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 nature of the nematic state in FeSe remains one of the major unsolved\nmysteries in Fe- based superconductors. Both spin and orbital physics have been\ninvoked to explain the origin of this phase. Here we present experimental\nevidence for frustrated, short-range magnetic order, as suggested by several\nrecent theoretical works, in the nematic state of FeSe. We use a combination of\nmagnetostriction, susceptibility and resistivity measurements to probe the\nin-plane anisotropies of the nematic state and its associated fluctuations.\nDespite the absence of long-range magnetic order in FeSe, we observe a sizable\nin-plane magnetic susceptibility anisotropy, which is responsible for the\nfield-induced in-plane distortion inferred from magnetostriction measurements.\nFurther we demonstrate that all three anisotropies in FeSe are very similar to\nthose of BaFe2As2, which strongly suggests that the nematic phase in FeSe is\nalso of magnetic origin.\n",
"title": "Evidence for short-range magnetic order in the nematic phase of FeSe from anisotropic in-plane magnetostriction and susceptibility measurements"
}
| null | null |
[
"Physics"
] | null | true | null |
10201
| null |
Validated
| null | null |
null |
{
"abstract": " We propose a simple modification to existing neural machine translation (NMT)\nmodels that enables using a single universal model to translate between\nmultiple languages while allowing for language specific parameterization, and\nthat can also be used for domain adaptation. Our approach requires no changes\nto the model architecture of a standard NMT system, but instead introduces a\nnew component, the contextual parameter generator (CPG), that generates the\nparameters of the system (e.g., weights in a neural network). This parameter\ngenerator accepts source and target language embeddings as input, and generates\nthe parameters for the encoder and the decoder, respectively. The rest of the\nmodel remains unchanged and is shared across all languages. We show how this\nsimple modification enables the system to use monolingual data for training and\nalso perform zero-shot translation. We further show it is able to surpass\nstate-of-the-art performance for both the IWSLT-15 and IWSLT-17 datasets and\nthat the learned language embeddings are able to uncover interesting\nrelationships between languages.\n",
"title": "Contextual Parameter Generation for Universal Neural Machine Translation"
}
| null | null | null | null | true | null |
10202
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of optimizing the placement of stubborn agents in a\nsocial network in order to maximally impact population opinions.\nWe assume individuals in a directed social network each have a latent opinion\nthat evolves over time in response to social media posts by their neighbors.\nThe individuals randomly communicate noisy versions of their latent opinion to\ntheir neighbors. Each individual updates his opinion using a time-varying\nupdate rule that has him become more stubborn with time and be less affected by\nnew posts. The dynamic update rule is a novel component of our model and\nreflects realistic behaviors observed in many psychological studies.\nWe show that in the presence of stubborn agents with immutable opinions and\nunder fairly general conditions on the stubbornness rate of the individuals,\nthe opinions converge to an equilibrium determined by a linear system. We give\nan interesting electrical network interpretation of the equilibrium. We also\nuse this equilibrium to present a simple closed form expression for harmonic\ninfluence centrality, which is a function that quantifies how much a node can\naffect the mean opinion in a network. We develop a discrete optimization\nformulation for the problem of maximally shifting opinions in a network by\ntargeting nodes with stubborn agents. We show that this is an optimization\nproblem with a monotone and submodular objective, allowing us to utilize a\ngreedy algorithm. Finally, we show that a small number of stubborn agents can\nnon-trivially influence a large population using simulated networks.\n",
"title": "Opinion Dynamics with Stubborn Agents"
}
| null | null | null | null | true | null |
10203
| null |
Default
| null | null |
null |
{
"abstract": " This work fits in the context of community microgrids, where members of a\ncommunity can exchange energy and services among themselves, without going\nthrough the usual channels of the public electricity grid. We introduce and\nanalyze a framework to operate a community microgrid, and to share the\nresulting revenues and costs among its members. A market-oriented pricing of\nenergy exchanges within the community is obtained by implementing an internal\nlocal market based on the marginal pricing scheme. The market aims at\nmaximizing the social welfare of the community, thanks to the more efficient\nallocation of resources, the reduction of the peak power to be paid, and the\nincreased amount of reserve, achieved at an aggregate level. A community\nmicrogrid operator, acting as a benevolent planner, redistributes revenues and\ncosts among the members, in such a way that the solution achieved by each\nmember within the community is not worse than the solution it would achieve by\nacting individually. In this way, each member is incentivized to participate in\nthe community on a voluntary basis. The overall framework is formulated in the\nform of a bilevel model, where the lower level problem clears the market, while\nthe upper level problem plays the role of the community microgrid operator.\nNumerical results obtained on a real test case implemented in Belgium show\nsignificant cost savings on a yearly scale for the community members, as\ncompared to the case when they act individually.\n",
"title": "A Community Microgrid Architecture with an Internal Local Market"
}
| null | null | null | null | true | null |
10204
| null |
Default
| null | null |
null |
{
"abstract": " We formulate a type B extended nilHecke algebra, following the type A\nconstruction of Naisse and Vaz. We describe an action of this algebra on\nextended polynomials and describe some results on the structure on the extended\nsymmetric polynomials. Finally, following Appel, Egilmez, Hogancamp, and Lauda,\nwe prove a result analogous to a classical theorem of Solomon connecting the\nextended symmetric polynomial ring to a ring of usual symmetric polynomials and\ntheir differentials.\n",
"title": "Extended nilHecke algebra and symmetric functions in type B"
}
| null | null | null | null | true | null |
10205
| null |
Default
| null | null |
null |
{
"abstract": " In unsupervised domain mapping, the learner is given two unmatched datasets\n$A$ and $B$. The goal is to learn a mapping $G_{AB}$ that translates a sample\nin $A$ to the analog sample in $B$. Recent approaches have shown that when\nlearning simultaneously both $G_{AB}$ and the inverse mapping $G_{BA}$,\nconvincing mappings are obtained. In this work, we present a method of learning\n$G_{AB}$ without learning $G_{BA}$. This is done by learning a mapping that\nmaintains the distance between a pair of samples. Moreover, good mappings are\nobtained, even by maintaining the distance between different parts of the same\nsample before and after mapping. We present experimental results that the new\nmethod not only allows for one sided mapping learning, but also leads to\npreferable numerical results over the existing circularity-based constraint.\nOur entire code is made publicly available at\nthis https URL .\n",
"title": "One-Sided Unsupervised Domain Mapping"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10206
| null |
Validated
| null | null |
null |
{
"abstract": " We present an approach to adaptively utilize deep neural networks in order to\nreduce the evaluation time on new examples without loss of accuracy. Rather\nthan attempting to redesign or approximate existing networks, we propose two\nschemes that adaptively utilize networks. We first pose an adaptive network\nevaluation scheme, where we learn a system to adaptively choose the components\nof a deep network to be evaluated for each example. By allowing examples\ncorrectly classified using early layers of the system to exit, we avoid the\ncomputational time associated with full evaluation of the network. We extend\nthis to learn a network selection system that adaptively selects the network to\nbe evaluated for each example. We show that computational time can be\ndramatically reduced by exploiting the fact that many examples can be correctly\nclassified using relatively efficient networks and that complex,\ncomputationally costly networks are only necessary for a small fraction of\nexamples. We pose a global objective for learning an adaptive early exit or\nnetwork selection policy and solve it by reducing the policy learning problem\nto a layer-by-layer weighted binary classification problem. Empirically, these\napproaches yield dramatic reductions in computational cost, with up to a 2.8x\nspeedup on state-of-the-art networks from the ImageNet image recognition\nchallenge with minimal (<1%) loss of top5 accuracy.\n",
"title": "Adaptive Neural Networks for Efficient Inference"
}
| null | null | null | null | true | null |
10207
| null |
Default
| null | null |
null |
{
"abstract": " The phylogenetic effective sample size is a parameter that has as its goal\nthe quantification of the amount of independent signal in a phylogenetically\ncorrelated sample. It was studied for Brownian motion and Ornstein-Uhlenbeck\nmodels of trait evolution. Here, we study this composite parameter when the\ntrait is allowed to jump at speciation points of the phylogeny. Our numerical\nstudy indicates that there is a non-trivial limit as the effect of jumps grows.\nThe limit depends on the value of the drift parameter of the Ornstein-Uhlenbeck\nprocess.\n",
"title": "The phylogenetic effective sample size and jumps"
}
| null | null | null | null | true | null |
10208
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we construct the additional symmetries of the fermionic\n$(2N,2M)$-Toda hierarchy basing on the generalization of the $N{=}(1|1)$\nsupersymmetric two dimensional Toda lattice hierarchy. These additional flows\nconstitute a $w_{\\infty}\\times w_{\\infty}$ Lie algebra. As a Bosonic reduction\nof the $N{=}(1|1)$ supersymmetric two dimensional Toda lattice hierarchy and\nthe fermionic $(2N,2M)$-Toda hierarchy, we define a new extended fermionic\n$(2N,2M)$-Toda hierarchy which admits a Bosonic Block type superconformal\nstructure.\n",
"title": "Bosonic symmetries of the extended fermionic $(2N,2M)$-Toda hierarchy"
}
| null | null | null | null | true | null |
10209
| null |
Default
| null | null |
null |
{
"abstract": " Measuring the full distribution of individual particles is of fundamental\nimportance to characterize many-body quantum systems through correlation\nfunctions at any order. Here we demonstrate the possibility to reconstruct the\nmomentum-space distribution of three-dimensional interacting lattice gases\natom-by-atom. This is achieved by detecting individual metastable Helium atoms\nin the far-field regime of expansion, when released from an optical lattice. We\nbenchmark our technique with Quantum Monte-Carlo calculations, demonstrating\nthe ability to resolve momentum distributions of superfluids occupying $10^5$\nlattice sites. It permits a direct measure of the condensed fraction across\nphase transitions, as we illustrate on the superfluid-to-normal transition. Our\nsingle-atom-resolved approach opens a new route to investigate interacting\nlattice gases through momentum correlations.\n",
"title": "Single-atom-resolved probing of lattice gases in momentum space"
}
| null | null |
[
"Physics"
] | null | true | null |
10210
| null |
Validated
| null | null |
null |
{
"abstract": " The branch of provability logic investigates the provability-based behavior\nof the mathematical theories. In a more precise way, it studies the relation\nbetween a mathematical theory $T$ and a modal logic $L$ via the provability\ninterpretation which interprets the modality as the provability predicate of\n$T$. In this paper we will extend this relation to investigate the\nprovability-based behavior of a hierarchy of theories. More precisely, using\nthe modal language with infinitely many modalities,\n$\\{\\Box_n\\}_{n=0}^{\\infty}$, we will define the hierarchical counterparts of\nsome of the classical modal theories such as $\\mathbf{K4}$, $\\mathbf{KD4}$,\n$\\mathbf{GL}$ and $\\mathbf{S4}$. Then we will define their canonical\nprovability interpretations and their corresponding soundness-completeness\ntheorems.\n",
"title": "Provability Logics of Hierarchies"
}
| null | null | null | null | true | null |
10211
| null |
Default
| null | null |
null |
{
"abstract": " Physical emergence - crystals, rocks, sandpiles, turbulent eddies, planets,\nstars - is fundamentally different from biological emergence - amoeba, cells,\nmice, humans - even though the latter is based in the former. This paper points\nout that an essential difference is that as well as involving physical\ncausation, causation in biological systems has a logical nature at each level\nof the hierarchy of emergence, from the biomolecular level up. The key link\nbetween physics and life enabling this to happen is provided by biomolecules,\nsuch as voltage gated ion channels, which enable branching logic to emerge from\nthe underlying physics and hence enable logically based cell processes to take\nplace in general, and in neurons in particular. These molecules can only have\ncome into being via the contextually dependent processes of natural selection,\nwhich selects them for their biological function. A further major difference is\nbetween life in general and intelligent life. We characterise intelligent\norganisms as being engaged in deductive causation, which enables them to\ntranscend the physical limitations of their bodies through the power of\nabstract thought, prediction, and planning. Ultimately this is enabled by the\nbiomolecules that underlie the propagation of action potentials in neuronal\naxons in the brain.\n",
"title": "On the Difference between Physics and Biology: Logical Branching and Biomolecules"
}
| null | null |
[
"Physics"
] | null | true | null |
10212
| null |
Validated
| null | null |
null |
{
"abstract": " The interplay between geometric frustration (GF) and bond disorder is studied\nin the Ising kagome lattice within a cluster approach. The model considers\nantiferromagnetic (AF) short-range couplings and long-range intercluster\ndisordered interactions. The replica formalism is used to obtain an effective\nsingle cluster model from where the thermodynamics is analyzed by exact\ndiagonalization. We found that the presence of GF can introduce cluster\nfreezing at very low levels of disorder. The system exhibits an entropy plateau\nfollowed by a large entropy drop close to the freezing temperature. In this\nscenario, a spin-liquid (SL) behavior prevents conventional long-range order,\nbut an infinitesimal disorder picks out uncompensated cluster states from the\nmulti degenerate SL regime, potentializing the intercluster disordered coupling\nand bringing the cluster spin-glass state. To summarize, our results suggest\nthat the SL state combined with low levels of disorder can activate small\nclusters, providing hypersensitivity to the freezing process in geometrically\nfrustrated materials and playing a key role in the glassy stabilization. We\npropose that this physical mechanism could be present in several geometrically\nfrustrated materials. In particular, we discuss our results in connection to\nthe recent experimental investigations of the Ising kagome compound\nCo$_3$Mg(OH)$_6$Cl$_2$.\n",
