the relies and\nguarantees that a heap region imposes on its context to maintain the local flow\ninvariant with respect to the global heap. Our main technical result is that\nthis notion leads to a new semantic model of separation logic. In this model,\nflow interfaces provide a general abstraction mechanism for describing complex\ndata structures. This abstraction mechanism admits proof rules that generalize\nover a wide variety of data structures. To demonstrate the versatility of our\napproach, we show how to extend the logic RGSep with flow interfaces. We have\nused this new logic to prove linearizability and memory safety of nontrivial\nconcurrent data structures. In particular, we obtain parametric linearizability\nproofs for concurrent dictionary algorithms that abstract from the details of\nthe underlying data structure representation. These proofs cannot be easily\nexpressed using the abstraction mechanisms provided by existing separation\nlogics.\n",
"title": "Go with the Flow: Compositional Abstractions for Concurrent Data Structures (Extended Version)"
}
| null | null | null | null | true | null |
5349
| null |
Default
| null | null |
null |
{
"abstract": " Ancient solutions arise in the study of parabolic blow-ups. If we can\ncategorize ancient solutions, we can better understand blow-up limits. Based on\nan argument of Giga and Kohn, we give a Liouville-type theorem restricting\nancient, type-I, non-collapsing two- dimensional mean curvature flows to either\nspheres or cylinders.\n",
"title": "A Liouville Theorem for Mean Curvature Flow"
}
| null | null |
[
"Mathematics"
] | null | true | null |
5350
| null |
Validated
| null | null |
null |
{
"abstract": " Solvothermal intercalation of ethylenediamine molecules into FeSe separates\nthe layers by 1078 pm and creates a different stacking. FeSe(en)0.3 is not\nsuperconducting although each layer exhibits the stripe-type crystal structure\nand the Fermi surface topology of superconducting FeSe. FeSe(en)0.3 requires\nelectron-doping for high-Tc similar to monolayers of FeSe@SrTiO3, whose much\nhigher Tc may arise from the proximity of the oxide surface.\n",
"title": "FeSe(en)0.3 - Separated FeSe layers with stripe-type crystal structure by intercalation of neutral spacer molecules"
}
| null | null |
[
"Physics"
] | null | true | null |
5351
| null |
Validated
| null | null |
null |
{
"abstract": " Tangles of quantized vortex line of initial density ${\\cal L}(0) \\sim 6\\times\n10^3$\\,cm$^{-2}$ and variable amplitude of fluctuations of flow velocity $U(0)$\nat the largest length scale were generated in superfluid $^4$He at $T=0.17$\\,K,\nand their free decay ${\\cal L}(t)$ was measured. If $U(0)$ is small, the excess\nrandom component of vortex line length firstly decays as ${\\cal L} \\propto\nt^{-1}$ until it becomes comparable with the structured component responsible\nfor the classical velocity field, and the decay changes to ${\\cal L} \\propto\nt^{-3/2}$. The latter regime always ultimately prevails, provided the classical\ndescription of $U$ holds. A quantitative model of coexisting cascades of\nquantum and classical energies describes all regimes of the decay.\n",
"title": "Coexistence of quantum and classical flows in quantum turbulence in the $T=0$ limit"
}
| null | null |
[
"Physics"
] | null | true | null |
5352
| null |
Validated
| null | null |
null |
{
"abstract": " We study the supersymmetric partition function on $S^1 \\times L(r, 1)$, or\nthe lens space index of four-dimensional $\\mathcal{N}=2$ superconformal field\ntheories and their connection to two-dimensional chiral algebras. We primarily\nfocus on free theories as well as Argyres-Douglas theories of type $(A_1, A_k)$\nand $(A_1, D_k)$. We observe that in specific limits, the lens space index is\nreproduced in terms of the (refined) character of an appropriately twisted\nmodule of the associated two-dimensional chiral algebra or a generalized vertex\noperator algebra. The particular twisted module is determined by the choice of\ndiscrete holonomies for the flavor symmetry in four-dimensions.\n",
"title": "Four-dimensional Lens Space Index from Two-dimensional Chiral Algebra"
}
| null | null | null | null | true | null |
5353
| null |
Default
| null | null |
null |
{
"abstract": " Let $ \\Omega$ be a bounded Lipschitz domain of $ \\mathbb{R}^{d}.$ The purpose\nof this paper is to establish Lions' formula for reproducing kernel Hilbert\nspaces $\\mathcal H^s(\\Omega)$ of real harmonic functions elements of the usual\nSobolev space $H^s(\\Omega)$ for $s\\geq 0.$ To this end, we provide a functional\ncharacterization of $\\mathcal H^s(\\Omega)$ via some new families of positive\nself-adjoint operators, describe their trace data and discuss the values of $s$\nfor which they are RKHSs. Also a construction of an orthonormal basis of\n$\\mathcal H^s(\\Omega)$ is established.\n",
"title": "Lions' formula for RKHSs of real harmonic functions on Lipschitz domains"
}
| null | null |
[
"Mathematics"
] | null | true | null |
5354
| null |
Validated
| null | null |
null |
{
"abstract": " With the rapidly growing interest in bifacial photovoltaics (PV), a worldwide\nmap of their potential performance can help assess and accelerate the global\ndeployment of this emerging technology. However, the existing literature only\nhighlights optimized bifacial PV for a few geographic locations or develops\nworldwide performance maps for very specific configurations, such as the\nvertical installation. It is still difficult to translate these location- and\nconfiguration-specific conclusions to a general optimized performance of this\ntechnology. In this paper, we present a global study and optimization of\nbifacial solar modules using a rigorous and comprehensive modeling framework.\nOur results demonstrate that with a low albedo of 0.25, the bifacial gain of\nground-mounted bifacial modules is less than 10% worldwide. However, increasing\nthe albedo to 0.5 and elevating modules 1 m above the ground can boost the\nbifacial gain to 30%. Moreover, we derive a set of empirical design rules,\nwhich optimize bifacial solar modules across the world, that provide the\ngroundwork for rapid assessment of the location-specific performance. We find\nthat ground-mounted, vertical, east-west-facing bifacial modules will\noutperform their south-north-facing, optimally tilted counterparts by up to 15%\nbelow the latitude of 30 degrees, for an albedo of 0.5. The relative energy\noutput is the reverse of this in latitudes above 30 degrees. A detailed and\nsystematic comparison with experimental data from Asia, Europe, and North\nAmerica validates the model presented in this paper. An online simulation tool\n(this https URL) based on the model developed in this paper is\nalso available for a user to predict and optimize bifacial modules in any\narbitrary location across the globe.\n",
"title": "Optimization and Performance of Bifacial Solar Modules: A Global Perspective"
}
| null | null | null | null | true | null |
5355
| null |
Default
| null | null |
null |
{
"abstract": " Simulation of wave propagation in a microearthquake environment is often\nchallenging due to small-scale structural and material heterogeneities. We\nsimulate wave propagation in three different real microearthquake environments\nusing a spectral-element method. In the first example, we compute the full\nwavefield in 2D and 3D models of an underground ore mine, namely the Pyhaesalmi\nmine in Finland. In the second example, we simulate wave propagation in a\nhomogeneous velocity model including the actual topography of an unstable rock\nslope at Aaknes in western Norway. Finally, we compute the full wavefield for a\nweakly anisotropic cylindrical sample at laboratory scale, which was used for\nan acoustic emission experiment under triaxial loading. We investigate the\ncharacteristic features of wave propagation in those models and compare\nsynthetic waveforms with observed waveforms wherever possible. We illustrate\nthe challenges associated with the spectral-element simulation in those models.\n",
