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": " We present a new class of decentralized first-order methods for nonsmooth and\nstochastic optimization problems defined over multiagent networks. Considering\nthat communication is a major bottleneck in decentralized optimization, our\nmain goal in this paper is to develop algorithmic frameworks which can\nsignificantly reduce the number of inter-node communications. We first propose\na decentralized primal-dual method which can find an $\\epsilon$-solution both\nin terms of functional optimality gap and feasibility residual in\n$O(1/\\epsilon)$ inter-node communication rounds when the objective functions\nare convex and the local primal subproblems are solved exactly. Our major\ncontribution is to present a new class of decentralized primal-dual type\nalgorithms, namely the decentralized communication sliding (DCS) methods, which\ncan skip the inter-node communications while agents solve the primal\nsubproblems iteratively through linearizations of their local objective\nfunctions. By employing DCS, agents can still find an $\\epsilon$-solution in\n$O(1/\\epsilon)$ (resp., $O(1/\\sqrt{\\epsilon})$) communication rounds for\ngeneral convex functions (resp., strongly convex functions), while maintaining\nthe $O(1/\\epsilon^2)$ (resp., $O(1/\\epsilon)$) bound on the total number of\nintra-node subgradient evaluations. We also present a stochastic counterpart\nfor these algorithms, denoted by SDCS, for solving stochastic optimization\nproblems whose objective function cannot be evaluated exactly. In comparison\nwith existing results for decentralized nonsmooth and stochastic optimization,\nwe can reduce the total number of inter-node communication rounds by orders of\nmagnitude while still maintaining the optimal complexity bounds on intra-node\nstochastic subgradient evaluations. The bounds on the subgradient evaluations\nare actually comparable to those required for centralized nonsmooth and\nstochastic optimization.\n",
"title": "Communication-Efficient Algorithms for Decentralized and Stochastic Optimization"
} | null | null | null | null | true | null | 19001 | null | Default | null | null |
null | {
"abstract": " Having the right assortment of shipping boxes in the fulfillment warehouse to\npack and ship customer's online orders is an indispensable and integral part of\nnowadays eCommerce business, as it will not only help maintain a profitable\nbusiness but also create great experiences for customers. However, it is an\nextremely challenging operations task to strategically select the best\ncombination of tens of box sizes from thousands of feasible ones to be\nresponsible for hundreds of thousands of orders daily placed on millions of\ninventory products. In this paper, we present a machine learning approach to\ntackle the task by formulating the box design problem prescriptively as a\ngeneralized version of weighted $k$-medoids clustering problem, where the\nparameters are estimated through a variety of descriptive analytics. We test\nthis machine learning approach on fulfillment data collected from Walmart U.S.\neCommerce, and our approach is shown to be capable of improving the box\nutilization rate by more than $10\\%$.\n",
"title": "A Machine Learning Approach to Shipping Box Design"
} | null | null | [
"Statistics"
]
| null | true | null | 19002 | null | Validated | null | null |
null | {
"abstract": " We present Flipper, a natural language interface for describing high-level\ntask specifications for robots that are compiled into robot actions. Flipper\nstarts with a formal core language for task planning that allows expressing\nrich temporal specifications and uses a semantic parser to provide a natural\nlanguage interface. Flipper provides immediate visual feedback by executing an\nautomatically constructed plan of the task in a graphical user interface. This\nallows the user to resolve potentially ambiguous interpretations. Flipper\nextends itself via naturalization: its users can add definitions for\nutterances, from which Flipper induces new rules and adds them to the core\nlanguage, gradually growing a more and more natural task specification\nlanguage. Flipper improves the naturalization by generalizing the definition\nprovided by users. Unlike other task-specification systems, Flipper enables\nnatural language interactions while maintaining the expressive power and formal\nprecision of a programming language. We show through an initial user study that\nnatural language interactions and generalization can considerably ease the\ndescription of tasks. Moreover, over time, users employ more and more concepts\noutside of the initial core language. Such extensions are available to the\nFlipper community, and users can use concepts that others have defined.\n",
"title": "Precise but Natural Specification for Robot Tasks"
} | null | null | null | null | true | null | 19003 | null | Default | null | null |