"title": "Spin liquid and infinitesimal-disorder-driven cluster spin glass in the kagome lattice"
}
| null | null | null | null | true | null |
10213
| null |
Default
| null | null |
null |
{
"abstract": " In acoustic scene classification researches, audio segment is usually split\ninto multiple samples. Majority voting is then utilized to ensemble the results\nof the samples. In this paper, we propose a punishment voting algorithm based\non the super categories construction method for acoustic scene classification.\nSpecifically, we propose a DenseNet-like model as the base classifier. The base\nclassifier is trained by the CQT spectrograms generated from the raw audio\nsegments. Taking advantage of the results of the base classifier, we propose a\nsuper categories construction method using the spectral clustering. Super\nclassifiers corresponding to the constructed super categories are further\ntrained. Finally, the super classifiers are utilized to enhance the majority\nvoting of the base classifier by punishment voting. Experiments show that the\npunishment voting obviously improves the performances on both the DCASE2017\nDevelopment dataset and the LITIS Rouen dataset.\n",
"title": "A punishment voting algorithm based on super categories construction for acoustic scene classification"
}
| null | null | null | null | true | null |
10214
| null |
Default
| null | null |
null |
{
"abstract": " A new definition of continuous-time equilibrium controls is introduced. As\nopposed to the standard definition, which involves a derivative-type operation,\nthe new definition parallels how a discrete-time equilibrium is defined, and\nallows for unambiguous economic interpretation. The terms \"strong equilibria\"\nand \"weak equilibria\" are coined for controls under the new and the standard\ndefinitions, respectively. When the state process is a time-homogeneous\ncontinuous-time Markov chain, a careful asymptotic analysis gives complete\ncharacterizations of weak and strong equilibria. Thanks to Kakutani-Fan's\nfixed-point theorem, general existence of weak and strong equilibria is also\nestablished, under additional compactness assumption. Our theoretic results are\napplied to a two-state model under non-exponential discounting. In particular,\nwe demonstrate explicitly that there can be incentive to deviate from a weak\nequilibrium, which justifies the need for strong equilibria. Our analysis also\nprovides new results for the existence and characterization of discrete-time\nequilibria under infinite horizon.\n",
"title": "Strong and Weak Equilibria for Time-Inconsistent Stochastic Control in Continuous Time"
}
| null | null | null | null | true | null |
10215
| null |
Default
| null | null |
null |
{
"abstract": " Alpha signals for statistical arbitrage strategies are often driven by latent\nfactors. This paper analyses how to optimally trade with latent factors that\ncause prices to jump and diffuse. Moreover, we account for the effect of the\ntrader's actions on quoted prices and the prices they receive from trading.\nUnder fairly general assumptions, we demonstrate how the trader can learn the\nposterior distribution over the latent states, and explicitly solve the latent\noptimal trading problem. We provide a verification theorem, and a methodology\nfor calibrating the model by deriving a variation of the\nexpectation-maximization algorithm. To illustrate the efficacy of the optimal\nstrategy, we demonstrate its performance through simulations and compare it to\nstrategies which ignore learning in the latent factors. We also provide\ncalibration results for a particular model using Intel Corporation stock as an\nexample.\n",
"title": "Trading algorithms with learning in latent alpha models"
}
| null | null | null | null | true | null |
10216
| null |
Default
| null | null |
null |
{
"abstract": " Deep convolutional networks have achieved great success for image\nrecognition. However, for action recognition in videos, their advantage over\ntraditional methods is not so evident. We present a general and flexible\nvideo-level framework for learning action models in videos. This method, called\ntemporal segment network (TSN), aims to model long-range temporal structures\nwith a new segment-based sampling and aggregation module. This unique design\nenables our TSN to efficiently learn action models by using the whole action\nvideos. The learned models could be easily adapted for action recognition in\nboth trimmed and untrimmed videos with simple average pooling and multi-scale\ntemporal window integration, respectively. We also study a series of good\npractices for the instantiation of TSN framework given limited training\nsamples. Our approach obtains the state-the-of-art performance on four\nchallenging action recognition benchmarks: HMDB51 (71.0%), UCF101 (94.9%),\nTHUMOS14 (80.1%), and ActivityNet v1.2 (89.6%). Using the proposed RGB\ndifference for motion models, our method can still achieve competitive accuracy\non UCF101 (91.0%) while running at 340 FPS. Furthermore, based on the temporal\nsegment networks, we won the video classification track at the ActivityNet\nchallenge 2016 among 24 teams, which demonstrates the effectiveness of TSN and\nthe proposed good practices.\n",
"title": "Temporal Segment Networks for Action Recognition in Videos"
}
| null | null | null | null | true | null |
10217
| null |
Default
| null | null |
null |
{
"abstract": " The current study applies deep learning to herbalism. Toward the goal, we\nacquired the de-identified health insurance reimbursements that were claimed in\na 10-year period from 2004 to 2013 in the National Health Insurance Database of\nTaiwan, the total number of reimbursement records equaling 340 millions. Two\nartificial intelligence techniques were applied to the dataset: residual\nconvolutional neural network multitask classifier and attention-based recurrent\nneural network. The former works to translate from herbal prescriptions to\ndiseases; and the latter from diseases to herbal prescriptions. Analysis of the\nclassification results indicates that herbal prescriptions are specific to:\nanatomy, pathophysiology, sex and age of the patient, and season and year of\nthe prescription. Further analysis identifies temperature and gross domestic\nproduct as the meteorological and socioeconomic factors that are associated\nwith herbal prescriptions. Analysis of the neural machine transitional result\nindicates that the recurrent neural network learnt not only syntax but also\nsemantics of diseases and herbal prescriptions.\n",
"title": "Neural Machine Translation between Herbal Prescriptions and Diseases"
}
| null | null | null | null | true | null |
10218
| null |
Default
| null | null |
null |
{
"abstract": " In privacy amplification, two mutually trusted parties aim to amplify the\nsecrecy of an initial shared secret $X$ in order to establish a shared private\nkey $K$ by exchanging messages over an insecure communication channel. If the\nchannel is authenticated the task can be solved in a single round of\ncommunication using a strong randomness extractor; choosing a quantum-proof\nextractor allows one to establish security against quantum adversaries.\nIn the case that the channel is not authenticated, Dodis and Wichs (STOC'09)\nshowed that the problem can be solved in two rounds of communication using a\nnon-malleable extractor, a stronger pseudo-random construction than a strong\nextractor.\nWe give the first construction of a non-malleable extractor that is secure\nagainst quantum adversaries. The extractor is based on a construction by Li\n(FOCS'12), and is able to extract from source of min-entropy rates larger than\n$1/2$. Combining this construction with a quantum-proof variant of the\nreduction of Dodis and Wichs, shown by Cohen and Vidick (unpublished), we\nobtain the first privacy amplification protocol secure against active quantum\nadversaries.\n",
"title": "A Quantum-Proof Non-Malleable Extractor, With Application to Privacy Amplification against Active Quantum Adversaries"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10219
| null |
Validated
| null | null |
null |
{
"abstract": " In a reversible language, any forward computation can be undone by a finite\nsequence of backward steps. Reversible computing has been studied in the\ncontext of different programming languages and formalisms, where it has been\nused for testing and verification, among others. In this paper, we consider a\nsubset of Erlang, a functional and concurrent programming language based on the\nactor model. We present a formal semantics for reversible computation in this\nlanguage and prove its main properties, including its causal consistency. We\nalso build on top of it a rollback operator that can be used to undo the\nactions of a process up to a given checkpoint.\n",
"title": "A Theory of Reversibility for Erlang"
}
| null | null | null | null | true | null |
10220
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we investigate the convergence and consistency properties of\nan Invariant-Extended Kalman Filter (RI-EKF) based Simultaneous Localization\nand Mapping (SLAM) algorithm. Basic convergence properties of this algorithm\nare proven. These proofs do not require the restrictive assumption that the\nJacobians of the motion and observation models need to be evaluated at the\nground truth. It is also shown that the output of RI-EKF is invariant under any\nstochastic rigid body transformation in contrast to $\\mathbb{SO}(3)$ based EKF\nSLAM algorithm ($\\mathbb{SO}(3)$-EKF) that is only invariant under\ndeterministic rigid body transformation. Implications of these invariance\nproperties on the consistency of the estimator are also discussed. Monte Carlo\nsimulation results demonstrate that RI-EKF outperforms $\\mathbb{SO}(3)$-EKF,\nRobocentric-EKF and the \"First Estimates Jacobian\" EKF, for 3D point feature\nbased SLAM.\n",
"title": "Convergence and Consistency Analysis for A 3D Invariant-EKF SLAM"
}
| null | null | null | null | true | null |
10221
| null |
Default
| null | null |
null |
{
"abstract": " Nanotubes of various kinds have been prepared in the last decade, starting\nfrom the discovery of carbon nanotubes. Recently other types of nanotubes\nincluding metallic (Au), inorganic (TiO2, HfS2, V7O16, CdSe, MoS2), and\npolymeric (polyaniline, polyacrylonitrile) have been produced. Herein we\npresent a novel synthetic procedure leading to a new kind of porous,\nhigh-surface-area nanoparticle nanotubes (NPNTs). This study characterizes the\nsynthesized silver nanotubes at optical wavelengths. The absorption spectrum of\nPAN washed silver nanotubes shows an extended absorption peak at visible\nwavelengths ranging from 350 to 700 nm. In addition, the absorption spectrum of\nrandomly oriented silver nanotubes showed plasmonic behavior, indicating high\nefficient surface enhanced Raman scattering (SERS) performance.\n",
"title": "Optical Characterization of Electro-spun Polymer Nanofiber based Silver Nanotubes"
}
| null | null | null | null | true | null |
10222
| null |
Default
| null | null |
null |
{
"abstract": " As with classic statistics, functional regression models are invaluable in\nthe analysis of functional data. While there are now extensive tools with\naccompanying theory available for linear models, there is still a great deal of\nwork to be done concerning nonlinear models for functional data. In this work\nwe consider the Additive Function-on-Function Regression model, a type of\nnonlinear model that uses an additive relationship between the functional\noutcome and functional covariate. We present an estimation methodology built\nupon Reproducing Kernel Hilbert Spaces, and establish optimal rates of\nconvergence for our estimates in terms of prediction error. We also discuss\ncomputational challenges that arise with such complex models, developing a\nrepresenter theorem for our estimate as well as a more practical and\ncomputationally efficient approximation. Simulations and an application to\nCumulative Intraday Returns around the 2008 financial crisis are also provided.\n",
"title": "Optimal Prediction for Additive Function-on-Function Regression"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
10223
| null |
Validated
| null | null |
null |
{
"abstract": " The star chromatic index of a multigraph $G$, denoted $\\chi'_{s}(G)$, is the\nminimum number of colors needed to properly color the edges of $G$ such that no\npath or cycle of length four is bi-colored. A multigraph $G$ is star\n$k$-edge-colorable if $\\chi'_{s}(G)\\le k$. Dvořák, Mohar and Šámal\n[Star chromatic index, J. Graph Theory 72 (2013), 313--326] proved that every\nsubcubic multigraph is star $7$-edge-colorable. They conjectured in the same\npaper that every subcubic multigraph should be star $6$-edge-colorable. In this\npaper, we first prove that it is NP-complete to determine whether\n$\\chi'_s(G)\\le3$ for an arbitrary graph $G$. This answers a question of Mohar.\nWe then establish some structure results on subcubic multigraphs $G$ with\n$\\delta(G)\\le2$ such that $\\chi'_s(G)>k$ but $\\chi'_s(G-v)\\le k$ for any $v\\in\nV(G)$, where $k\\in\\{5,6\\}$. We finally apply the structure results, along with\na simple discharging method, to prove that every subcubic multigraph $G$ is\nstar $6$-edge-colorable if $mad(G)<5/2$, and star $5$-edge-colorable if\n$mad(G)<24/11$, respectively, where $mad(G)$ is the maximum average degree of a\nmultigraph $G$. This partially confirms the conjecture of Dvořák, Mohar\nand Šámal.\n",
"title": "Star chromatic index of subcubic multigraphs"
}
| null | null |
[
"Mathematics"
] | null | true | null |
10224
| null |
Validated
| null | null |
null |
{
"abstract": " Absolutely Koszul algebras are a class of rings over which any finite graded\nmodule has a rational Poincaré series. We provide a criterion to detect\nnon-absolutely Koszul rings. Combining the criterion with machine computations,\nwe identify large families of Veronese subrings and Segre products of\npolynomial rings which are not absolutely Koszul. In particular, we classify\ncompletely the absolutely Koszul algebras among Segre products of polynomial\nrings, at least in characteristic $0$.\n",
"title": "The absolutely Koszul property of Veronese subrings and Segre products"
}
| null | null | null | null | true | null |
10225
| null |
Default
| null | null |
null |
{
"abstract": " Online advertisement is the main source of revenue for Internet business.\nAdvertisers are typically ranked according to a score that takes into account\ntheir bids and potential click-through rates(eCTR). Generally, the likelihood\nthat a user clicks on an ad is often modeled by optimizing for the click\nthrough rates rather than the performance of the auction in which the click\nthrough rates will be used. This paper attempts to eliminate this\ndis-connection by proposing loss functions for click modeling that are based on\nfinal auction performance.In this paper, we address two feasible metrics (AUC^R\nand SAUC) to evaluate the on-line RPM (revenue per mille) directly rather than\nthe CTR. And then, we design an explicit ranking function by incorporating the\ncalibration fac-tor and price-squashed factor to maximize the revenue. Given\nthe power of deep networks, we also explore an implicit optimal ranking\nfunction with deep model. Lastly, various experiments with two real world\ndatasets are presented. In particular, our proposed methods perform better than\nthe state-of-the-art methods with regard to the revenue of the platform.\n",