"title": "Wave propagation modelling in various microearthquake environments using a spectral-element method"
}
| null | null | null | null | true | null |
5356
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a new concurrent heap algorithm, based on a stateless\nshape property, which efficiently maintains balance during insert and removeMin\noperations implemented with hand-over-hand locking. It also provides a O(1)\nlinearizable snapshot operation based on lazy copy-on-write semantics. Such\nsnapshots can be used to provide consistent views of the heap during iteration,\nas well as to make speculative updates (which can later be dropped).\nThe simplicity of the algorithm allows it to be easily proven correct, and\nthe choice of shape property provides priority queue performance which is\ncompetitive with highly optimized skiplist implementations (and has stronger\nbounds on worst-case time complexity).\nA Scala reference implementation is provided.\n",
"title": "Fast Snapshottable Concurrent Braun Heaps"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5357
| null |
Validated
| null | null |
null |
{
"abstract": " This article presents GuideR, a user-guided rule induction algorithm, which\novercomes the largest limitation of the existing methods-the lack of the\npossibility to introduce user's preferences or domain knowledge to the rule\nlearning process. Automatic selection of attributes and attribute ranges often\nleads to the situation in which resulting rules do not contain interesting\ninformation. We propose an induction algorithm which takes into account user's\nrequirements. Our method uses the sequential covering approach and is suitable\nfor classification, regression, and survival analysis problems. The\neffectiveness of the algorithm in all these tasks has been verified\nexperimentally, confirming guided rule induction to be a powerful data analysis\ntool.\n",
"title": "GuideR: a guided separate-and-conquer rule learning in classification, regression, and survival settings"
}
| null | null | null | null | true | null |
5358
| null |
Default
| null | null |
null |
{
"abstract": " Over short time intervals planetary ephemerides have been traditionally\nrepresented in analytical form as finite sums of periodic terms or sums of\nPoisson terms that are periodic terms with polynomial amplitudes. Nevertheless,\nthis representation is not well adapted for the evolution of the planetary\norbits in the solar system over million of years as they present drifts in\ntheir main frequencies, due to the chaotic nature of their dynamics. The aim of\nthe present paper is to develop a numerical algorithm for slowly diffusing\nsolutions of a perturbed integrable Hamiltonian system that will apply to the\nrepresentation of the chaotic planetary motions with varying frequencies. By\nsimple analytical considerations, we first argue that it is possible to recover\nexactly a single varying frequency. Then, a function basis involving\ntime-dependent fundamental frequencies is formulated in a semi-analytical way.\nFinally, starting from a numerical solution, a recursive algorithm is used to\nnumerically decompose the solution on the significant elements of the function\nbasis. Simple examples show that this algorithm can be used to give compact\nrepresentations of different types of slowly diffusing solutions. As a test\nexample, we show how this algorithm can be successfully applied to obtain a\nvery compact approximation of the La2004 solution of the orbital motion of the\nEarth over 40 Myr ([-35Myr,5Myr]). This example has been chosen as this\nsolution is widely used for the reconstruction of the climates of the past.\n",
"title": "Frequency analysis and the representation of slowly diffusing planetary solutions"
}
| null | null |
[
"Physics"
] | null | true | null |
5359
| null |
Validated
| null | null |
null |
{
"abstract": " Given two sets of points $A$ and $B$ in a normed plane, we prove that there\nare two linearly separable sets $A'$ and $B'$ such that $\\mathrm{diam}(A')\\leq\n\\mathrm{diam}(A)$, $\\mathrm{diam}(B')\\leq \\mathrm{diam}(B)$, and $A'\\cup\nB'=A\\cup B.$ This extends a result for the Euclidean distance to symmetric\nconvex distance functions. As a consequence, some Euclidean $k$-clustering\nalgorithms are adapted to normed planes, for instance, those that minimize the\nmaximum, the sum, or the sum of squares of the $k$ cluster diameters. The\n2-clustering problem when two different bounds are imposed to the diameters is\nalso solved. The Hershberger-Suri's data structure for managing ball hulls can\nbe useful in this context.\n",
"title": "Geometric clustering in normed planes"
}
| null | null | null | null | true | null |
5360
| null |
Default
| null | null |
null |
{
"abstract": " Rural areas in the developing countries are predominantly devoid of Internet\naccess as it is not viable for operators to provide broadband service in these\nareas. To solve this problem, we propose a middle mile Long erm Evolution\nAdvanced (LTE-A) network operating in TV white space to connect villages to an\noptical Point of Presence (PoP) located in the vicinity of a rural area. We\nstudy the problem of spectrum sharing for the middle mile networks deployed by\nmultiple operators. A graph theory based Fairness Constrained Channel\nAllocation (FCCA) algorithm is proposed, employing Carrier Aggregation (CA) and\nListen Before Talk (LBT) features of LTE-A. We perform extensive system level\nsimulations to demonstrate that FCCA not only increases spectral efficiency but\nalso improves system fairness.\n",
"title": "Spectrum Sharing for LTE-A Network in TV White Space"
}
| null | null | null | null | true | null |
5361
| null |
Default
| null | null |
null |
{
"abstract": " We construct an obstruction for the existence of embeddings of homology\n$3$-sphere into homology $S^3\\times S^1$ under some cohomological condition.\nThe obstruction is defined as an element in the filtered version of the\ninstanton Floer cohomology due to R.Fintushel-R.Stern. We make use of the\n$\\mathbb{Z}$-fold covering space of homology $S^3\\times S^1$ and the instantons\non it.\n",
"title": "Instantons for 4-manifolds with periodic ends and an obstruction to embeddings of 3-manifolds"
}
| null | null | null | null | true | null |
5362
| null |
Default
| null | null |
null |
{
"abstract": " Inspired by river networks and other structures formed by Laplacian growth,\nwe use the Loewner equation to investigate the growth of a network of thin\nfingers in a diffusion field. We first review previous contributions to\nillustrate how this formalism reduces the network's expansion to three rules,\nwhich respectively govern the velocity, the direction, and the nucleation of\nits growing branches. This framework allows us to establish the mathematical\nequivalence between three formulations of the direction rule, namely geodesic\ngrowth, growth that maintains local symmetry and growth that maximizes flux\ninto tips for a given amount of growth. Surprisingly, we find that this growth\nrule may result in a network different from the static configuration that\noptimizes flux into tips.\n",
"title": "Laplacian networks: growth, local symmetry and shape optimization"
}
| null | null |
[
"Physics"
] | null | true | null |
5363
| null |
Validated
| null | null |
null |
{
"abstract": " Multiview representation learning is very popular for latent factor analysis.\nIt naturally arises in many data analysis, machine learning, and information\nretrieval applications to model dependent structures among multiple data\nsources. For computational convenience, existing approaches usually formulate\nthe multiview representation learning as convex optimization problems, where\nglobal optima can be obtained by certain algorithms in polynomial time.\nHowever, many pieces of evidence have corroborated that heuristic nonconvex\napproaches also have good empirical computational performance and convergence\nto the global optima, although there is a lack of theoretical justification.\nSuch a gap between theory and practice motivates us to study a nonconvex\nformulation for multiview representation learning, which can be efficiently\nsolved by a simple stochastic gradient descent (SGD) algorithm. We first\nillustrate the geometry of the nonconvex formulation; Then, we establish\nasymptotic global rates of convergence to the global optima by diffusion\napproximations. Numerical experiments are provided to support our theory.\n",