null | {
"abstract": " This article presents a rigorous analysis for efficient statistically\naccurate algorithms for solving the Fokker-Planck equations associated with\nhigh-dimensional nonlinear turbulent dynamical systems with conditional\nGaussian structures. Despite the conditional Gaussianity, these nonlinear\nsystems contain many strong non-Gaussian features such as intermittency and\nfat-tailed probability density functions (PDFs). The algorithms involve a\nhybrid strategy that requires only a small number of samples $L$ to capture\nboth the transient and the equilibrium non-Gaussian PDFs with high accuracy.\nHere, a conditional Gaussian mixture in a high-dimensional subspace via an\nextremely efficient parametric method is combined with a judicious Gaussian\nkernel density estimation in the remaining low-dimensional subspace. Rigorous\nanalysis shows that the mean integrated squared error in the recovered PDFs in\nthe high-dimensional subspace is bounded by the inverse square root of the\ndeterminant of the conditional covariance, where the conditional covariance is\ncompletely determined by the underlying dynamics and is independent of $L$.\nThis is fundamentally different from a direct application of kernel methods to\nsolve the full PDF, where $L$ needs to increase exponentially with the\ndimension of the system and the bandwidth shrinks. A detailed comparison\nbetween different methods justifies that the efficient statistically accurate\nalgorithms are able to overcome the curse of dimensionality. It is also shown\nwith mathematical rigour that these algorithms are robust in long time provided\nthat the system is controllable and stochastically stable. Particularly,\ndynamical systems with energy-conserving quadratic nonlinearity as in many\ngeophysical and engineering turbulence are proved to have these properties.\n",
"title": "Rigorous Analysis for Efficient Statistically Accurate Algorithms for Solving Fokker-Planck Equations in Large Dimensions"
} | null | null | null | null | true | null | 19004 | null | Default | null | null |
null | {
"abstract": " We report a methodology for measuring 85Kr/Kr isotopic abundances using Atom\nTrap Trace Analysis (ATTA) that increases sample measurement throughput by over\nan order of magnitude to 6 samples per 24 hours. The noble gas isotope 85Kr\n(half-life = 10.7 yr) is a useful tracer for young groundwater in the age range\nof 5-50 years. ATTA, an efficient and selective laser-based atom counting\nmethod, has recently been applied to 85Kr/Kr isotopic abundance measurements,\nrequiring 5-10 microliters of krypton gas at STP extracted from 50-100 L of\nwater. Previously a single such measurement required 48 hours. Our new method\ndemonstrates that we can measure 85Kr/Kr ratios with 3-5% relative uncertainty\nevery 4 hours, on average, with the same sample requirements.\n",
"title": "Rapid processing of 85Kr/Kr ratios using Atom Trap Trace Analysis"
} | null | null | null | null | true | null | 19005 | null | Default | null | null |
null | {
"abstract": " We address the problem of executing tool-using manipulation skills in\nscenarios where the objects to be used may vary. We assume that point clouds of\nthe tool and target object can be obtained, but no interpretation or further\nknowledge about these objects is provided. The system must interpret the point\nclouds and decide how to use the tool to complete a manipulation task with a\ntarget object; this means it must adjust motion trajectories appropriately to\ncomplete the task. We tackle three everyday manipulations: scraping material\nfrom a tool into a container, cutting, and scooping from a container. Our\nsolution encodes these manipulation skills in a generic way, with parameters\nthat can be filled in at run-time via queries to a robot perception module; the\nperception module abstracts the functional parts for the tool and extracts key\nparameters that are needed for the task. The approach is evaluated in\nsimulation and with selected examples on a PR2 robot.\n",
"title": "Adapting Everyday Manipulation Skills to Varied Scenarios"
} | null | null | null | null | true | null | 19006 | null | Default | null | null |
null | {
"abstract": " Motivated by understanding Majorana zero modes in topological superfluids in\nparticle-number conserving framework beyond the present framework, we study the\neffect of particle number conservation on the Berry phase resulting from\ntransport of a bound quasiparticle around a superfluid vortex. We find that\nparticle-number non-conserving calculations based on Bogoliubov-de Gennes (BdG)\nequations are unable to capture the correct physics when the quasiparticle is\nwithin the penetration depth of the vortex core where the superfluid velocity\nis non-zero. Particle number conservation is crucial for deriving the correct\nBerry phase in this context, and the Berry phase takes non-universal values\ndepending on the system parameters and the external trap imposed to bind the\nquasiparticle. Of particular relevance to Majorana physics are the findings\nthat superfluid condensate affects the part of the Berry phase not accounted\nfor in the standard BdG framework, and that the superfluid many-body ground\nstate of odd number of fermions involves superfluid condensate deformation due\nto the presence of the bound quasiparticle - an effect which is beyond the\ndescription of the BdG equations.\n",