"title": "Learning Theory and Algorithms for Revenue Management in Sponsored Search"
}
| null | null | null | null | true | null |
10226
| null |
Default
| null | null |
null |
{
"abstract": " To a smooth and symmetric function $f$ defined on a symmetric open set\n$\\Gamma\\subset\\mathbb{R}^{n}$ and a real $n$-dimensional vector space $V$ we\nassign an associated operator function $F$ defined on an open subset\n$\\Omega\\subset\\mathcal{L}(V)$ of linear transformations of $V$, such that for\neach inner product $g$ on $V$, on the subspace\n$\\Sigma_{g}(V)\\subset\\mathcal{L}(V)$ of $g$-selfadjoint operators,\n$F_{g}=F_{|\\Sigma_{g}(V)}$ is the isotropic function associated to $f$, which\nmeans that $F_{g}(A)=f(\\mathrm{EV}(A))$, where $\\mathrm{EV}(A)$ denotes the\nordered $n$-tuple of real eigenvalues of $A$. We extend some well known\nrelations between the derivatives of $f$ and each $F_{g}$ to relations between\n$f$ and $F$. By means of an example we show that well known regularity\nproperties of $F_{g}$ do not carry over to $F$.\n",
"title": "Isotropic functions revisited"
}
| null | null | null | null | true | null |
10227
| null |
Default
| null | null |
null |
{
"abstract": " We demonstrate an enhancement in the vortex generation when artificial gauge\npotential is introduced to condensates confined in a double well potential.\nThis is due to the lower energy required to create a vortex in the low\ncondensate density region within the barrier. Furthermore, we study the\ntransport of vortices between the two wells, and show that the traverse time\nfor vortices is longer for the lower height of the well. We also show that the\ncritical value of synthetic magnetic field to inject vortices into the bulk of\nthe condensate is lower in the double-well potential compared to the harmonic\nconfining potential.\n",
"title": "Condensates in double-well potential with synthetic gauge potentials and vortex seeding"
}
| null | null | null | null | true | null |
10228
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the sub-Gaussian property for almost surely bounded random\nvariables. If sub-Gaussianity per se is de facto ensured by the bounded support\nof said random variables, then exciting research avenues remain open. Among\nthese questions is how to characterize the optimal sub-Gaussian proxy variance?\nAnother question is how to characterize strict sub-Gaussianity, defined by a\nproxy variance equal to the (standard) variance? We address the questions in\nproposing conditions based on the study of functions variations. A particular\nfocus is given to the relationship between strict sub-Gaussianity and symmetry\nof the distribution. In particular, we demonstrate that symmetry is neither\nsufficient nor necessary for strict sub-Gaussianity. In contrast, simple\nnecessary conditions on the one hand, and simple sufficient conditions on the\nother hand, for strict sub-Gaussianity are provided. These results are\nillustrated via various applications to a number of bounded random variables,\nincluding Bernoulli, beta, binomial, uniform, Kumaraswamy, and triangular\ndistributions.\n",
"title": "On strict sub-Gaussianity, optimal proxy variance and symmetry for bounded random variables"
}
| null | null | null | null | true | null |
10229
| null |
Default
| null | null |
null |
{
"abstract": " Finding a maximum cut is a fundamental task in many computational settings.\nSurprisingly, it has been insufficiently studied in the classic distributed\nsettings, where vertices communicate by synchronously sending messages to their\nneighbors according to the underlying graph, known as the $\\mathcal{LOCAL}$ or\n$\\mathcal{CONGEST}$ models. We amend this by obtaining almost optimal\nalgorithms for Max-Cut on a wide class of graphs in these models. In\nparticular, for any $\\epsilon > 0$, we develop randomized approximation\nalgorithms achieving a ratio of $(1-\\epsilon)$ to the optimum for Max-Cut on\nbipartite graphs in the $\\mathcal{CONGEST}$ model, and on general graphs in the\n$\\mathcal{LOCAL}$ model.\nWe further present efficient deterministic algorithms, including a\n$1/3$-approximation for Max-Dicut in our models, thus improving the best known\n(randomized) ratio of $1/4$. Our algorithms make non-trivial use of the greedy\napproach of Buchbinder et al. (SIAM Journal on Computing, 2015) for maximizing\nan unconstrained (non-monotone) submodular function, which may be of\nindependent interest.\n",
"title": "Fast Distributed Approximation for Max-Cut"
}
| null | null | null | null | true | null |
10230
| null |
Default
| null | null |
null |
{
"abstract": " Let $S$ be a $p$-group for an odd prime $p$, Oliver proposed the conjecture\nthat the Thompson subgroup $J(S)$ is always contained in the Oliver subgroup\n$\\mathfrak{X}(S)$. That means he conjectured that\n$|J(S)\\mathfrak{X}(S):\\mathfrak{X}(S)|=1$. Let $\\mathfrak{X}_1(S)$ be a\nsubgroup of $S$ such that $\\mathfrak{X}_1(S)/\\mathfrak{X}(S)$ is the center of\n$S/\\mathfrak{X}(S)$. In this short note, we prove that $J(S)\\leq\n\\mathfrak{X}(S)$ if and only if $J(S)\\leq \\mathfrak{X}_1(S)$.\nAs an easy application, we prove that\n$|J(S)\\mathfrak{X}(S):\\mathfrak{X}(S)|\\neq p$.\n",
"title": "A note on Oliver's p-group conjecture"
}
| null | null |
[
"Mathematics"
] | null | true | null |
10231
| null |
Validated
| null | null |
null |
{
"abstract": " Suppose we have a elliptic curve over a number field whose mod $l$\nrepresentation has image isomorphic to $SL_2(\\mathbb{F}_l)$. We present a\nmethod to determine Frobenius elements of the associated Galois group which\nincorporates the linear structure available. We are able to distinguish\n$SL_n(\\mathbb{F}_l)$-conjugacy from $GL_n(\\mathbb{F}_l)$-conjugacy; this can be\nthought of as being analogous to a result which distinguishes $A_n$-conjugacy\nfrom $S_n$-conjugacy when the Galois group is considered as a permutation\ngroup.\n",
"title": "Frobenius elements in Galois representations with SL_n image"
}
| null | null | null | null | true | null |
10232
| null |
Default
| null | null |
null |
{
"abstract": " The precise modeling of subatomic particle interactions and propagation\nthrough matter is paramount for the advancement of nuclear and particle physics\nsearches and precision measurements. The most computationally expensive step in\nthe simulation pipeline of a typical experiment at the Large Hadron Collider\n(LHC) is the detailed modeling of the full complexity of physics processes that\ngovern the motion and evolution of particle showers inside calorimeters. We\nintroduce \\textsc{CaloGAN}, a new fast simulation technique based on generative\nadversarial networks (GANs). We apply these neural networks to the modeling of\nelectromagnetic showers in a longitudinally segmented calorimeter, and achieve\nspeedup factors comparable to or better than existing full simulation\ntechniques on CPU ($100\\times$-$1000\\times$) and even faster on GPU (up to\n$\\sim10^5\\times$). There are still challenges for achieving precision across\nthe entire phase space, but our solution can reproduce a variety of geometric\nshower shape properties of photons, positrons and charged pions. This\nrepresents a significant stepping stone toward a full neural network-based\ndetector simulation that could save significant computing time and enable many\nanalyses now and in the future.\n",
"title": "CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks"
}
| null | null | null | null | true | null |
10233
| null |
Default
| null | null |
null |
{
"abstract": " We demonstrate temporal measurements of subpicosecond optical pulses via\ntime-to-frequency conversion in a 45cm long CMOS compatible high index glass\nspiral waveguide. The measurements are based on efficient four wave mixing in\nthe C-band, using around 1W of peak pump power. We achieve a resolution of\n400fs over a time window of 100ps, representing a time-bandwidth product > 250.\n",
"title": "Measurement of ultrashort optical pulses via time lens imaging in CMOS compatible waveguides"
}
| null | null | null | null | true | null |
10234
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we present several values for the next-to-minimal weights of\nprojective Reed-Muller codes. We work over $\\mathbb{F}_q$ with $q \\geq 3$ since\nin IEEE-IT 62(11) p. 6300-6303 (2016) we have determined the complete values\nfor the next-to-minimal weights of binary projective Reed-Muller codes. As in\nloc. cit. here we also find examples of codewords with next-to-minimal weight\nwhose set of zeros is not in a hyperplane arrangement.\n",
"title": "On the next-to-minimal weight of projective Reed-Muller codes"
}
| null | null | null | null | true | null |
10235
| null |
Default
| null | null |
null |
{
"abstract": " The AliEn (ALICE Environment) file catalogue is a global unique namespace\nproviding mapping between a UNIX-like logical name structure and the\ncorresponding physical files distributed over 80 storage elements worldwide.\nPowerful search tools and hierarchical metadata information are integral parts\nof the system and are used by the Grid jobs as well as local users to store and\naccess all files on the Grid storage elements. The catalogue has been in\nproduction since 2005 and over the past 11 years has grown to more than 2\nbillion logical file names. The backend is a set of distributed relational\ndatabases, ensuring smooth growth and fast access. Due to the anticipated fast\nfuture growth, we are looking for ways to enhance the performance and\nscalability by simplifying the catalogue schema while keeping the functionality\nintact. We investigated different backend solutions, such as distributed key\nvalue stores, as replacement for the relational database. This contribution\ncovers the architectural changes in the system, together with the technology\nevaluation, benchmark results and conclusions.\n",
"title": "Scalable Global Grid catalogue for LHC Run3 and beyond"
}
| null | null | null | null | true | null |
10236
| null |
Default
| null | null |
null |
{
"abstract": " We report on a versatile mini ultra-high vacuum (UHV) chamber which is\ndesigned to be used on the MAgnetic Reflectometer with high Incident Angle of\nthe Jülich Centre for Neutron Science at Heinz Maier-Leibnitz Zentrum in\nGarching, Germany. Samples are prepared in the adjacent thin film laboratory by\nmolecular beam epitaxy and moved into the compact chamber for transfer without\nexposure to ambient air. The chamber is based on DN 40 CF flanges and equipped\nwith sapphire view ports, a small getter pump, and a wobble stick, which serves\nalso as sample holder. Here, we present polarized neutron reflectivity\nmeasurements which have been performed on Co thin films at room temperature in\nUHV and in ambient air in a magnetic field of 200 mT and in the Q-range of 0.18\n\\AA$^{-1}$. The results confirm that the Co film is not contaminated during the\npolarized neutron reflectivity measurement. Herewith it is demonstrated that\nthe mini UHV transport chamber also works as a measurement chamber which opens\nnew possibilities for polarized neutron measurements under UHV conditions.\n",
"title": "A versatile UHV transport and measurement chamber for neutron reflectometry under UHV conditions"
}
| null | null | null | null | true | null |
10237
| null |
Default
| null | null |
null |
{
"abstract": " The dislocation-mediated quantum melting of solids into quantum liquid\ncrystals is extended from two to three spatial dimensions, using a\ngeneralization of boson-vortex or Abelian-Higgs duality. Dislocations are now\nBurgers-vector-valued strings that trace out worldsheets in spacetime while the\nphonons of the solid dualize into two-form (Kalb-Ramond) gauge fields. We\npropose an effective dual Higgs potential that allows for restoring\ntranslational symmetry in either one, two or three directions, leading to the\nquantum analogues of columnar, smectic or nematic liquid crystals. In these\nphases, transverse phonons turn into gapped, propagating modes while\ncompressional stress remains massless. Rotational Goldstone modes emerge\nwhenever translational symmetry is restored. We also consider electrically\ncharged matter, and find amongst others that as a hard principle only two out\nof the possible three rotational Goldstone modes are observable using\nelectromagnetic means.\n",
"title": "Dual gauge field theory of quantum liquid crystals in three dimensions"
}
| null | null | null | null | true | null |
10238
| null |
Default
| null | null |
null |
{
"abstract": " The mechanical properties and deformation mechanisms of Cu/Nb nanoscale\nmetallic multilayers (NMMs) manufactured by accumulative roll bonding (ARB) are\nstudied at 25C and 400C. Cu/Nb NMMs with individual layer thicknesses between 7\nand 63 nm were tested by in-situ micropillar compression inside a scanning\nelectron microscope Yield strength, strain-rate sensitivities and activation\nvolumes were obtained from the pillar compression tests. The deformed\nmicropillars were examined under scanning and transmission electron microscopy\nin order to examine the deformation mechanisms active for different layer\nthicknesses and temperatures. The analysis suggests that room temperature\ndeformation was determined by dislocation glide at larger layer thicknesses and\ninterface-related mechanisms at the thinner layer thicknesses. The high\ntemperature compression tests, in contrast, revealed superior thermo-mechanical\nstability and strength retention for the NMMs with larger layer thicknesses\nwith deformation controlled by dislocation glide. A remarkable transition in\ndeformation mechanism occurred as the layer thickness decreased, to a\ndeformation response controlled by diffusion processes along the interfaces,\nwhich resulted in temperature-induced softening. A deformation mechanism map,\nin terms of layer thickness and temperature, is proposed from the results\nobtained in this investigation.\n",
"title": "Deformation mechanism map of Cu/Nb nanoscale metallic multilayers as a function of temperature and layer thickness"
}
| null | null | null | null | true | null |
10239
| null |
Default
| null | null |
null |
{
"abstract": " We give necessary and sufficient conditions for the embeddings\n$\\Lambda\\text{BV}^{(p)}\\subseteq \\Gamma\\text{BV}^{(q_n\\uparrow q)}$ and\n$\\Phi\\text{BV}\\subseteq\\text{BV}^{(q_n\\uparrow q)}$. As a consequence, a number\nof results in the literature, including a fundamental theorem of Perlman and\nWaterman, are simultaneously extended.\n",
"title": "Relations between Schramm spaces and generalized Wiener classes"
}
| null | null | null | null | true | null |
10240
| null |
Default
| null | null |
null |
{