"title": "Dropping Convexity for More Efficient and Scalable Online Multiview Learning"
}
| null | null | null | null | true | null |
5364
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we studied a SLAM method for vector-based road structure\nmapping using multi-beam LiDAR. We propose to use the polyline as the primary\nmapping element instead of grid cell or point cloud, because the vector-based\nrepresentation is precise and lightweight, and it can directly generate\nvector-based High-Definition (HD) driving map as demanded by autonomous driving\nsystems. We explored: 1) the extraction and vectorization of road structures\nbased on local probabilistic fusion. 2) the efficient vector-based matching\nbetween frames of road structures. 3) the loop closure and optimization based\non the pose-graph. In this study, we took a specific road structure, the road\nboundary, as an example. We applied the proposed matching method in three\ndifferent scenes and achieved the average absolute matching error of 0.07. We\nfurther applied the mapping system to the urban road with the length of 860\nmeters and achieved an average global accuracy of 0.466 m without the help of\nhigh precision GPS.\n",
"title": "Automatic Vector-based Road Structure Mapping Using Multi-beam LiDAR"
}
| null | null | null | null | true | null |
5365
| null |
Default
| null | null |
null |
{
"abstract": " To a complex projective structure $\\Sigma$ on a surface, Thurston associates\na locally convex pleated surface. We derive bounds on the geometry of both in\nterms of the norms $\\|\\phi_\\Sigma\\|_\\infty$ and $\\|\\phi_\\Sigma\\|_2$ of the\nquadratic differential $\\phi_\\Sigma$ of $\\Sigma$ given by the Schwarzian\nderivative of the associated locally univalent map. We show that these give a\nunifying approach that generalizes a number of important, well known results\nfor convex cocompact hyperbolic structures on 3-manifolds, including bounds on\nthe Lipschitz constant for the nearest-point retraction and the length of the\nbending lamination. We then use these bounds to begin a study of the\nWeil-Petersson gradient flow of renormalized volume on the space $CC(N)$ of\nconvex cocompact hyperbolic structures on a compact manifold $N$ with\nincompressible boundary, leading to a proof of the conjecture that the\nrenormalized volume has infimum given by one-half the simplicial volume of\n$DN$, the double of $N$.\n",
"title": "Schwarzian derivatives, projective structures, and the Weil-Petersson gradient flow for renormalized volume"
}
| null | null | null | null | true | null |
5366
| null |
Default
| null | null |
null |
{
"abstract": " This paper is concerned with paraphrase detection. The ability to detect\nsimilar sentences written in natural language is crucial for several\napplications, such as text mining, text summarization, plagiarism detection,\nauthorship authentication and question answering. Given two sentences, the\nobjective is to detect whether they are semantically identical. An important\ninsight from this work is that existing paraphrase systems perform well when\napplied on clean texts, but they do not necessarily deliver good performance\nagainst noisy texts. Challenges with paraphrase detection on user generated\nshort texts, such as Twitter, include language irregularity and noise. To cope\nwith these challenges, we propose a novel deep neural network-based approach\nthat relies on coarse-grained sentence modeling using a convolutional neural\nnetwork and a long short-term memory model, combined with a specific\nfine-grained word-level similarity matching model. Our experimental results\nshow that the proposed approach outperforms existing state-of-the-art\napproaches on user-generated noisy social media data, such as Twitter texts,\nand achieves highly competitive performance on a cleaner corpus.\n",
"title": "A Deep Network Model for Paraphrase Detection in Short Text Messages"
}
| null | null | null | null | true | null |
5367
| null |
Default
| null | null |
null |
{
"abstract": " Lead halide perovskite solar cells have recently emerged as a very promising\nphotovoltaic technology due to their excellent power conversion efficiencies;\nhowever, the toxicity of lead and the poor stability of perovskite materials\nremain two main challenges that need to be addressed. Here, for the first time,\nwe report a lead-free, highly stable C6H4NH2CuBr2I compound. The C6H4NH2CuBr2I\nfilms exhibit extraordinary hydrophobic behavior with a contact angle of\napproximately 90 degree, and their X-ray diffraction patterns remain unchanged\neven after four hours of water immersion. UV-Vis absorption spectrum shows that\nC6H4NH2CuBr2I compound has an excellent optical absorption over the entire\nvisible spectrum. We applied this copper-based light absorber in printable\nmesoscopic solar cell for the initial trial and achieved a power conversion\nefficiency of 0.5%. Our study represents an alternative pathway to develop\nlow-toxic and highly stable organic-inorganic hybrid materials for photovoltaic\napplication.\n",
"title": "Organic-inorganic Copper(II)-based Material: a Low-Toxic, Highly Stable Light Absorber beyond Organolead Perovskites"
}
| null | null | null | null | true | null |
5368
| null |
Default
| null | null |
null |
{
"abstract": " We establish a bijective correspondence between certain non-self-intersecting\ncurves in an $n$-punctured disc and positive ${\\mathbf c}$-vectors of acyclic\ncluster algebras whose quivers have multiple arrows between every pair of\nvertices. As a corollary, we obtain a proof of a conjecture by K.-H. Lee and K.\nLee (arXiv:1703.09113) on the combinatorial description of real Schur roots for\nacyclic quivers with multiple arrows, and give a combinatorial characterization\nof seeds in terms of curves in an $n$-punctured disc.\n",
"title": "Acyclic cluster algebras, reflection groups, and curves on a punctured disc"
}
| null | null | null | null | true | null |
5369
| null |
Default
| null | null |
null |
{
"abstract": " Knowing the structure of an offline social network facilitates a variety of\nanalyses, including studying the rate at which infectious diseases may spread\nand identifying a subset of actors to immunize in order to reduce, as much as\npossible, the rate of spread. Offline social network topologies are typically\nestimated by surveying actors and asking them to list their neighbours. While\nidentifying close friends and family (i.e., strong ties) can typically be done\nreliably, listing all of one's acquaintances (i.e., weak ties) is subject to\nerror due to respondent fatigue. This issue is commonly circumvented through\nthe use of so-called \"fixed choice\" surveys where respondents are asked to name\na fixed, small number of their weak ties (e.g., two or ten). Of course, the\nresulting crude observed network will omit many ties, and using this crude\nnetwork to infer properties of the network, such as its degree distribution or\nclustering coefficient, will lead to biased estimates. This paper develops\nestimators, based on the method of moments, for a number of network\ncharacteristics including those related to the first and second moments of the\ndegree distribution as well as the network size, using fixed-choice survey\ndata. Experiments with simulated data illustrate that the proposed estimators\nperform well across a variety of network topologies and measurement scenarios,\nand the resulting estimates are significantly more accurate than those obtained\ndirectly using the crude observed network, which are commonly used in the\nliterature. We also describe a variation of the Jackknife procedure that can be\nused to obtain an estimate of the estimator variance.\n",