"title": "Effect of Particle Number Conservation on the Berry Phase Resulting from Transport of a Bound Quasiparticle around a Superfluid Vortex"
} | null | null | null | null | true | null | 19007 | null | Default | null | null |
null | {
"abstract": " We have investigated the band structure of the bulk crystal and the (001)\nsurface of the \\beta-InTe layered crystal subjected to biaxial stretching in\nthe layer plane. The calculation has been carried out using the full-potential\nlinearized augmented plane wave method (FP LAPW) implemented in WIEN2k. It has\nbeen shown that at the strain \\Deltaa/a=0.06, where a is the lattice parameter\nin the layer plane, the band gap in the electronic spectrum collapses. With\nfurther strain increase a band inversion occurs. The inclusion of the\nspin-orbit interaction reopens the gap in the electronic spectrum of a bulk\ncrystal, and our calculations show that the spectrum of the surface states has\nthe form of a Dirac cone, typical for topological insulators.\n",
"title": "A new topological insulator - β-InTe strained in the layer plane"
} | null | null | null | null | true | null | 19008 | null | Default | null | null |
null | {
"abstract": " There is a great need to stock materials for production, but storing\nmaterials comes at a cost. Lack of organization in the inventory can result in\na very high cost for the final product, in addition to generating other\nproblems in the production chain. In this work we present mathematical and\nstatistical methods applicable to stock management. The stock analysis using\nABC curves serves to identify which are the priority items, the most expensive\nand with the highest turnover (demand), and thus determine, through stock\ncontrol models, the purchase lot size and the periodicity that minimize the\ntotal costs of storing these materials. Using the Economic Order Quantity (EOQ)\nmodel and the (Q,R) model, the inventory costs of a company were minimized. The\ncomparison of the results provided by the models was performed.\n",
"title": "Stock management (Gestão de estoques)"
} | null | null | null | null | true | null | 19009 | null | Default | null | null |
null | {
"abstract": " The autoencoder is an effective unsupervised learning model which is widely\nused in deep learning. It is well known that an autoencoder with a single\nfully-connected hidden layer, a linear activation function and a squared error\ncost function trains weights that span the same subspace as the one spanned by\nthe principal component loading vectors, but that they are not identical to the\nloading vectors. In this paper, we show how to recover the loading vectors from\nthe autoencoder weights.\n",
"title": "From Principal Subspaces to Principal Components with Linear Autoencoders"
} | null | null | null | null | true | null | 19010 | null | Default | null | null |
null | {
"abstract": " In this article we are interested in the nonlocal regional Schrödinger\nequation with critical exponent \\begin{eqnarray*} &\\epsilon^{2\\alpha}\n(-\\Delta)_{\\rho}^{\\alpha}u + u = \\lambda u^q + u^{2_{\\alpha}^{*}-1} \\mbox{ in }\n\\mathbb{R}^{N}, \\\\ & u \\in H^{\\alpha}(\\mathbb{R}^{N}), \\end{eqnarray*} where\n$\\epsilon$ is a small positive parameter, $\\alpha \\in (0,1)$, $q\\in\n(1,2_{\\alpha}^{*}-1)$, $2_{\\alpha}^{*} = \\frac{2N}{N-2\\alpha}$ is the critical\nSobolev exponent, $\\lambda >0$ is a parameter and $(-\\Delta)_{\\rho}^{\\alpha}$\nis a variational version of the regional laplacian, whose range of scope is a\nball with radius $\\rho(x)>0$. We study the existence of a ground state and we\nanalyze the behavior of semi-classical solutions as $\\varepsilon\\to 0$.\n",
"title": "A critical nonlinear elliptic equation with non local regional diffusion"
} | null | null | null | null | true | null | 19011 | null | Default | null | null |
null | {