"abstract": " In variable or graph selection problems, finding a right-sized model or\ncontrolling the number of false positives is notoriously difficult. Recently, a\nmeta-algorithm called Stability Selection was proposed that can provide\nreliable finite-sample control of the number of false positives. Its benefits\nwere demonstrated when used in conjunction with the lasso and orthogonal\nmatching pursuit algorithms.\nIn this paper, we investigate the applicability of stability selection to\nstructured selection algorithms: the group lasso and the structured\ninput-output lasso. We find that using stability selection often increases the\npower of both algorithms, but that the presence of complex structure reduces\nthe reliability of error control under stability selection. We give strategies\nfor setting tuning parameters to obtain a good model size under stability\nselection, and highlight its strengths and weaknesses compared to competing\nmethods screen and clean and cross-validation. We give guidelines about when to\nuse which error control method.\n",
"title": "Stability Selection for Structured Variable Selection"
}
| null | null | null | null | true | null |
10241
| null |
Default
| null | null |
null |
{
"abstract": " We propose a definition of Vafa-Witten invariants counting semistable Higgs\npairs on a polarised surface. We use virtual localisation applied to\nMochizuki/Joyce-Song pairs.\nFor $K_S\\le0$ we expect our definition coincides with an alternative\ndefinition using weighted Euler characteristics. We prove this for deg $K_S<0$\nhere, and it is proved for $S$ a K3 surface in \\cite{MT}.\nFor K3 surfaces we calculate the invariants in terms of modular forms which\ngeneralise and prove conjectures of Vafa and Witten.\n",
"title": "Vafa-Witten invariants for projective surfaces II: semistable case"
}
| null | null | null | null | true | null |
10242
| null |
Default
| null | null |
null |
{
"abstract": " The aggregation of many independent estimates can outperform the most\naccurate individual judgment. This centenarian finding, popularly known as the\nwisdom of crowds, has been applied to problems ranging from the diagnosis of\ncancer to financial forecasting. It is widely believed that social influence\nundermines collective wisdom by reducing the diversity of opinions within the\ncrowd. Here, we show that if a large crowd is structured in small independent\ngroups, deliberation and social influence within groups improve the crowd's\ncollective accuracy. We asked a live crowd (N=5180) to respond to\ngeneral-knowledge questions (e.g., what is the height of the Eiffel Tower?).\nParticipants first answered individually, then deliberated and made consensus\ndecisions in groups of five, and finally provided revised individual estimates.\nWe found that averaging consensus decisions was substantially more accurate\nthan aggregating the initial independent opinions. Remarkably, combining as few\nas four consensus choices outperformed the wisdom of thousands of individuals.\n",
"title": "Aggregated knowledge from a small number of debates outperforms the wisdom of large crowds"
}
| null | null | null | null | true | null |
10243
| null |
Default
| null | null |
null |
{
"abstract": " In temperate climates, mortality is seasonal with a winter-dominant pattern,\ndue in part to pneumonia and influenza. Cardiac causes, which are the leading\ncause of death in the United States, are also winter-seasonal although it is\nnot clear why. Interactions between circulating respiratory viruses (f.e.,\ninfluenza) and cardiac conditions have been suggested as a cause of\nwinter-dominant mortality patterns. We propose and implement a way to estimate\nan upper bound on mortality attributable to winter-dominant viruses like\ninfluenza. We calculate 'pseudo-seasonal' life expectancy, dividing the year\ninto two six-month spans, one encompassing winter the other summer. During the\nsummer when the circulation of respiratory viruses is drastically reduced, life\nexpectancy is about one year longer. We also quantify the seasonal mortality\ndifference in terms of seasonal \"equivalent ages\" (defined herein) and\nproportional hazards. We suggest that even if viruses cause excess winter\ncardiac mortality, the population-level mortality reduction of a perfect\ninfluenza vaccine would be much more modest than is often recognized.\n",
"title": "Summertime, and the livin is easy: Winter and summer pseudoseasonal life expectancy in the United States"
}
| null | null |
[
"Statistics"
] | null | true | null |
10244
| null |
Validated
| null | null |
null |
{
"abstract": " We address the problem of verifying the satisfiability of Constrained Horn\nClauses (CHCs) based on theories of inductively defined data structures, such\nas lists and trees. We propose a transformation technique whose objective is\nthe removal of these data structures from CHCs, hence reducing their\nsatisfiability to a satisfiability problem for CHCs on integers and booleans.\nWe propose a transformation algorithm and identify a class of clauses where it\nalways succeeds. We also consider an extension of that algorithm, which\ncombines clause transformation with reasoning on integer constraints. Via an\nexperimental evaluation we show that our technique greatly improves the\neffectiveness of applying the Z3 solver to CHCs. We also show that our\nverification technique based on CHC transformation followed by CHC solving, is\ncompetitive with respect to CHC solvers extended with induction. This paper is\nunder consideration for acceptance in TPLP.\n",
"title": "Solving Horn Clauses on Inductive Data Types Without Induction"
}
| null | null | null | null | true | null |
10245
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the power of non-determinism in purely functional programming\nlanguages with higher-order types. Specifically, we consider cons-free programs\nof varying data orders, equipped with explicit non-deterministic choice.\nCons-freeness roughly means that data constructors cannot occur in function\nbodies and all manipulation of storage space thus has to happen indirectly\nusing the call stack.\nWhile cons-free programs have previously been used by several authors to\ncharacterise complexity classes, the work on non-deterministic programs has\nalmost exclusively considered programs of data order 0. Previous work has shown\nthat adding explicit non-determinism to cons-free programs taking data of order\n0 does not increase expressivity; we prove that this - dramatically - is not\nthe case for higher data orders: adding non-determinism to programs with data\norder at least 1 allows for a characterisation of the entire class of\nelementary-time decidable sets.\nFinally we show how, even with non-deterministic choice, the original\nhierarchy of characterisations is restored by imposing different restrictions.\n",
"title": "The Power of Non-Determinism in Higher-Order Implicit Complexity"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10246
| null |
Validated
| null | null |
null |
{
"abstract": " The Advanced Virgo detector uses two monolithic optical cavities at its\noutput port to suppress higher order modes and radio frequency sidebands from\nthe carrier light used for gravitational wave detection. These two cavities in\nseries form the output mode cleaner. We present a measured upper limit on the\nlength noise of these cavities that is consistent with the thermo-refractive\nnoise prediction of $8 \\times 10^{-16}\\,\\textrm{m/Hz}^{1/2}$ at 15 Hz. The\ncavity length is controlled using Peltier cells and piezo-electric actuators to\nmaintain resonance on the incoming light. A length lock precision of $3.5\n\\times 10^{-13}\\,\\textrm{m}$ is achieved. These two results are combined to\ndemonstrate that the broadband length noise of the output mode cleaner in the\n10-60 Hz band is at least a factor 10 below other expected noise sources in the\nAdvanced Virgo detector design configuration.\n",
"title": "Upper-limit on the Advanced Virgo output mode cleaner cavity length noise"
}
| null | null |
[
"Physics"
] | null | true | null |
10247
| null |
Validated
| null | null |
null |
{
"abstract": " We use a mobile impurity or depleton model to study elementary excitations in\none-dimensional integrable systems. For Lieb-Liniger and bosonic Yang-Gaudin\nmodels we express two phenomenological parameters characterising renormalised\ninter- actions of mobile impurities with superfluid background: the number of\ndepleted particles, $N$ and the superfluid phase drop $\\pi J$ in terms of the\ncorresponding Bethe Ansatz solution and demonstrate, in the leading order, the\nabsence of two-phonon scattering resulting in vanishing rates of inelastic\nprocesses such as viscosity experienced by the mobile impurities\n",
"title": "Mobile impurities in integrable models"
}
| null | null | null | null | true | null |
10248
| null |
Default
| null | null |
null |
{
"abstract": " We study the applicability of the time-dependent variational principle in\nmatrix product state manifolds for the long time description of quantum\ninteracting systems. By studying integrable and nonintegrable systems for which\nthe long time dynamics are known we demonstrate that convergence of long time\nobservables is subtle and needs to be examined carefully. Remarkably, for the\ndisordered nonintegrable system we consider the long time dynamics are in good\nagreement with the rigorously obtained short time behavior and with previous\nobtained numerically exact results, suggesting that at least in this case the\napparent convergence of this approach is reliable. Our study indicates that\nwhile great care must be exercised in establishing the convergence of the\nmethod, it may still be asymptotically accurate for a class of disordered\nnonintegrable quantum systems.\n",
"title": "Time-dependent variational principle in matrix-product state manifolds: pitfalls and potential"
}
| null | null | null | null | true | null |
10249
| null |
Default
| null | null |
null |
{
"abstract": " The paper presents a topology optimization approach that designs an optimal\nstructure, called a self-supporting structure, which is ready to be fabricated\nvia additive manufacturing without the usage of additional support structures.\nSuch supports in general have to be created during the fabricating process so\nthat the primary object can be manufactured layer by layer without collapse,\nwhich is very time-consuming and waste of material.\nThe proposed approach resolves this problem by formulating the\nself-supporting requirements as a novel explicit quadratic continuous\nconstraint in the topology optimization problem, or specifically, requiring the\nnumber of unsupported elements (in terms of the sum of squares of their\ndensities) to be zero. Benefiting form such novel formulations, computing\nsensitivity of the self-supporting constraint with respect to the design\ndensity is straightforward, which otherwise would require lots of research\nefforts in general topology optimization studies. The derived sensitivity for\neach element is only linearly dependent on its sole density, which, different\nfrom previous layer-based sensitivities, consequently allows for a parallel\nimplementation and possible higher convergence rate. In addition, a discrete\nconvolution operator is also designed to detect the unsupported elements as\ninvolved in each step of optimization iteration, and improves the detection\nprocess 100 times as compared with simply enumerating these elements. The\napproach works for cases of general overhang angle, or general domain, and\nproduces an optimized structures, and their associated optimal compliance, very\nclose to that of the reference structure obtained without considering the\nself-supporting constraint, as demonstrated by extensive 2D and 3D benchmark\nexamples.\n",
"title": "Self-supporting Topology Optimization for Additive Manufacturing"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10250
| null |
Validated
| null | null |
null |
{
"abstract": " Reward shaping is one of the most effective methods to tackle the crucial yet\nchallenging problem of credit assignment in Reinforcement Learning (RL).\nHowever, designing shaping functions usually requires much expert knowledge and\nhand-engineering, and the difficulties are further exacerbated given multiple\nsimilar tasks to solve. In this paper, we consider reward shaping on a\ndistribution of tasks, and propose a general meta-learning framework to\nautomatically learn the efficient reward shaping on newly sampled tasks,\nassuming only shared state space but not necessarily action space. We first\nderive the theoretically optimal reward shaping in terms of credit assignment\nin model-free RL. We then propose a value-based meta-learning algorithm to\nextract an effective prior over the optimal reward shaping. The prior can be\napplied directly to new tasks, or provably adapted to the task-posterior while\nsolving the task within few gradient updates. We demonstrate the effectiveness\nof our shaping through significantly improved learning efficiency and\ninterpretable visualizations across various settings, including notably a\nsuccessful transfer from DQN to DDPG.\n",
"title": "Reward Shaping via Meta-Learning"
}
| null | null | null | null | true | null |
10251
| null |
Default
| null | null |
null |
{
"abstract": " Among recently introduced new notions in real algebraic geometry is that of\nregulous functions. Such functions form a foundation for the development of\nregulous geometry. Several interesting results on regulous varieties and\nregulous sheaves are already available. In this paper, we define and\ninvestigate regulous vector bundles. We establish algebraic and geometric\nproperties of such vector bundles, and identify them with stratified-algebraic\nvector bundles. Furthermore, using new results on curve-rational functions, we\ncharacterize regulous vector bundles among families of vector spaces\nparametrized by an affine regulous variety. We also study relationships between\nregulous and topological vector bundles.\n",
"title": "Regulous vector bundles"
}
| null | null | null | null | true | null |
10252
| null |
Default
| null | null |
null |
{
"abstract": " We revise the operator-norm convergence of the Trotter product formula for a\npair {A,B} of generators of semigroups on a Banach space. Operator-norm\nconvergence holds true if the dominating operator A generates a holomorphic\ncontraction semigroup and B is a A-infinitesimally small generator of a\ncontraction semigroup, in particular, if B is a bounded operator. Inspired by\nstudies of evolution semigroups it is shown in the present paper that the\noperator-norm convergence generally fails even for bounded operators B if A is\nnot a holomorphic generator. Moreover, it is shown that operator norm\nconvergence of the Trotter product formula can be arbitrary slow.\n",
"title": "Remarks on the operator-norm convergence of the Trotter product formula"
}
| null | null | null | null | true | null |
10253
| null |
Default
| null | null |
null |
{
"abstract": " Nowadays data compressors are applied to many problems of text analysis, but\nmany such applications are developed outside of the framework of mathematical\nstatistics. In this paper we overcome this obstacle and show how several\nmethods of classical mathematical statistics can be developed based on\napplications of the data compressors.\n",
"title": "Using data-compressors for statistical analysis of problems on homogeneity testing and classification"