"title": "Inferring Structural Characteristics of Networks with Strong and Weak Ties from Fixed-Choice Surveys"
}
| null | null | null | null | true | null |
5370
| null |
Default
| null | null |
null |
{
"abstract": " This work investigated the detection of gravitational wave (GW) from\nsimulated damped sinusoid signals contaminated with Gaussian noise. We proposed\nto treat it as a classification problem with one class bearing our special\nattentions. Two successive steps of the proposed scheme are as following:\nfirst, decompose the data using a wavelet packet and represent the GW signal\nand noise using the derived decomposition coefficients; Second, detect the\nexistence of GW using a convolutional neural network (CNN). To reflect our\nspecial attention on searching GW signals, the performance is evaluated using\nnot only the traditional classification accuracy (correct ratio), but also\nreceiver operating characteristic (ROC) curve, and experiments show excelllent\nperformances on both evaluation measures. The generalization of a proposed\nsearching scheme on GW model parameter and possible extensions to other data\nanalysis tasks are crucial for a machine learning based approach. On this\naspect, experiments shows that there is no significant difference between GW\nmodel parameters on identification performances by our proposed scheme.\nTherefore, the proposed scheme has excellent generalization and could be used\nto search for non-trained and un-known GW signals or glitches in the future GW\nastronomy era.\n",
"title": "A Method Of Detecting Gravitational Wave Based On Time-frequency Analysis And Convolutional Neural Networks"
}
| null | null |
[
"Physics"
] | null | true | null |
5371
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we demonstrate the connection between a magnetic storage ring\nwith additional sextupole fields set so that the x and y chromaticities vanish\nand the maximizing of the lifetime of in-plane polarization (IPP) for a\n0.97-GeV/c deuteron beam. The IPP magnitude was measured by continuously\nmonitoring the down-up scattering asymmetry (sensitive to sideways\npolarization) in an in-beam, carbon-target polarimeter and unfolding the\nprecession of the IPP due to the magnetic anomaly of the deuteron. The optimum\noperating conditions for a long IPP lifetime were made by scanning the field of\nthe storage ring sextupole magnet families while observing the rate of IPP loss\nduring storage of the beam. The beam was bunched and electron cooled. The IPP\nlosses appear to arise from the change of the orbit circumference, and\nconsequently the particle speed and spin tune, due to the transverse betatron\noscillations of individual particles in the beam. The effects of these changes\nare canceled by an appropriate sextupole field setting.\n",
"title": "The connection between zero chromaticity and long in-plane polarization lifetime in a magnetic storage ring"
}
| null | null | null | null | true | null |
5372
| null |
Default
| null | null |
null |
{
"abstract": " Training automatic speech recognition (ASR) systems requires large amounts of\ndata in the target language in order to achieve good performance. Whereas large\ntraining corpora are readily available for languages like English, there exists\na long tail of languages which do suffer from a lack of resources. One method\nto handle data sparsity is to use data from additional source languages and\nbuild a multilingual system. Recently, ASR systems based on recurrent neural\nnetworks (RNNs) trained with connectionist temporal classification (CTC) have\ngained substantial research interest. In this work, we extended our previous\napproach towards training CTC-based systems multilingually. Our systems feature\na global phone set, based on the joint phone sets of each source language. We\nevaluated the use of different language combinations as well as the addition of\nLanguage Feature Vectors (LFVs). As contrastive experiment, we built systems\nbased on graphemes as well. Systems having a multilingual phone set are known\nto suffer in performance compared to their monolingual counterparts. With our\nproposed approach, we could reduce the gap between these mono- and multilingual\nsetups, using either graphemes or phonemes.\n",
"title": "Phonemic and Graphemic Multilingual CTC Based Speech Recognition"
}
| null | null | null | null | true | null |
5373
| null |
Default
| null | null |
null |
{
"abstract": " Dynamic networks are a general language for describing time-evolving complex\nsystems, and discrete time network models provide an emerging statistical\ntechnique for various applications. It is a fundamental research question to\ndetect the community structure in time-evolving networks. However, due to\nsignificant computational challenges and difficulties in modeling communities\nof time-evolving networks, there is little progress in the current literature\nto effectively find communities in time-evolving networks. In this work, we\npropose a novel model-based clustering framework for time-evolving networks\nbased on discrete time exponential-family random graph models. To choose the\nnumber of communities, we use conditional likelihood to construct an effective\nmodel selection criterion. Furthermore, we propose an efficient variational\nexpectation-maximization (EM) algorithm to find approximate maximum likelihood\nestimates of network parameters and mixing proportions. By using variational\nmethods and minorization-maximization (MM) techniques, our method has appealing\nscalability for large-scale time-evolving networks. The power of our method is\ndemonstrated in simulation studies and empirical applications to international\ntrade networks and the collaboration networks of a large American research\nuniversity.\n",
"title": "Model-Based Clustering of Time-Evolving Networks through Temporal Exponential-Family Random Graph Models"
}
| null | null | null | null | true | null |
5374
| null |
Default
| null | null |
null |
{
"abstract": " Many robotic applications, such as search-and-rescue, require multiple agents\nto search for and perform actions on targets. However, such missions present\nseveral challenges, including cooperative exploration, task selection and\nallocation, time limitations, and computational complexity. To address this, we\npropose a decentralized multi-agent decision-making framework for the search\nand action problem with time constraints. The main idea is to treat time as an\nallocated budget in a setting where each agent action incurs a time cost and\nyields a certain reward. Our approach leverages probabilistic reasoning to make\nnear-optimal decisions leading to maximized reward. We evaluate our method in\nthe search, pick, and place scenario of the Mohamed Bin Zayed International\nRobotics Challenge (MBZIRC), by using a probability density map and reward\nprediction function to assess actions. Extensive simulations show that our\nalgorithm outperforms benchmark strategies, and we demonstrate system\nintegration in a Gazebo-based environment, validating the framework's readiness\nfor field application.\n",
"title": "Multi-agent Time-based Decision-making for the Search and Action Problem"
}
| null | null | null | null | true | null |
5375
| null |
Default
| null | null |
null |
{
"abstract": " The missing phase problem in X-ray crystallography is commonly solved using\nthe technique of molecular replacement, which borrows phases from a previously\nsolved homologous structure, and appends them to the measured Fourier\nmagnitudes of the diffraction patterns of the unknown structure. More recently,\nmolecular replacement has been proposed for solving the missing orthogonal\nmatrices problem arising in Kam's autocorrelation analysis for single particle\nreconstruction using X-ray free electron lasers and cryo-EM. In classical\nmolecular replacement, it is common to estimate the magnitudes of the unknown\nstructure as twice the measured magnitudes minus the magnitudes of the\nhomologous structure, a procedure known as `twicing'. Mathematically, this is\nequivalent to finding an unbiased estimator for a complex-valued scalar. We\ngeneralize this scheme for the case of estimating real or complex valued\nmatrices arising in single particle autocorrelation analysis. We name this\napproach \"Anisotropic Twicing\" because unlike the scalar case, the unbiased\nestimator is not obtained by a simple magnitude isotropic correction. We\ncompare the performance of the least squares, twicing and anisotropic twicing\nestimators on synthetic and experimental datasets. We demonstrate 3D homology\nmodeling in cryo-EM directly from experimental data without iterative\nrefinement or class averaging, for the first time.\n",
"title": "Anisotropic twicing for single particle reconstruction using autocorrelation analysis"
}
| null | null | null | null | true | null |
5376
| null |
Default
| null | null |
null |
{
"abstract": " While a variety of fundamental differences are known to separate\ntwo-dimensional (2D) and three-dimensional (3D) fluid flows, it is not well\nunderstood how they are related. Conventionally, dimensional reduction is\njustified by an \\emph{a priori} geometrical framework; i.e., 2D flows occur\nunder some geometrical constraint such as shallowness. However, deeper inquiry\ninto 3D flow often finds the presence of local 2D-like structures without such\na constraint, where 2D-like behavior may be identified by the integrability of\nvortex lines or vanishing local helicity. Here we propose a new paradigm of\nflow structure by introducing an intermediate class, termed epi-2-dimensional\nflow, and thereby build a topological bridge between 2D and 3D flows. The\nepi-2D property is local, and is preserved in fluid elements obeying ideal\n(inviscid and barotropic) mechanics; a local epi-2D flow may be regarded as a\n`particle' carrying a generalized enstrophy as its charge. A finite viscosity\nmay cause `fusion' of two epi-2D particles, generating helicity from their\ncharges giving rise to 3D flow.\n",
"title": "Epi-two-dimensional fluid flow: a new topological paradigm for dimensionality"
}
| null | null | null | null | true | null |
5377
| null |
Default
| null | null |
null |
{
"abstract": " The critical behavior of the random field $O(N)$ model driven at a uniform\nvelocity is investigated at zero-temperature. From naive phenomenological\narguments, we introduce a dimensional reduction property, which relates the\nlarge-scale behavior of the $D$-dimensional driven random field $O(N)$ model to\nthat of the $(D-1)$-dimensional pure $O(N)$ model. This is an analogue of the\ndimensional reduction property in equilibrium cases, which states that the\nlarge-scale behavior of $D$-dimensional random field models is identical to\nthat of $(D-2)$-dimensional pure models. However, the dimensional reduction\nproperty breaks down in low enough dimensions due to the presence of multiple\nmeta-stable states. By employing the non-perturbative renormalization group\napproach, we calculate the critical exponents of the driven random field $O(N)$\nmodel near three-dimensions and determine the range of $N$ in which the\ndimensional reduction breaks down.\n",
"title": "Dimensional reduction and its breakdown in the driven random field O(N) model"
}
| null | null | null | null | true | null |
5378
| null |
Default
| null | null |
null |
{
"abstract": " Four types of explicit estimators are proposed here to estimate the loss\nrates of the links in a network with the tree topology and all of them are\nderived by the maximum likelihood principle. One of the four is developed from\nan estimator that was used but neglected because it was suspected to have a\nhigher variance. All of the estimators are proved to be either unbiased or\nasymptotic unbiased. In addition, a set of formulae are derived to compute the\nefficiencies and variances of the estimates obtained by the estimators. One of\nthe formulae shows that if a path is divided into two segments, the variance of\nthe estimates obtained for the pass rate of a segment is equal to the variance\nof the pass rate of the path divided by the square of the pass rate of the\nother segment. A number of theorems and corollaries are derived from the\nformulae that can be used to evaluate the performance of an estimator. Using\nthe theorems and corollaries, we find the estimators from the neglected one are\nthe best estimator for the networks with the tree topology in terms of\nefficiency and computation complexity.\n",
"title": "Statistical Properties of Loss Rate Estimators in Tree Topology (2)"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5379
| null |
Validated
| null | null |
null |
{
"abstract": " We study the multiparty communication complexity of high dimensional\npermutations, in the Number On the Forehead (NOF) model. This model is due to\nChandra, Furst and Lipton (CFL) who also gave a nontrivial protocol for the\nExactly-n problem where three players receive integer inputs and need to decide\nif their inputs sum to a given integer $n$. There is a considerable body of\nliterature dealing with the same problem, where $(\\mathbb{N},+)$ is replaced by\nsome other abelian group. Our work can be viewed as a far-reaching extension of\nthis line of work.\nWe show that the known lower bounds for that group-theoretic problem apply to\nall high dimensional permutations. We introduce new proof techniques that\nappeal to recent advances in Additive Combinatorics and Ramsey theory. We\nreveal new and unexpected connections between the NOF communication complexity\nof high dimensional permutations and a variety of well known and thoroughly\nstudied problems in combinatorics.\nPrevious protocols for Exactly-n all rely on the construction of large sets\nof integers without a 3-term arithmetic progression. No direct algorithmic\nprotocol was previously known for the problem, and we provide the first such\nalgorithm. This suggests new ways to significantly improve the CFL protocol.\nMany new open questions are presented throughout.\n",
"title": "On The Communication Complexity of High-Dimensional Permutations"
}
| null | null | null | null | true | null |
5380
| null |
Default
| null | null |
null |
{
"abstract": " Many engineers wish to deploy modern neural networks in memory-limited\nsettings; but the development of flexible methods for reducing memory use is in\nits infancy, and there is little knowledge of the resulting cost-benefit. We\npropose structural model distillation for memory reduction using a strategy\nthat produces a student architecture that is a simple transformation of the\nteacher architecture: no redesign is needed, and the same hyperparameters can\nbe used. Using attention transfer, we provide Pareto curves/tables for\ndistillation of residual networks with four benchmark datasets, indicating the\nmemory versus accuracy payoff. We show that substantial memory savings are\npossible with very little loss of accuracy, and confirm that distillation\nprovides student network performance that is better than training that student\narchitecture directly on data.\n",
"title": "Moonshine: Distilling with Cheap Convolutions"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
5381