}
| null | null | null | null | true | null |
10254
| null |
Default
| null | null |
null |
{
"abstract": " Two-dimensional embeddings remain the dominant approach to visualize high\ndimensional data. The choice of embeddings ranges from highly non-linear ones,\nwhich can capture complex relationships but are difficult to interpret\nquantitatively, to axis-aligned projections, which are easy to interpret but\nare limited to bivariate relationships. Linear project can be considered as a\ncompromise between complexity and interpretability, as they allow explicit axes\nlabels, yet provide significantly more degrees of freedom compared to\naxis-aligned projections. Nevertheless, interpreting the axes directions, which\nare linear combinations often with many non-trivial components, remains\ndifficult. To address this problem we introduce a structure aware decomposition\nof (multiple) linear projections into sparse sets of axis aligned projections,\nwhich jointly capture all information of the original linear ones. In\nparticular, we use tools from Dempster-Shafer theory to formally define how\nrelevant a given axis aligned project is to explain the neighborhood relations\ndisplayed in some linear projection. Furthermore, we introduce a new approach\nto discover a diverse set of high quality linear projections and show that in\npractice the information of $k$ linear projections is often jointly encoded in\n$\\sim k$ axis aligned plots. We have integrated these ideas into an interactive\nvisualization system that allows users to jointly browse both linear\nprojections and their axis aligned representatives. Using a number of case\nstudies we show how the resulting plots lead to more intuitive visualizations\nand new insight.\n",
"title": "Exploring High-Dimensional Structure via Axis-Aligned Decomposition of Linear Projections"
}
| null | null | null | null | true | null |
10255
| null |
Default
| null | null |
null |
{
"abstract": " Graph matching or quadratic assignment, is the problem of labeling the\nvertices of two graphs so that they are as similar as possible. A common method\nfor approximately solving the NP-hard graph matching problem is relaxing it to\na convex optimization problem over the set of doubly stochastic (DS) matrices.\nRecent analysis has shown that for almost all pairs of isomorphic and\nasymmetric graphs, the DS relaxation succeeds in correctly retrieving the\nisomorphism between the graphs. Our goal in this paper is to analyze the case\nof symmetric isomorphic graphs. This goal is motivated by shape matching\napplications where the graphs of interest usually have reflective symmetry.\nFor symmetric problems the graph matching problem has multiple isomorphisms\nand so convex relaxations admit all convex combinations of these isomorphisms\nas viable solutions. If the convex relaxation does not admit any additional\nsuperfluous solution we say that it is convex exact. In this case there are\ntractable algorithms to retrieve an isomorphism from the convex relaxation.\nWe show that convex exactness depends strongly on the symmetry group of the\ngraphs; For a fixed symmetry group $G$, either the DS relaxation will be convex\nexact for almost all pairs of isomorphic graphs with symmetry group $G$, or the\nDS relaxation will fail for all such pairs. We show that for reflective groups\nwith at least one full orbit convex exactness holds almost everywhere, and\nprovide some simple examples of non-reflective symmetry groups for which convex\nexactness always fails.\nWhen convex exactness holds, the isomorphisms of the graphs are the extreme\npoints of the convex solution set. We suggest an efficient algorithm for\nretrieving an isomorphism in this case. We also show that the \"convex to\nconcave\" projection method will also retrieve an isomorphism in this case.\n",
"title": "Exact Recovery with Symmetries for the Doubly-Stochastic Relaxation"
}
| null | null | null | null | true | null |
10256
| null |
Default
| null | null |
null |
{
"abstract": " Extensive cooperation among unrelated individuals is unique to humans, who\noften sacrifice personal benefits for the common good and work together to\nachieve what they are unable to execute alone. The evolutionary success of our\nspecies is indeed due, to a large degree, to our unparalleled other-regarding\nabilities. Yet, a comprehensive understanding of human cooperation remains a\nformidable challenge. Recent research in social science indicates that it is\nimportant to focus on the collective behavior that emerges as the result of the\ninteractions among individuals, groups, and even societies. Non-equilibrium\nstatistical physics, in particular Monte Carlo methods and the theory of\ncollective behavior of interacting particles near phase transition points, has\nproven to be very valuable for understanding counterintuitive evolutionary\noutcomes. By studying models of human cooperation as classical spin models, a\nphysicist can draw on familiar settings from statistical physics. However,\nunlike pairwise interactions among particles that typically govern solid-state\nphysics systems, interactions among humans often involve group interactions,\nand they also involve a larger number of possible states even for the most\nsimplified description of reality. The complexity of solutions therefore often\nsurpasses that observed in physical systems. Here we review experimental and\ntheoretical research that advances our understanding of human cooperation,\nfocusing on spatial pattern formation, on the spatiotemporal dynamics of\nobserved solutions, and on self-organization that may either promote or hinder\nsocially favorable states.\n",
"title": "Statistical physics of human cooperation"
}
| null | null | null | null | true | null |
10257
| null |
Default
| null | null |
null |
{
"abstract": " We study Bayesian hypernetworks: a framework for approximate Bayesian\ninference in neural networks. A Bayesian hypernetwork $\\h$ is a neural network\nwhich learns to transform a simple noise distribution, $p(\\vec\\epsilon) =\n\\N(\\vec 0,\\mat I)$, to a distribution $q(\\pp) := q(h(\\vec\\epsilon))$ over the\nparameters $\\pp$ of another neural network (the \"primary network\")\\@. We train\n$q$ with variational inference, using an invertible $\\h$ to enable efficient\nestimation of the variational lower bound on the posterior $p(\\pp | \\D)$ via\nsampling. In contrast to most methods for Bayesian deep learning, Bayesian\nhypernets can represent a complex multimodal approximate posterior with\ncorrelations between parameters, while enabling cheap iid sampling of~$q(\\pp)$.\nIn practice, Bayesian hypernets can provide a better defense against\nadversarial examples than dropout, and also exhibit competitive performance on\na suite of tasks which evaluate model uncertainty, including regularization,\nactive learning, and anomaly detection.\n",
"title": "Bayesian Hypernetworks"
}
| null | null | null | null | true | null |
10258
| null |
Default
| null | null |
null |
{
"abstract": " Plate motions are governed by equilibrium between basal and edge forces.\nGreat earthquakes may induce differential static stress changes across tectonic\nplates, enabling a new equilibrium state. Here we consider the torque balance\nfor idealized circular plates and find a simple scalar relationship for changes\nin relative plate speed as a function of its size, upper mantle viscosity, and\ncoseismic stress changes. Applied to Japan, the 2011\n$\\mathrm{M}_{\\mathrm{W}}=9.0$ Tohoku earthquake generated coseismic stresses of\n$10^2-10^5$~Pa that could have induced changes in motion of small (radius\n$\\sim100$~km) crustal blocks within Honshu. Analysis of time-dependent GPS\nvelocities, with corrections for earthquake cycle effects, reveals that plate\nspeeds may have changed by up to $\\sim3$ mm/yr between $\\sim3.75$-year epochs\nbracketing this earthquake, consistent with an upper mantle viscosity of $\\sim\n5\\times10^{18}$Pa$\\cdot$s, suggesting that great earthquakes may modulate\nmotions of proximal crustal blocks at frequencies as high as $10^-8$~Hz.\n",
"title": "Block Motion Changes in Japan Triggered by the 2011 Great Tohoku Earthquake"
}
| null | null | null | null | true | null |
10259
| null |
Default
| null | null |
null |
{
"abstract": " One of the most challenging tasks when adopting Bayesian Networks (BNs) is\nthe one of learning their structure from data. This task is complicated by the\nhuge search space of possible solutions, and by the fact that the problem is\nNP-hard. Hence, full enumeration of all the possible solutions is not always\nfeasible and approximations are often required. However, to the best of our\nknowledge, a quantitative analysis of the performance and characteristics of\nthe different heuristics to solve this problem has never been done before.\nFor this reason, in this work, we provide a detailed comparison of many\ndifferent state-of-the-arts methods for structural learning on simulated data\nconsidering both BNs with discrete and continuous variables, and with different\nrates of noise in the data. In particular, we investigate the performance of\ndifferent widespread scores and algorithmic approaches proposed for the\ninference and the statistical pitfalls within them.\n",
"title": "Learning the structure of Bayesian Networks: A quantitative assessment of the effect of different algorithmic schemes"
}
| null | null | null | null | true | null |
10260
| null |
Default
| null | null |
null |
{
"abstract": " We simulate the stresses induced by temperature changes in a putative hard\nlayer near the surface of comet 67P/Churyumov--Gerasimenko with a\nthermo-viscoelastic model. Such a layer could be formed by the recondensation\nor sintering of water ice (and dust grains), as suggested by laboratory\nexperiments and computer simulations, and would explain the high compressive\nstrength encountered by experiments on board the Philae lander. Changes in\ntemperature from seasonal insolation variation penetrate into the comet's\nsurface to depths controlled by the thermal inertia, causing the material to\nexpand and contract. Modelling this with a Maxwellian viscoelastic response on\na spherical nucleus, we show that a hard, icy layer with similar properties to\nMartian permafrost will experience high stresses: up to tens of MPa, which\nexceed its material strength (a few MPa), down to depths of centimetres to a\nmetre. The stress distribution with latitude is confirmed qualitatively when\ntaking into account the comet's complex shape but neglecting thermal inertia.\nStress is found to be comparable to the material strength everywhere for\nsufficient thermal inertia ($\\gtrsim50$ J m$^{-2}$ K$^{-1}$ s$^{-1/2}$) and ice\ncontent ($\\gtrsim 45\\%$ at the equator). In this case, stresses penetrate to a\ntypical depth of $\\sim0.25$ m, consistent with the detection of metre-scale\nthermal contraction crack polygons all over the comet. Thermal fracturing may\nbe an important erosion process on cometary surfaces which breaks down material\nand weakens cliffs.\n",
"title": "Thermal fracturing on comets. Applications to 67P/Churyumov-Gerasimenko"
}
| null | null |
[
"Physics"
] | null | true | null |
10261
| null |
Validated
| null | null |
null |
{
"abstract": " Forwarding data by name has been assumed to be a necessary aspect of an\ninformation-centric redesign of the current Internet architecture that makes\ncontent access, dissemination, and storage more efficient. The Named Data\nNetworking (NDN) and Content-Centric Networking (CCNx) architectures are the\nleading examples of such an approach. However, forwarding data by name incurs\nstorage and communication complexities that are orders of magnitude larger than\nsolutions based on forwarding data using addresses. Furthermore, the specific\nalgorithms used in NDN and CCNx have been shown to have a number of\nlimitations. The Addressable Data Networking (ADN) architecture is introduced\nas an alternative to NDN and CCNx. ADN is particularly attractive for\nlarge-scale deployments of the Internet of Things (IoT), because it requires\nfar less storage and processing in relaying nodes than NDN. ADN allows things\nand data to be denoted by names, just like NDN and CCNx do. However, instead of\nreplacing the waist of the Internet with named-data forwarding, ADN uses an\naddress-based forwarding plane and introduces an information plane that\nseamlessly maps names to addresses without the involvement of end-user\napplications. Simulation results illustrate the order of magnitude savings in\ncomplexity that can be attained with ADN compared to NDN.\n",
"title": "ADN: An Information-Centric Networking Architecture for the Internet of Things"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10262
| null |
Validated
| null | null |
null |
{
"abstract": " Delay Tolerant Networking (DTN) is an approach to networking which handles\nnetwork disruptions and high delays that may occur in many kinds of\ncommunication networks. The major reasons for high delay include partial\nconnectivity of networks as can be seen in many types of ad hoc wireless\nnetworks with frequent network partitions, long propagation time as experienced\nin inter-planetary and deep space networks, and frequent link disruptions due\nto the mobility of nodes as observed in terrestrial wireless network\nenvironments. Experimenting network architectures, protocols, and mobility\nmodels in such real-world scenarios is difficult due to the complexities\ninvolved in the network environment. Therefore, in this document, we present\nthe documentation of an Urban Delay Tolerant Network Simulator (UDTNSim)\nversion 0.1, capable of simulating urban road network environments with DTN\ncharacteristics including mobility models and routing protocols. The mobility\nmodels included in this version of UDTNSim are (i) Stationary Movement, (ii)\nSimple Random Movement, (iii) Path Type Based Movememt, (iv) Path Memory Based\nMovement, (v) Path Type with Restricted Movement, and (vi) Path Type with Wait\nMovement. In addition to mobility models, we also provide three routing and\ndata hand-off protocols: (i) Epidemic Routing, (ii) Superior Only Handoff, and\n(iii) Superior Peer Handoff. UDTNSim v0.1 is designed using object-oriented\nprogramming approach in order to provide flexibility in addition of new\nfeatures to the DTN environment. UDTNSim v0.1 is distributed as an open source\nsimulator for the use of the research community.\n",
"title": "Urban Delay Tolerant Network Simulator (UDTNSim v0.1)"
}
| null | null | null | null | true | null |
10263
| null |
Default
| null | null |
null |
{
"abstract": " Regularization techniques such as the lasso (Tibshirani 1996) and elastic net\n(Zou and Hastie 2005) can be used to improve regression model coefficient\nestimation and prediction accuracy, as well as to perform variable selection.\nOrdinal regression models are widely used in applications where the use of\nregularization could be beneficial; however, these models are not included in\nmany popular software packages for regularized regression. We propose a\ncoordinate descent algorithm to fit a broad class of ordinal regression models\nwith an elastic net penalty. Furthermore, we demonstrate that each model in\nthis class generalizes to a more flexible form, for instance to accommodate\nunordered categorical data. We introduce an elastic net penalty class that\napplies to both model forms. Additionally, this penalty can be used to shrink a\nnon-ordinal model toward its ordinal counterpart. Finally, we introduce the R\npackage ordinalNet, which implements the algorithm for this model class.\n",