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, a new wiretap channel model is proposed, where the legitimate\ntransmitter and receiver communicate over a discrete memoryless channel. The\nwiretapper has perfect access to a fixed-length subset of the transmitted\ncodeword symbols of her choosing. Additionally, she observes the remainder of\nthe transmitted symbols through a discrete memoryless channel. This new model\nsubsumes the classical wiretap channel and wiretap channel II with noisy main\nchannel as its special cases. The strong secrecy capacity of the proposed\nchannel model is identified. Achievability is established by solving a dual\nsecret key agreement problem in the source model, and converting the solution\nto the original channel model using probability distribution approximation\narguments. In the dual problem, a source encoder and decoder, who observe\nrandom sequences independent and identically distributed according to the input\nand output distributions of the legitimate channel in the original problem,\ncommunicate a confidential key over a public error-free channel using a single\nforward transmission, in the presence of a compound wiretapping source who has\nperfect access to the public discussion. The security of the key is guaranteed\nfor the exponentially many possibilities of the subset chosen at wiretapper by\nderiving a lemma which provides a doubly-exponential convergence rate for the\nprobability that, for a fixed choice of the subset, the key is uniform and\nindependent from the public discussion and the wiretapping source's\nobservation. The converse is derived by using Sanov's theorem to upper bound\nthe secrecy capacity of the new wiretap channel model by the secrecy capacity\nwhen the tapped subset is randomly chosen by nature.\n",
"title": "A New Wiretap Channel Model and its Strong Secrecy Capacity"
}
| null | null |
[
"Computer Science"
] | null | true | null |
5382
| null |
Validated
| null | null |
null |
{
"abstract": " Route selection based on performance measurements is an essential task in\ninter-domain Traffic Engineering. It can benefit from the detection of\nsignificant changes in RTT measurements and the understanding on potential\ncauses of change. Among the extensive works on change detection methods and\ntheir applications in various domains, few focus on RTT measurements. It is\nthus unclear which approach works the best on such data.\nIn this paper, we present an evaluation framework for change detection on RTT\ntimes series, consisting of: 1) a carefully labelled 34,008-hour RTT dataset as\nground truth; 2) a scoring method specifically tailored for RTT measurements.\nFurthermore, we proposed a data transformation that improves the detection\nperformance of existing methods. Path changes are as well attended to. We fix\nshortcomings of previous works by distinguishing path changes due to routing\nprotocols (IGP and BGP) from those caused by load balancing.\nFinally, we apply our change detection methods to a large set of measurements\nfrom RIPE Atlas. The characteristics of both RTT and path changes are analyzed;\nthe correlation between the two are also illustrated. We identify extremely\nfrequent AS path changes yet with few consequences on RTT, which has not been\nreported before.\n",
"title": "One-to-One Matching of RTT and Path Changes"
}
| null | null | null | null | true | null |
5383
| null |
Default
| null | null |
null |
{
"abstract": " We study infinite-horizon asymptotic average optimality for parallel server\nnetwork with multiple classes of jobs and multiple server pools in the\nHalfin-Whitt regime. Three control formulations are considered: 1) minimizing\nthe queueing and idleness cost, 2) minimizing the queueing cost under a\nconstraints on idleness at each server pool, and 3) fairly allocating the idle\nservers among different server pools. For the third problem, we consider a\nclass of bounded-queue, bounded-state (BQBS) stable networks, in which any\nmoment of the state is bounded by that of the queue only (for both the limiting\ndiffusion and diffusion-scaled state processes). We show that the optimal\nvalues for the diffusion-scaled state processes converge to the corresponding\nvalues of the ergodic control problems for the limiting diffusion. We present a\nfamily of state-dependent Markov balanced saturation policies (BSPs) that\nstabilize the controlled diffusion-scaled state processes. It is shown that\nunder these policies, the diffusion-scaled state process is exponentially\nergodic, provided that at least one class of jobs has a positive abandonment\nrate. We also establish useful moment bounds, and study the ergodic properties\nof the diffusion-scaled state processes, which play a crucial role in proving\nthe asymptotic optimality.\n",
"title": "Infinite horizon asymptotic average optimality for large-scale parallel server networks"
}
| null | null | null | null | true | null |
5384
| null |
Default
| null | null |
null |
{
"abstract": " Two well-known turbulence models to describe the inertial and dissipative\nranges simultaneously are by Pao~[Phys. Fluids {\\bf 8}, 1063 (1965)] and\nPope~[{\\em Turbulent Flows.} Cambridge University Press, 2000]. In this paper,\nwe compute energy spectrum $E(k)$ and energy flux $\\Pi(k)$ using spectral\nsimulations on grids up to $4096^3$, and show consistency between the numerical\nresults and predictions by the aforementioned models. We also construct a model\nfor laminar flows that predicts $E(k)$ and $\\Pi(k)$ to be of the form\n$\\exp(-k)$, and verify the model predictions using numerical simulations. The\nshell-to-shell energy transfers for the turbulent flows are {\\em forward and\nlocal} for both inertial and dissipative range, but those for the laminar flows\nare {\\em forward and nonlocal}.\n",
"title": "Energy fluxes and spectra for turbulent and laminar flows"
}
| null | null |
[
"Physics"
] | null | true | null |
5385
| null |
Validated
| null | null |
null |
{
"abstract": " We present a new model to explain the difference between the transport and\nspectroscopy gaps in samarium hexaboride (SmB$_6$), which has been a mystery\nfor some time. We propose that SmB$_6$ can be modeled as an intrinsic\nsemiconductor with a depletion length that diverges at cryogenic temperatures.\nIn this model, we find a self-consistent solution to Poisson's equation in the\nbulk, with boundary conditions based on Fermi energy pinning due to surface\ncharges. The solution yields band bending in the bulk; this explains the\ndifference between the two gaps because spectroscopic methods measure the gap\nnear the surface, while transport measures the average over the bulk. We also\nconnect the model to transport parameters, including the Hall coefficient and\nthermopower, using semiclassical transport theory. The divergence of the\ndepletion length additionally explains the 10-12 K feature in data for these\nparameters, demonstrating a crossover from bulk dominated transport above this\ntemperature to surface-dominated transport below this temperature. We find good\nagreement between our model and a collection of transport data from 4-40 K.\nThis model can also be generalized to materials with similar band structure.\n",
"title": "Understanding low-temperature bulk transport in samarium hexaboride without relying on in-gap bulk states"
}
| null | null | null | null | true | null |
5386
| null |
Default
| null | null |
null |
{
"abstract": " We investigate a graph probing problem in which an agent has only an\nincomplete view $G' \\subsetneq G$ of the network and wishes to explore the\nnetwork with least effort. In each step, the agent selects a node $u$ in $G'$\nto probe. After probing $u$, the agent gains the information about $u$ and its\nneighbors. All the neighbors of $u$ become \\emph{observed} and are\n\\emph{probable} in the subsequent steps (if they have not been probed). What is\nthe best probing strategy to maximize the number of nodes explored in $k$\nprobes? This problem serves as a fundamental component for other\ndecision-making problems in incomplete networks such as information harvesting\nin social networks, network crawling, network security, and viral marketing\nwith incomplete information.\nWhile there are a few methods proposed for the problem, none can perform\nconsistently well across different network types. In this paper, we establish a\nstrong (in)approximability for the problem, proving that no algorithm can\nguarantees finite approximation ratio unless P=NP. On the bright side, we\ndesign learning frameworks to capture the best probing strategies for\nindividual network. Our extensive experiments suggest that our framework can\nlearn efficient probing strategies that \\emph{consistently} outperform previous\nheuristics and metric-based approaches.\n",
"title": "Towards Optimal Strategy for Adaptive Probing in Incomplete Networks"
}
| null | null | null | null | true | null |
5387
| null |
Default
| null | null |
null |
{
"abstract": " In recent years, work has been done to develop the theory of General\nReinforcement Learning (GRL). However, there are few examples demonstrating\nthese results in a concrete way. In particular, there are no examples\ndemonstrating the known results regarding gener- alised discounting. We have\nadded to the GRL simulation platform AIXIjs the functionality to assign an\nagent arbitrary discount functions, and an environment which can be used to\ndetermine the effect of discounting on an agent's policy. Using this, we\ninvestigate how geometric, hyperbolic and power discounting affect an informed\nagent in a simple MDP. We experimentally reproduce a number of theoretical\nresults, and discuss some related subtleties. It was found that the agent's\nbehaviour followed what is expected theoretically, assuming appropriate\nparameters were chosen for the Monte-Carlo Tree Search (MCTS) planning\nalgorithm.\n",
"title": "Generalised Discount Functions applied to a Monte-Carlo AImu Implementation"
}
| null | null | null | null | true | null |
5388
| null |
Default
| null | null |
null |
{
"abstract": " We have observed the Vela pulsar for one year using a Phased Array Feed (PAF)\nreceiver on the 12-metre antenna of the Parkes Test-Bed Facility. These\nobservations have allowed us to investigate the stability of the PAF\nbeam-weights over time, to demonstrate that pulsars can be timed over long\nperiods using PAF technology and to detect and study the most recent glitch\nevent that occurred on 12 December 2016. The beam-weights are shown to be\nstable to 1% on time scales on the order of three weeks. We discuss the\nimplications of this for monitoring pulsars using PAFs on single dish\ntelescopes.\n",
"title": "One year of monitoring the Vela pulsar using a Phased Array Feed"
}
| null | null | null | null | true | null |
5389
| null |
Default
| null | null |
null |
{
"abstract": " Elasticity is one of the key features of cloud computing that attracts many\nSaaS providers to minimize their services' cost. Cost is minimized by\nautomatically provision and release computational resources depend on actual\ncomputational needs. However, delay of starting up new virtual resources can\ncause Service Level Agreement violation. Consequently, predicting cloud\nresources provisioning gains a lot of attention to scale computational\nresources in advance. However, most of current approaches do not consider\nmulti-seasonality in cloud workloads. This paper proposes cloud resource\nprovisioning prediction algorithm based on Holt-Winters exponential smoothing\nmethod. The proposed algorithm extends Holt-Winters exponential smoothing\nmethod to model cloud workload with multi-seasonal cycles. Prediction accuracy\nof the proposed algorithm has been improved by employing Artificial Bee Colony\nalgorithm to optimize its parameters. Performance of the proposed algorithm has\nbeen evaluated and compared with double and triple exponential smoothing\nmethods. Our results have shown that the proposed algorithm outperforms other\nmethods.\n",
"title": "Using Multiple Seasonal Holt-Winters Exponential Smoothing to Predict Cloud Resource Provisioning"
}
| null | null | null | null | true | null |
5390
| null |
Default
| null | null |
null |
{
"abstract": " Cooper pairs in superconductors are normally spin singlet. Nevertheless,\nrecent studies suggest that spin-triplet Cooper pairs can be created at\ncarefully engineered superconductor-ferromagnet interfaces. If Cooper pairs are\nspin-polarized they would transport not only charge but also a net spin\ncomponent, but without dissipation, and therefore minimize the heating effects\nassociated with spintronic devices. Although it is now established that triplet\nsupercurrents exist, their most interesting property - spin - is only inferred\nindirectly from transport measurements. In conventional spintronics, it is well\nknown that spin currents generate spin-transfer torques that alter\nmagnetization dynamics and switch magnetic moments. The observation of similar\neffects due to spin-triplet supercurrents would not only confirm the net spin\nof triplet pairs but also pave the way for applications of superconducting\nspintronics. Here, we present a possible evidence for spin-transfer torques\ninduced by triplet supercurrents in superconductor/ferromagnet/superconductor\n(S/F/S) Josephson junctions. Below the superconducting transition temperature\nT_c, the ferromagnetic resonance (FMR) field at X-band (~ 9.0 GHz) shifts\nrapidly to a lower field with decreasing temperature due to the spin-transfer\ntorques induced by triplet supercurrents. In contrast, this phenomenon is\nabsent in ferromagnet/superconductor (F/S) bilayers and\nsuperconductor/insulator/ferromagnet/superconductor (S/I/F/S) multilayers where\nno supercurrents pass through the ferromagnetic layer. These experimental\nobservations are discussed with theoretical predictions for ferromagnetic\nJosephson junctions with precessing magnetization.\n",
"title": "Possible evidence for spin-transfer torque induced by spin-triplet supercurrent"
}
| null | null |
[
"Physics"
] | null | true | null |
5391
| null |
Validated
| null | null |
null |
{
"abstract": " Quantum parameter estimation plays a key role in many fields like quantum\ncomputation, communication and metrology. Optimal estimation allows one to\nachieve the most precise parameter estimates, but requires accurate knowledge\nof the model. Any inevitable uncertainty in the model parameters may heavily\ndegrade the quality of the estimate. It is therefore desired to make the\nestimation process robust to such uncertainties. Robust estimation was\npreviously studied for a varying phase, where the goal was to estimate the\nphase at some time in the past, using the measurement results from both before\nand after that time within a fixed time interval up to current time. Here, we\nconsider a robust guaranteed-cost filter yielding robust estimates of a varying\nphase in real time, where the current phase is estimated using only past\nmeasurements. Our filter minimizes the largest (worst-case) variance in the\nallowable range of the uncertain model parameter(s) and this determines its\nguaranteed cost. It outperforms in the worst case the optimal Kalman filter\ndesigned for the model with no uncertainty, that corresponds to the center of\nthe possible range of the uncertain parameter(s). Moreover, unlike the Kalman\nfilter, our filter in the worst case always performs better than the best\nachievable variance for heterodyne measurements, that we consider as the\ntolerable threshold for our system. Furthermore, we consider effective quantum\nefficiency and effective noise power, and show that our filter provides the\nbest results by these measures in the worst case.\n",
"title": "Robust Guaranteed-Cost Adaptive Quantum Phase Estimation"
}
| null | null | null | null | true | null |
5392
| null |
Default
| null | null |
null |
{
"abstract": " We propose a multi-view network for text classification. Our method\nautomatically creates various views of its input text, each taking the form of\nsoft attention weights that distribute the classifier's focus among a set of\nbase features. For a bag-of-words representation, each view focuses on a\ndifferent subset of the text's words. Aggregating many such views results in a\nmore discriminative and robust representation. Through a novel architecture\nthat both stacks and concatenates views, we produce a network that emphasizes\nboth depth and width, allowing training to converge quickly. Using our\nmulti-view architecture, we establish new state-of-the-art accuracies on two\nbenchmark tasks.\n",