"title": "Regularized Ordinal Regression and the ordinalNet R Package"
}
| null | null | null | null | true | null |
10264
| null |
Default
| null | null |
null |
{
"abstract": " In multi-object tracking applications, model parameter tuning is a\nprerequisite for reliable performance. In particular, it is difficult to know\nstatistics of false measurements due to various sensing conditions and changes\nin the field of views. In this paper we are interested in designing a\nmulti-object tracking algorithm that handles unknown false measurement rate.\nRecently proposed robust multi-Bernoulli filter is employed for clutter\nestimation while generalized labeled multi-Bernoulli filter is considered for\ntarget tracking. Performance evaluation with real videos demonstrates the\neffectiveness of the tracking algorithm for real-world scenarios.\n",
"title": "Visual Multiple-Object Tracking for Unknown Clutter Rate"
}
| null | null |
[
"Computer Science"
] | null | true | null |
10265
| null |
Validated
| null | null |
null |
{
"abstract": " We consider a general monotone regression estimation where we allow for\nindependent and dependent regressors. We propose a modification of the\nclassical isotonic least squares estimator and establish its rate of\nconvergence for the integrated $L_1$-loss function. The methodology captures\nthe shape of the data without assuming additivity or a parametric form for the\nregression function. Furthermore, the degree of smoothing is chosen\nautomatically and no auxiliary tuning is required for the theoretical analysis.\nSome simulations and two real data illustrations complement the study of the\nproposed estimator.\n",
"title": "On Integrated $L^{1}$ Convergence Rate of an Isotonic Regression Estimator for Multivariate Observations"
}
| null | null | null | null | true | null |
10266
| null |
Default
| null | null |
null |
{
"abstract": " Distributed network optimization has been studied for well over a decade.\nHowever, we still do not have a good idea of how to design schemes that can\nsimultaneously provide good performance across the dimensions of utility\noptimality, convergence speed, and delay. To address these challenges, in this\npaper, we propose a new algorithmic framework with all these metrics\napproaching optimality. The salient features of our new algorithm are\nthree-fold: (i) fast convergence: it converges with only $O(\\log(1/\\epsilon))$\niterations that is the fastest speed among all the existing algorithms; (ii)\nlow delay: it guarantees optimal utility with finite queue length; (iii) simple\nimplementation: the control variables of this algorithm are based on virtual\nqueues that do not require maintaining per-flow information. The new technique\nbuilds on a kind of inexact Uzawa method in the Alternating Directional Method\nof Multiplier, and provides a new theoretical path to prove global and linear\nconvergence rate of such a method without requiring the full rank assumption of\nthe constraint matrix.\n",
"title": "Towards Fast-Convergence, Low-Delay and Low-Complexity Network Optimization"
}
| null | null | null | null | true | null |
10267
| null |
Default
| null | null |
null |
{
"abstract": " The purpose of this note is to revive in $L^p$ spaces the original A. Markov\nideas to study monotonicity of zeros of orthogonal polynomials. This allows us\nto prove and improve in a simple and unified way our previous result [Electron.\nTrans. Numer. Anal., 44 (2015), pp. 271-280] concerning the discrete version of\nA. Markov's theorem on monotonicity of zeros.\n",
"title": "On zeros of polynomials in best $L^p$-approximation and inserting mass points"
}
| null | null | null | null | true | null |
10268
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we apply empirical likelihood method to inference for the\nregression parameters in the partial functional linear regression models based\non B spline. We prove that the empirical log likelihood ratio for the\nregression parameters converges in law to a weighted sum of independent chi\nsquare distributions and run simulations to assess the finite sample\nperformance of our method.\n",
"title": "Empirical likelihood inference for partial functional linear regression models based on B spline"
}
| null | null |
[
"Statistics"
] | null | true | null |
10269
| null |
Validated
| null | null |
null |
{
"abstract": " To obtain uncertainty estimates with real-world Bayesian deep learning\nmodels, practical inference approximations are needed. Dropout variational\ninference (VI) for example has been used for machine vision and medical\napplications, but VI can severely underestimates model uncertainty.\nAlpha-divergences are alternative divergences to VI's KL objective, which are\nable to avoid VI's uncertainty underestimation. But these are hard to use in\npractice: existing techniques can only use Gaussian approximating\ndistributions, and require existing models to be changed radically, thus are of\nlimited use for practitioners. We propose a re-parametrisation of the\nalpha-divergence objectives, deriving a simple inference technique which,\ntogether with dropout, can be easily implemented with existing models by simply\nchanging the loss of the model. We demonstrate improved uncertainty estimates\nand accuracy compared to VI in dropout networks. We study our model's epistemic\nuncertainty far away from the data using adversarial images, showing that these\ncan be distinguished from non-adversarial images by examining our model's\nuncertainty.\n",
"title": "Dropout Inference in Bayesian Neural Networks with Alpha-divergences"
}
| null | null | null | null | true | null |
10270
| null |
Default
| null | null |
null |
{
"abstract": " In this article, we propose two classes of semiparametric mixture regression\nmodels with single-index for model based clustering. Unlike many\nsemiparametric/nonparametric mixture regression models that can only be applied\nto low dimensional predictors, the new semiparametric models can easily\nincorporate high dimensional predictors into the nonparametric components. The\nproposed models are very general, and many of the recently proposed\nsemiparametric/nonparametric mixture regression models are indeed special cases\nof the new models. Backfitting estimates and the corresponding modified EM\nalgorithms are proposed to achieve optimal convergence rates for both\nparametric and nonparametric parts. We establish the identifiability results of\nthe proposed two models and investigate the asymptotic properties of the\nproposed estimation procedures. Simulation studies are conducted to demonstrate\nthe finite sample performance of the proposed models. An application of NBA\ndata by new models reveals some new findings.\n",
"title": "Semiparametric Mixtures of Regressions with Single-index for Model Based Clustering"
}
| null | null | null | null | true | null |
10271
| null |
Default
| null | null |
null |
{
"abstract": " We apply a convolutional neural network (CNN) to classify and detect quasars\nin the Sloan Digital Sky Survey Stripe 82 and also to predict the photometric\nredshifts of quasars. The network takes the variability of objects into account\nby converting light curves into images. The width of the images, noted w,\ncorresponds to the five magnitudes ugriz and the height of the images, noted h,\nrepresents the date of the observation. The CNN provides good results since its\nprecision is 0.988 for a recall of 0.90, compared to a precision of 0.985 for\nthe same recall with a random forest classifier. Moreover 175 new quasar\ncandidates are found with the CNN considering a fixed recall of 0.97. The\ncombination of probabilities given by the CNN and the random forest makes good\nperformance even better with a precision of 0.99 for a recall of 0.90.\nFor the redshift predictions, the CNN presents excellent results which are\nhigher than those obtained with a feature extraction step and different\nclassifiers (a K-nearest-neighbors, a support vector machine, a random forest\nand a gaussian process classifier). Indeed, the accuracy of the CNN within\n|\\Delta z|<0.1 can reach 78.09%, within |\\Delta z|<0.2 reaches 86.15%, within\n|\\Delta z|<0.3 reaches 91.2% and the value of rms is 0.359. The performance of\nthe KNN decreases for the three |\\Delta z| regions, since within the accuracy\nof |\\Delta z|<0.1, |\\Delta z|<0.2 and |\\Delta z|<0.3 is 73.72%, 82.46% and\n90.09% respectively, and the value of rms amounts to 0.395. So the CNN\nsuccessfully reduces the dispersion and the catastrophic redshifts of quasars.\nThis new method is very promising for the future of big databases like the\nLarge Synoptic Survey Telescope.\n",
"title": "Deep learning Approach for Classifying, Detecting and Predicting Photometric Redshifts of Quasars in the Sloan Digital Sky Survey Stripe 82"
}
| null | null |
[
"Physics"
] | null | true | null |
10272
| null |
Validated
| null | null |
null |
{
"abstract": " We analyze the source of inter-model scatter in the surface temperature\nresponse to quadrupling CO2 in two sets of GCM simulations from the Tropical\nRain Belts with an Annual cycle and a Continent Model Intercomparison Project\n(TRACMIP; Voigt et al, 2016). TRACMIP provides simulations of idealized\nclimates that allow for studying the fundamental dynamics of tropical rainfall\nand its response to climate change. One configuration is an aquaplanet\natmosphere (i.e., with zonally-symmetric boundary conditions) coupled to a slab\nocean (AquaCTL and Aqua4x). The other includes an equatorial continent\nrepresented by a thin slab ocean with increased surface albedo and decreased\nevaporation (LandCTL and Land4x).\n",
"title": "Sources of inter-model scatter in TRACMIP, the Tropical Rain belts with an Annual cycle and a Continent Model Intercomparison Project"
}
| null | null | null | null | true | null |
10273
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a novel representation of decomposable graphs based on\nsemi-latent tree-dependent bipartite graphs. The novel representation has two\nmain benefits. First, it enables a form of sub-clustering within maximal\ncliques of the graph, adding informational richness to the general use of\ndecomposable graphs that could be harnessed in applications with behavioural\ntype of data. Second, it allows for a new node-driven Markov chain Monte Carlo\nsampler of decomposable graphs that can easily parallelize and scale. The\nproposed sampler also benefits from the computational efficiency of\njunction-tree-based samplers of decomposable graphs.\n",
"title": "Sub-clustering in decomposable graphs and size-varying junction trees"
}
| null | null | null | null | true | null |
10274
| null |
Default
| null | null |
null |
{
"abstract": " The causal inference problem consists in determining whether a probability\ndistribution over a set of observed variables is compatible with a given causal\nstructure. In [arXiv:1609.00672], one of us introduced a hierarchy of necessary\nlinear programming constraints which all the observed distributions compatible\nwith the considered causal structure must satisfy. In this work, we prove that\nthe inflation hierarchy is complete, i.e., any distribution of the observed\nvariables which does not admit a realization within the considered causal\nstructure will fail one of the inflation tests. More quantitatively, we show\nthat any distribution of measurable events satisfying the $n^{th}$ inflation\ntest is $O\\left(\\frac{1}{\\sqrt{n}}\\right)$-close in Euclidean norm to a\ndistribution realizable within the given causal structure. In addition, we show\nthat the corresponding $n^{th}$-order relaxation of the dual problem consisting\nin maximizing a $k^{th}$ degree polynomial on the observed variables is\n$O\\left(\\frac{k^2}{n}\\right)$-close to the optimal solution.\n",
"title": "The inflation technique solves completely the classical inference problem"
}
| null | null | null | null | true | null |
10275
| null |
Default
| null | null |
null |
{
"abstract": " The complexity embedded in condensed matter fertilizes the discovery of new\nstates of matter, enriched by ingredients like frustration. Illustrating\nexamples in magnetic systems are Kitaev spin liquids, skyrmions phases, or spin\nices. These unconventional ground states support exotic excitations, for\nexample the magnetic charges in spin ices, also called monopoles. Beyond their\ndiscovery, an important challenge is to be able to control and manipulate them.\nHere, we propose a new mechanism to inject monopoles in a spin ice through a\nstaggered magnetic field. We show theoretically, and demonstrate experimentally\nin the Ho$_2$Ir$_2$O$_7$ pyrochlore iridate, that it results in the\nstabilization of a monopole crystal, which exhibits magnetic fragmentation. In\nthis new state of matter, the magnetic moment fragments into an ordered part\nand a persistently fluctuating one. Compared to conventional spin ices, the\ndifferent nature of the excitations in this fragmented state opens the way to\nnovel tunable field-induced and dynamical behaviors.\n",
"title": "Magnetic charge injection in spin ice: a new way to fragmentation"
}
| null | null | null | null | true | null |
10276
| null |
Default
| null | null |
null |
{
"abstract": " Robotic systems, working together as a team, are becoming valuable players in\ndifferent real-world applications, from disaster response to warehouse\nfulfillment services. Centralized solutions for coordinating multi-robot teams\noften suffer from poor scalability and vulnerability to communication\ndisruptions. This paper develops a decentralized multi-agent task allocation\n(Dec-MATA) algorithm for multi-robot applications. The task planning problem is\nposed as a maximum-weighted matching of a bipartite graph, the solution of\nwhich using the blossom algorithm allows each robot to autonomously identify\nthe optimal sequence of tasks it should undertake. The graph weights are\ndetermined based on a soft clustering process, which also plays a problem\ndecomposition role seeking to reduce the complexity of the individual-agents'\ntask assignment problems. To evaluate the new Dec-MATA algorithm, a series of\ncase studies (of varying complexity) are performed, with tasks being\ndistributed randomly over an observable 2D environment. A centralized approach,\nbased on a state-of-the-art MILP formulation of the multi-Traveling Salesman\nproblem is used for comparative analysis. While getting within 7-28% of the\noptimal cost obtained by the centralized algorithm, the Dec-MATA algorithm is\nfound to be 1-3 orders of magnitude faster and minimally sensitive to\ntask-to-robot ratios, unlike the centralized algorithm.\n",
"title": "Decentralized Task Allocation in Multi-Robot Systems via Bipartite Graph Matching Augmented with Fuzzy Clustering"
}
| null | null | null | null | true | null |
10277
| null |
Default
| null | null |
null |
{