"title": "End-to-End Multi-View Networks for Text Classification"
}
| null | null | null | null | true | null |
5393
| null |
Default
| null | null |
null |
{
"abstract": " To understand the multiple relations between developers and projects on\nGitHub as a whole, we model them as a multilayer bipartite network and analyze\nthe degree distributions, the nearest neighbors' degree distributions and their\ncorrelations with degree, and the collaborative similarity distributions and\ntheir correlations with degree. Our results show that all degree distributions\nhave a power-law form, especially, the degree distribution of projects in\nwatching layer has double power-law form. Negative correlations between nearest\nneighbors' degree and degree for both developers and projects are observed in\nboth layers, exhibiting a disassortative mixing pattern. The collaborative\nsimilarity of both developers and projects negatively correlates with degree in\nwatching layer, while a positive correlations is observed for developers in\nforking layer and no obvious correlation is observed for projects in forking\nlayer.\n",
"title": "Collaborative similarity analysis of multilayer developer-project bipartite network"
}
| null | null |
[
"Computer Science",
"Physics"
] | null | true | null |
5394
| null |
Validated
| null | null |
null |
{
"abstract": " We consider a class of participation rights, i.e. obligations issued by a\ncompany to investors who are interested in performance-based compensation.\nAlbeit having desirable economic properties equity-based debt obligations\n(EbDO) pose challenges in accounting and contract pricing. We formulate and\nsolve the associated mathematical problem in a discrete time, as well as a\ncontinuous time setting. In the latter case the problem is reduced to a\nforward-backward stochastic differential equation (FBSDE) and solved using the\nmethod of decoupling fields.\n",
"title": "Evaluation of equity-based debt obligations"
}
| null | null | null | null | true | null |
5395
| null |
Default
| null | null |
null |
{
"abstract": " Online sparse linear regression is an online problem where an algorithm\nrepeatedly chooses a subset of coordinates to observe in an adversarially\nchosen feature vector, makes a real-valued prediction, receives the true label,\nand incurs the squared loss. The goal is to design an online learning algorithm\nwith sublinear regret to the best sparse linear predictor in hindsight. Without\nany assumptions, this problem is known to be computationally intractable. In\nthis paper, we make the assumption that data matrix satisfies restricted\nisometry property, and show that this assumption leads to computationally\nefficient algorithms with sublinear regret for two variants of the problem. In\nthe first variant, the true label is generated according to a sparse linear\nmodel with additive Gaussian noise. In the second, the true label is chosen\nadversarially.\n",
"title": "Adaptive Feature Selection: Computationally Efficient Online Sparse Linear Regression under RIP"
}
| null | null | null | null | true | null |
5396
| null |
Default
| null | null |
null |
{
"abstract": " Image-guided radiation therapy can benefit from accurate motion tracking by\nultrasound imaging, in order to minimize treatment margins and radiate moving\nanatomical targets, e.g., due to breathing. One way to formulate this tracking\nproblem is the automatic localization of given tracked anatomical landmarks\nthroughout a temporal ultrasound sequence. For this, we herein propose a\nfully-convolutional Siamese network that learns the similarity between pairs of\nimage regions containing the same landmark. Accordingly, it learns to localize\nand thus track arbitrary image features, not only predefined anatomical\nstructures. We employ a temporal consistency model as a location prior, which\nwe combine with the network-predicted location probability map to track a\ntarget iteratively in ultrasound sequences. We applied this method on the\ndataset of the Challenge on Liver Ultrasound Tracking (CLUST) with competitive\nresults, where our work is the first to effectively apply CNNs on this tracking\nproblem, thanks to our temporal regularization.\n",
"title": "Siamese Networks with Location Prior for Landmark Tracking in Liver Ultrasound Sequences"
}
| null | null | null | null | true | null |
5397
| null |
Default
| null | null |
null |
{
"abstract": " We report 3D coherent diffractive imaging of Au/Pd core-shell nanoparticles\nwith 6 nm resolution on 5-6 femtosecond timescales. We measured single-shot\ndiffraction patterns of core-shell nanoparticles using very intense and short\nx-ray free electron laser pulses. By taking advantage of the curvature of the\nEwald sphere and the symmetry of the nanoparticle, we reconstructed the 3D\nelectron density of 34 core-shell structures from single-shot diffraction\npatterns. We determined the size of the Au core and the thickness of the Pd\nshell to be 65.0 +/- 1.0 nm and 4.0 +/- 0.5 nm, respectively, and identified\nthe 3D elemental distribution inside the nanoparticles with an accuracy better\nthan 2%. We anticipate this method can be used for quantitative 3D imaging of\nsymmetrical nanostructures and virus particles.\n",
"title": "Single-Shot 3D Diffractive Imaging of Core-Shell Nanoparticles with Elemental Specificity"
}
| null | null | null | null | true | null |
5398
| null |
Default
| null | null |
null |
{
"abstract": " Context: We describe the new SEPIA (Swedish-ESO PI Instrument for APEX)\nreceiver, which was designed and built by the Group for Advanced Receiver\nDevelopment (GARD), at Onsala Space Observatory (OSO) in collaboration with\nESO. It was installed and commissioned at the APEX telescope during 2015 with\nan ALMA Band 5 receiver channel and updated with a new frequency channel (ALMA\nBand 9) in February 2016. Aims: This manuscript aims to provide, for observers\nwho use the SEPIA receiver, a reference in terms of the hardware description,\noptics and performance as well as the commissioning results. Methods: Out of\nthree available receiver cartridge positions in SEPIA, the two current\nfrequency channels, corresponding to ALMA Band 5, the RF band 158--211 GHz, and\nBand 9, the RF band 600--722 GHz, provide state-of-the-art dual polarization\nreceivers. The Band 5 frequency channel uses 2SB SIS mixers with an average SSB\nnoise temperature around 45K with IF (intermediate frequency) band 4--8 GHz for\neach sideband providing total 4x4 GHz IF band. The Band 9 frequency channel\nuses DSB SIS mixers with a noise temperature of 75--125K with IF band 4--12 GHz\nfor each polarization. Results: Both current SEPIA receiver channels are\navailable to all APEX observers.\n",
"title": "SEPIA - a new single pixel receiver at the APEX Telescope"
}
| null | null | null | null | true | null |
5399
| null |
Default
| null | null |
null |
{
"abstract": " We present an algorithm to identify sparse dependence structure in continuous\nand non-Gaussian probability distributions, given a corresponding set of data.\nThe conditional independence structure of an arbitrary distribution can be\nrepresented as an undirected graph (or Markov random field), but most\nalgorithms for learning this structure are restricted to the discrete or\nGaussian cases. Our new approach allows for more realistic and accurate\ndescriptions of the distribution in question, and in turn better estimates of\nits sparse Markov structure. Sparsity in the graph is of interest as it can\naccelerate inference, improve sampling methods, and reveal important\ndependencies between variables. The algorithm relies on exploiting the\nconnection between the sparsity of the graph and the sparsity of transport\nmaps, which deterministically couple one probability measure to another.\n",
"title": "Beyond normality: Learning sparse probabilistic graphical models in the non-Gaussian setting"
}
| null | null | null | null | true | null |
5400
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.