"abstract": " Model precision in a classification task is highly dependent on the feature\nspace that is used to train the model. Moreover, whether the features are\nsequential or static will dictate which classification method can be applied as\nmost of the machine learning algorithms are designed to deal with either one or\nanother type of data. In real-life scenarios, however, it is often the case\nthat both static and dynamic features are present, or can be extracted from the\ndata. In this work, we demonstrate how generative models such as Hidden Markov\nModels (HMM) and Long Short-Term Memory (LSTM) artificial neural networks can\nbe used to extract temporal information from the dynamic data. We explore how\nthe extracted information can be combined with the static features in order to\nimprove the classification performance. We evaluate the existing techniques and\nsuggest a hybrid approach, which outperforms other methods on several public\ndatasets.\n",
"title": "Combining Static and Dynamic Features for Multivariate Sequence Classification"
}
| null | null | null | null | true | null |
10278
| null |
Default
| null | null |
null |
{
"abstract": " Our goal is to improve variance reducing stochastic methods through better\ncontrol variates. We first propose a modification of SVRG which uses the\nHessian to track gradients over time, rather than to recondition, increasing\nthe correlation of the control variates and leading to faster theoretical\nconvergence close to the optimum. We then propose accurate and computationally\nefficient approximations to the Hessian, both using a diagonal and a low-rank\nmatrix. Finally, we demonstrate the effectiveness of our method on a wide range\nof problems.\n",
"title": "Tracking the gradients using the Hessian: A new look at variance reducing stochastic methods"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
10279
| null |
Validated
| null | null |
null |
{
"abstract": " We consider plasmon resonances and cloaking for the elastostatic system in\n$\\mathbb{R}^3$ via the spectral theory of Neumann-Poincaré operator. We first\nderive the full spectral properties of the Neumann-Poincaré operator for the\n3D elastostatic system in the spherical geometry. The spectral result is of\nsignificant interest for its own sake, and serves as a highly nontrivial\nextension of the corresponding 2D study in [8]. The derivation of the spectral\nresult in 3D involves much more complicated and subtle calculations and\narguments than that for the 2D case. Then we consider a 3D plasmonic structure\nin elastostatics which takes a general core-shell-matrix form with the\nmetamaterial located in the shell. Using the obtained spectral result, we\nprovide an accurate characterisation of the anomalous localised resonance and\ncloaking associated to such a plasmonic structure.\n",
"title": "On spectral properties of Neuman-Poincare operator and plasmonic resonances in 3D elastostatics"
}
| null | null | null | null | true | null |
10280
| null |
Default
| null | null |
null |
{
"abstract": " We introduce a hybridizable discontinuous Galerkin method for the\nincompressible Navier--Stokes equations for which the approximate velocity\nfield is pointwise divergence-free. The method builds on the method presented\nby Labeur and Wells [SIAM J. Sci. Comput., vol. 34 (2012), pp. A889--A913]. We\nshow that with modifications of the function spaces in the method of Labeur and\nWells it is possible to formulate a simple method with pointwise\ndivergence-free velocity fields which is momentum conserving, energy stable,\nand pressure-robust. Theoretical results are supported by two- and\nthree-dimensional numerical examples and for different orders of polynomial\napproximation.\n",
"title": "A hybridizable discontinuous Galerkin method for the Navier--Stokes equations with pointwise divergence-free velocity field"
}
| null | null | null | null | true | null |
10281
| null |
Default
| null | null |
null |
{
"abstract": " Artificial neural networks (ANNs) have gained a well-deserved popularity\namong machine learning tools upon their recent successful applications in\nimage- and sound processing and classification problems. ANNs have also been\napplied for predicting the family or function of a protein, knowing its residue\nsequence. Here we present two new ANNs with multi-label classification ability,\nshowing impressive accuracy when classifying protein sequences into 698 UniProt\nfamilies (AUC=99.99%) and 983 Gene Ontology classes (AUC=99.45%).\n",
"title": "Near Perfect Protein Multi-Label Classification with Deep Neural Networks"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
10282
| null |
Validated
| null | null |
null |
{
"abstract": " Connectionist Temporal Classification has recently attracted a lot of\ninterest as it offers an elegant approach to building acoustic models (AMs) for\nspeech recognition. The CTC loss function maps an input sequence of observable\nfeature vectors to an output sequence of symbols. Output symbols are\nconditionally independent of each other under CTC loss, so a language model\n(LM) can be incorporated conveniently during decoding, retaining the\ntraditional separation of acoustic and linguistic components in ASR. For fixed\nvocabularies, Weighted Finite State Transducers provide a strong baseline for\nefficient integration of CTC AMs with n-gram LMs. Character-based neural LMs\nprovide a straight forward solution for open vocabulary speech recognition and\nall-neural models, and can be decoded with beam search. Finally,\nsequence-to-sequence models can be used to translate a sequence of individual\nsounds into a word string. We compare the performance of these three\napproaches, and analyze their error patterns, which provides insightful\nguidance for future research and development in this important area.\n",
"title": "Comparison of Decoding Strategies for CTC Acoustic Models"
}
| null | null | null | null | true | null |
10283
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a novel method to describe battery degradation. We use\nthe concept of degradation maps to model the incremental charge capacity loss\nas a function of discrete battery control actions and state of charge. The maps\ncan be scaled to represent any battery system in size and power. Their convex\npiece-wise affine representations allow for tractable optimal control\nformulations and can be used in power system simulations to incorporate battery\ndegradation. The map parameters for different battery technologies are\npublished making them an useful basis to benchmark different battery\ntechnologies in case studies.\n",
"title": "Battery Degradation Maps for Power System Optimization and as a Benchmark Reference"
}
| null | null | null | null | true | null |
10284
| null |
Default
| null | null |
null |
{
"abstract": " We consider the problem of locating a point-source heart arrhythmia using\ndata from a standard diagnostic procedure, where a reference catheter is placed\nin the heart, and arrival times from a second diagnostic catheter are recorded\nas the diagnostic catheter moves around within the heart. We model this\nsituation as a nonconvex feasibility problem, where given a set of arrival\ntimes, we look for a source location that is consistent with the available\ndata. We develop a new optimization approach and fast algorithm to obtain\nonline proposals for the next location to suggest to the operator as she\ncollects data. We validate the procedure using a Monte Carlo simulation based\non patients' electrophysiological data. The proposed procedure robustly and\nquickly locates the source of arrhythmias without any prior knowledge of heart\nanatomy.\n",
"title": "Computer Assisted Localization of a Heart Arrhythmia"
}
| null | null | null | null | true | null |
10285
| null |
Default
| null | null |
null |
{
"abstract": " In this note we propose a method based on artificial neural network to study\nthe transition between states governed by stochastic processes. In particular,\nwe aim for numerical schemes for the committor function, the central object of\ntransition path theory, which satisfies a high-dimensional Fokker-Planck\nequation. By working with the variational formulation of such partial\ndifferential equation and parameterizing the committor function in terms of a\nneural network, approximations can be obtained via optimizing the neural\nnetwork weights using stochastic algorithms. The numerical examples show that\nmoderate accuracy can be achieved for high-dimensional problems.\n",
"title": "Solving for high dimensional committor functions using artificial neural networks"
}
| null | null | null | null | true | null |
10286
| null |
Default
| null | null |
null |
{
"abstract": " The possibility of calculation of the conditional and unconditional\ncomplexity of description of information objects in the algorithmic theory of\ninformation is connected with the limitations for the set of the used languages\nof programming (description). The results of calculation of the conditional\ncomplexity allow introducing the fundamental information dimensions and the\npartial ordering in the set of information objects, and the requirement of\nequality of languages allows introducing the vector space. In case of optimum\ncompression, the \"prefix\" contains the regular part of the information about\nthe object, and is analogous to the classical trajectory of a material point in\nthe physical space, and the \"suffix\" contains the random part of the\ninformation, the quantity of which is analogous to the physical time in the\nintrinsic reference system. Analysis of the mechanism of the \"Einstein's clock\"\nallows representing the result of observation of the material point as a word,\nwritten down in a binary alphabet, thus making the aforesaid analogies more\nclear. The kinematics of the information trajectories is described by the\nLorentz's transformations, identically to its physical analog. At the same\ntime, various languages of description are associated with various reference\nsystems in physics. In the present paper, the information analog of the\nprinciple of least action is found and the main problems of information\ndynamics in the constructed space are formulated.\n",
"title": "The possibility of constructing a relativistic space of information states based on the theory of complexity and analogies with physical space-time"
}
| null | null | null | null | true | null |
10287
| null |
Default
| null | null |
null |
{
"abstract": " Let $G$ be a finite simple graph. For $X \\subset V(G)$, the difference of\n$X$, $d(X) := |X| - |N (X)|$ where $N(X)$ is the neighborhood of $X$ and $\\max\n\\, \\{d(X):X\\subset V(G)\\}$ is called the critical difference of $G$. $X$ is\ncalled a critical set if $d(X)$ equals the critical difference and ker$(G)$ is\nthe intersection of all critical sets. It is known that ker$(G)$ is an\nindependent (vertex) set of $G$. diadem$(G)$ is the union of all critical\nindependent sets. An independent set $S$ is an inclusion minimal set with $d(S)\n> 0$ if no proper subset of $S$ has positive difference.\nA graph $G$ is called König-Egerváry if the sum of its independence\nnumber ($\\alpha (G)$) and matching number ($\\mu (G)$) equals $|V(G)|$. It is\nknown that bipartite graphs are König-Egerváry.\nIn this paper, we study independent sets with positive difference for which\nevery proper subset has a smaller difference and prove a result conjectured by\nLevit and Mandrescu in 2013. The conjecture states that for any graph, the\nnumber of inclusion minimal sets $S$ with $d(S) > 0$ is at least the critical\ndifference of the graph. We also give a short proof of the inequality\n$|$ker$(G)| + |$diadem$(G)| \\le 2\\alpha (G)$ (proved by Short in 2016).\nA characterization of unicyclic non-König-Egerváry graphs is also\npresented and a conjecture which states that for such a graph $G$, the critical\ndifference equals $\\alpha (G) - \\mu (G)$, is proved.\nWe also make an observation about ker$G)$ using Edmonds-Gallai Structure\nTheorem as a concluding remark.\n",
"title": "Problems on Matchings and Independent Sets of a Graph"
}
| null | null | null | null | true | null |
10288
| null |
Default
| null | null |
null |
{
"abstract": " The collision-ionization mechanism of nonsequential double ionization (NSDI)\nprocess in IR+XUV two-color laser fields [\\PRA \\textbf{93}, 043417 (2016)] has\nbeen investigated by us recently. Here we extend this work to study the\ncollision-excitation-ionization (CEI) mechanism of NSDI processes in the\ntwo-color laser fields with different laser conditions. It is found that the\nCEI mechanism makes a dominant contribution to the NSDI as the XUV photon\nenergy is smaller than the ionization threshold of the He$^+$ ion, and the\nmomentum spectrum shows complex interference patterns and symmetrical\nstructures. By channel analysis, we find that, as the energy carried by the\nrecollision electron is not enough to excite the bound electron, the bound\nelectron will absorb XUV photons during their collision, as a result, both\nforward and backward collisions make a comparable contributions to the NSDI\nprocesses. However, it is found that, as the energy carried by the recollision\nelectron is large enough to excite the bound electron, the bound electron does\nnot absorb any XUV photon and it is excited only by sharing the energy carried\nby the recollsion electron, hence the forward collision plays a dominant role\non the NSDI processes. Moreover, we find that the interference patterns of the\nNSDI spectra can be reconstructed by the spectra of two above-threshold\nionization (ATI) processes, which may be used to analyze the structure of the\ntwo separate ATI spectra by NSDI processes.\n",
"title": "Nonsequential double ionization of helium in IR+XUV two-color laser fields II: Collision-excitation ionization process"
}
| null | null | null | null | true | null |
10289
| null |
Default
| null | null |
null |
{
"abstract": " What is chaos? Despite several decades of research on this ubiquitous and\nfundamental phenomenon there is yet no agreed-upon answer to this question.\nRecently, it was realized that all stochastic and deterministic differential\nequations, describing all natural and engineered dynamical systems, possess a\ntopological supersymmetry. It was then suggested that its spontaneous breakdown\ncould be interpreted as the stochastic generalization of deterministic chaos.\nThis conclusion stems from the fact that such phenomenon encompasses features\nthat are traditionally associated with chaotic dynamics such as\nnon-integrability, positive topological entropy, sensitivity to initial\nconditions, and the Poincare-Bendixson theorem. Here, we strengthen and\ncomplete this picture by showing that the hallmarks of set-theoretic chaos --\ntopological transitivity/mixing and dense periodic orbits -- can also be\nattributed to the spontaneous breakdown of topological supersymmetry. We also\ndemonstrate that these features, which highlight the noisy character of chaotic\ndynamics, do not actually admit a stochastic generalization. We therefore\nconclude that spontaneous topological symmetry breaking can be considered as\nthe most general definition of continuous-time dynamical chaos. Contrary to the\ncommon perception and semantics of the word \"chaos\", this phenomenon should\nthen be truly interpreted as the low-symmetry, or ordered phase of the\ndynamical systems that manifest it. Since the long-range order in this case is\ntemporal, we then suggest the word \"chronotaxis\" as a better representation of\nthis phenomenon.\n",
"title": "Chaos or Order?"
}
| null | null |
[
"Physics",
"Mathematics"
] | null | true | null |
10290
| null |
Validated
| null | null |
null |
{
"abstract": " Context: The first Gaia data release (DR1) delivered a catalogue of\nastrometry and photometry for over a billion astronomical sources. Within the\npanoply of methods used for data exploration, visualisation is often the\nstarting point and even the guiding reference for scientific thought. However,\nthis is a volume of data that cannot be efficiently explored using traditional\ntools, techniques, and habits.\nAims: We aim to provide a global visual exploration service for the Gaia\narchive, something that is not possible out of the box for most people. The\nservice has two main goals. The first is to provide a software platform for\ninteractive visual exploration of the archive contents, using common personal\ncomputers and mobile devices available to most users. The second aim is to\nproduce intelligible and appealing visual representations of the enormous\ninformation content of the archive.\nMethods: The interactive exploration service follows a client-server design.\nThe server runs close to the data, at the archive, and is responsible for\nhiding as far as possible the complexity and volume of the Gaia data from the\nclient. This is achieved by serving visual detail on demand. Levels of detail\nare pre-computed using data aggregation and subsampling techniques. For DR1,\nthe client is a web application that provides an interactive multi-panel\nvisualisation workspace as well as a graphical user interface.\nResults: The Gaia archive Visualisation Service offers a web-based\nmulti-panel interactive visualisation desktop in a browser tab. It currently\nprovides highly configurable 1D histograms and 2D scatter plots of Gaia DR1 and\nthe Tycho-Gaia Astrometric Solution (TGAS) with linked views. An innovative\nfeature is the creation of ADQL queries from visually defined regions in plots.\n[abridged]\n",
"title": "Gaia Data Release 1: The archive visualisation service"
}
| null | null | null | null | true | null |
10291
| null |
Default
| null | null |
null |
{
"abstract": " Applications that require substantial computational resources today cannot\navoid the use of heavily parallel machines. Embracing the opportunities of\nparallel computing and especially the possibilities provided by a new\ngeneration of massively parallel accelerator devices such as GPUs, Intel's Xeon\nPhi or even FPGAs enables applications and studies that are inaccessible to\nserial programs. Here we outline the opportunities and challenges of massively\nparallel computing for Monte Carlo simulations in statistical physics, with a\nfocus on the simulation of systems exhibiting phase transitions and critical\nphenomena. This covers a range of canonical ensemble Markov chain techniques as\nwell as generalized ensembles such as multicanonical simulations and population\nannealing. While the examples discussed are for simulations of spin systems,\nmany of the methods are more general and moderate modifications allow them to\nbe applied to other lattice and off-lattice problems including polymers and\nparticle systems. We discuss important algorithmic requirements for such highly\nparallel simulations, such as the challenges of random-number generation for\nsuch cases, and outline a number of general design principles for parallel\nMonte Carlo codes to perform well.\n",
"title": "Monte Carlo methods for massively parallel computers"
}
| null | null | null | null | true | null |
10292
| null |
Default
| null | null |
null |
{
"abstract": " Kiva is an online non-profit crowdsouring microfinance platform that raises\nfunds for the poor in the third world. The borrowers on Kiva are small business\nowners and individuals in urgent need of money. To raise funds as fast as\npossible, they have the option to form groups and post loan requests in the\nname of their groups. While it is generally believed that group loans pose less\nrisk for investors than individual loans do, we study whether this is the case\nin a philanthropic online marketplace. In particular, we measure the effect of\ngroup loans on funding time while controlling for the loan sizes and other\nfactors. Because loan descriptions (in the form of texts) play an important\nrole in lenders' decision process on Kiva, we make use of this information\nthrough deep learning in natural language processing. In this aspect, this is\nthe first paper that uses one of the most advanced deep learning techniques to\ndeal with unstructured data in a way that can take advantage of its superior\nprediction power to answer causal questions. We find that on average, forming\ngroup loans speeds up the funding time by about 3.3 days.\n",
"title": "A Deep Causal Inference Approach to Measuring the Effects of Forming Group Loans in Online Non-profit Microfinance Platform"
}
| null | null | null | null | true | null |
10293
| null |
Default
| null | null |
null |
{
"abstract": " In the ICS, WUT a platform for simulation of cooperation of physical and\nvirtual mobile agents is under development. The paper describes the motivation\nof the research, an organization of the platform, a model of agent, and the\nprinciples of design of the platform. Several experimental simulations are\nbriefly described.\n",
"title": "JADE - A Platform for Research on Cooperation of Physical and Virtual Agents"
}
| null | null | null | null | true | null |
10294
| null |
Default
| null | null |
null |
{
"abstract": " We analytically derive the expressions for the structure of the inner region\nof protoplanetary disks based on the results from the recent hydrodynamical\nsimulations. The inner part of a disk can be divided into four regions:\ndust-free region with gas temperature in the optically thin limit, optically\nthin dust halo, optically thick condensation front and the classical optically\nthick region in order from the inside. We derive the dust-to-gas mass ratio\nprofile in the dust halo using the fact that partial dust condensation\nregulates the temperature to the dust evaporation temperature. Beyond the dust\nhalo, there is an optically thick condensation front where all the available\nsilicate gas condenses out. The curvature of the condensation surface is\ndetermined by the condition that the surface temperature must be nearly equal\nto the characteristic temperature $\\sim 1200{\\,\\rm K}$. We derive the mid-plane\ntemperature in the outer two regions using the two-layer approximation with the\nadditional heating by the condensation front for the outermost region. As a\nresult, the overall temperature profile is step-like with steep gradients at\nthe borders between the outer three regions. The borders might act as planet\ntraps where the inward migration of planets due to gravitational interaction\nwith the gas disk stops. The temperature at the border between the two\noutermost regions coincides with the temperature needed to activate\nmagnetorotational instability, suggesting that the inner edge of the dead zone\nmust lie at this border. The radius of the dead-zone inner edge predicted from\nour solution is $\\sim$ 2-3 times larger than that expected from the classical\noptically thick temperature.\n",
"title": "Analytic Expressions for the Inner-Rim Structure of Passively Heated Protoplanetary Disks"
}
| null | null | null | null | true | null |
10295
| null |
Default
| null | null |
null |
{
"abstract": " The X-ray emission spectrum of liquid ethanol was calculated using density\nfunctional theory and a semi-classical approximation to the Kramers-Heisenberg\nformula including core-hole-induced dynamics. Our spectrum agrees well with the\nexperimental spectrum. We found that the intensity ratio between the two peaks\nat 526 and 527 eV assigned as 10a' and 3a\" depends not only on the hydrogen\nbonding network around the target molecule, but also on the intramolecular\nconformation. This effect is absent in liquid methanol and demonstrates the\nhigh sensitivity of X-ray emission to molecular structure. The dependence of\nspectral features on hydrogen-bonding as well as on dynamical effects following\ncore-excitation are also discussed.\n",
"title": "X-ray Emission Spectrum of Liquid Ethanol : Origin of Split Peaks"
}
| null | null | null | null | true | null |
10296
| null |
Default
| null | null |
null |
{
"abstract": " Class labels have been empirically shown useful in improving the sample\nquality of generative adversarial nets (GANs). In this paper, we mathematically\nstudy the properties of the current variants of GANs that make use of class\nlabel information. With class aware gradient and cross-entropy decomposition,\nwe reveal how class labels and associated losses influence GAN's training.\nBased on that, we propose Activation Maximization Generative Adversarial\nNetworks (AM-GAN) as an advanced solution. Comprehensive experiments have been\nconducted to validate our analysis and evaluate the effectiveness of our\nsolution, where AM-GAN outperforms other strong baselines and achieves\nstate-of-the-art Inception Score (8.91) on CIFAR-10. In addition, we\ndemonstrate that, with the Inception ImageNet classifier, Inception Score\nmainly tracks the diversity of the generator, and there is, however, no\nreliable evidence that it can reflect the true sample quality. We thus propose\na new metric, called AM Score, to provide a more accurate estimation of the\nsample quality. Our proposed model also outperforms the baseline methods in the\nnew metric.\n",
"title": "Activation Maximization Generative Adversarial Nets"
}
| null | null | null | null | true | null |
10297
| null |
Default
| null | null |
null |
{
"abstract": " High Performance Computing is often performed on scarce and shared computing\nresources. To ensure computers are used to their full capacity, administrators\noften incentivize large workloads that are not possible on smaller systems.\nMeasurements in Lattice QCD frequently do not scale to machine-size workloads.\nBy bundling tasks together we can create large jobs suitable for gigantic\npartitions. We discuss METAQ and mpi_jm, software developed to dynamically\ngroup computational tasks together, that can intelligently backfill to consume\nidle time without substantial changes to users' current workflows or\nexecutables.\n",
"title": "Job Management and Task Bundling"
}
| null | null | null | null | true | null |
10298
| null |
Default
| null | null |
null |
{
"abstract": " We propose a new method to detect off-pulse (unpulsed and/or continuous)\nemission from pulsars, using the intensity modulations associated with\ninterstellar scintillation. Our technique involves obtaining the dynamic\nspectra, separately for on-pulse window and off-pulse region, with time and\nfrequency resolutions to properly sample the intensity variations due to\ndiffractive scintillation, and then estimating their mutual correlation as a\nmeasure of off-pulse emission, if any. We describe and illustrate the essential\ndetails of this technique with the help of simulations, as well as real data.\nWe also discuss advantages of this method over earlier approaches to detect\noff-pulse emission. In particular, we point out how certain non-idealities\ninherent to measurement set-ups could potentially affect estimations in earlier\napproaches, and argue that the present technique is immune to such\nnon-idealities. We verify both of the above situations with relevant\nsimulations. We apply this method to observation of PSR B0329+54 at frequencies\n730 and 810 MHz, made with the Green Bank Telescope and present upper limits\nfor the off-pulse intensity at the two frequencies. We expect this technique to\npave way for extensive investigations of off-pulse emission with the help of\neven existing dynamic spectral data on pulsars and of course with more\nsensitive long-duration data from new observations.\n",
"title": "Scintillation based search for off-pulse radio emission from pulsars"
}
| null | null | null | null | true | null |
10299
| null |
Default
| null | null |
null |
{
"abstract": " Structural and topological information play a key role in modeling flow and\ntransport through fractured rock in the subsurface. Discrete fracture network\n(DFN) computational suites such as dfnWorks are designed to simulate flow and\ntransport in such porous media. Flow and transport calculations reveal that a\nsmall backbone of fractures exists, where most flow and transport occurs.\nRestricting the flowing fracture network to this backbone provides a\nsignificant reduction in the network's effective size. However, the particle\ntracking simulations needed to determine the reduction are computationally\nintensive. Such methods may be impractical for large systems or for robust\nuncertainty quantification of fracture networks, where thousands of forward\nsimulations are needed to bound system behavior.\nIn this paper, we develop an alternative network reduction approach to\ncharacterizing transport in DFNs, by combining graph theoretical and machine\nlearning methods. We consider a graph representation where nodes signify\nfractures and edges denote their intersections. Using random forest and support\nvector machines, we rapidly identify a subnetwork that captures the flow\npatterns of the full DFN, based primarily on node centrality features in the\ngraph. Our supervised learning techniques train on particle-tracking backbone\npaths found by dfnWorks, but run in negligible time compared to those\nsimulations. We find that our predictions can reduce the network to\napproximately 20% of its original size, while still generating breakthrough\ncurves consistent with those of the original network.\n",
"title": "Machine learning for graph-based representations of three-dimensional discrete fracture networks"
}
| null | null |
[
"Computer Science",
"Physics",
"Statistics"
] | null | true | null |
10300
| null |
Validated
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.