"abstract": " It is common practice to decay the learning rate. Here we show one can\nusually obtain the same learning curve on both training and test sets by\ninstead increasing the batch size during training. This procedure is successful\nfor stochastic gradient descent (SGD), SGD with momentum, Nesterov momentum,\nand Adam. It reaches equivalent test accuracies after the same number of\ntraining epochs, but with fewer parameter updates, leading to greater\nparallelism and shorter training times. We can further reduce the number of\nparameter updates by increasing the learning rate $\\epsilon$ and scaling the\nbatch size $B \\propto \\epsilon$. Finally, one can increase the momentum\ncoefficient $m$ and scale $B \\propto 1/(1-m)$, although this tends to slightly\nreduce the test accuracy. Crucially, our techniques allow us to repurpose\nexisting training schedules for large batch training with no hyper-parameter\ntuning. We train ResNet-50 on ImageNet to $76.1\\%$ validation accuracy in under\n30 minutes.\n",
"title": "Don't Decay the Learning Rate, Increase the Batch Size"
} | null | null | null | null | true | null | 19012 | null | Default | null | null |
null | {
"abstract": " If QCD axions form a large fraction of the total mass of dark matter, then\naxion stars could be very abundant in galaxies. As a result, collisions with\neach other, and with other astrophysical bodies, can occur. We calculate the\nrate and analyze the consequences of three classes of collisions, those\noccurring between a dilute axion star and: another dilute axion star, an\nordinary star, or a neutron star. In all cases we attempt to quantify the most\nimportant astrophysical uncertainties; we also pay particular attention to\nscenarios in which collisions lead to collapse of otherwise stable axion stars,\nand possible subsequent decay through number changing interactions. Collisions\nbetween two axion stars can occur with a high total rate, but the low relative\nvelocity required for collapse to occur leads to a very low total rate of\ncollapses. On the other hand, collisions between an axion star and an ordinary\nstar have a large rate, $\\Gamma_\\odot \\sim 3000$ collisions/year/galaxy, and\nfor sufficiently heavy axion stars, it is plausible that most or all such\ncollisions lead to collapse. We identify in this case a parameter space which\nhas a stable region and a region in which collision triggers collapse, which\ndepend on the axion number ($N$) in the axion star, and a ratio of mass to\nradius cubed characterizing the ordinary star ($M_s/R_s^3$). Finally, we\nrevisit the calculation of collision rates between axion stars and neutron\nstars, improving on previous estimates by taking cylindrical symmetry of the\nneutron star distribution into account. Collapse and subsequent decay through\ncollision processes, if occurring with a significant rate, can affect dark\nmatter phenomenology and the axion star mass distribution.\n",
"title": "Collisions of Dark Matter Axion Stars with Astrophysical Sources"
} | null | null | [
"Physics"
]
| null | true | null | 19013 | null | Validated | null | null |
null | {
"abstract": " In the Best-$K$ identification problem (Best-$K$-Arm), we are given $N$\nstochastic bandit arms with unknown reward distributions. Our goal is to\nidentify the $K$ arms with the largest means with high confidence, by drawing\nsamples from the arms adaptively. This problem is motivated by various\npractical applications and has attracted considerable attention in the past\ndecade. In this paper, we propose new practical algorithms for the Best-$K$-Arm\nproblem, which have nearly optimal sample complexity bounds (matching the lower\nbound up to logarithmic factors) and outperform the state-of-the-art algorithms\nfor the Best-$K$-Arm problem (even for $K=1$) in practice.\n",
"title": "Practical Algorithms for Best-K Identification in Multi-Armed Bandits"
} | null | null | null | null | true | null | 19014 | null | Default | null | null |
null | {
"abstract": " In a recent paper Baker and Bowler introduced matroids over hyperfields,\noffering a common generalization of matroids, oriented matroids, and linear\nsubspaces of based vector spaces. This paper introduces the notion of a\ntopological hyperfield and explores the generalization of Grassmannians and\nrealization spaces to this context, particularly in relating the (hyper)fields\nR and C to hyperfields arising in matroid theory and in tropical geometry.\n",
"title": "Hyperfield Grassmannians"
} | null | null | null | null | true | null | 19015 | null | Default | null | null |
null | {
"abstract": " We consider a binary branching process structured by a stochastic trait that\nevolves according to a diffusion process that triggers the branching events, in\nthe spirit of Kimmel's model of cell division with parasite infection. Based on\nthe observation of the trait at birth of the first n generations of the\nprocess, we construct nonparametric estimator of the transition of the\nassociated bifurcating chain and study the parametric estimation of the\nbranching rate. In the limit, as n tends to infinity, we obtain asymptotic\nefficiency in the parametric case and minimax optimality in the nonparametric\ncase.\n",
"title": "Statistical estimation in a randomly structured branching population"
} | null | null | null | null | true | null | 19016 | null | Default | null | null |
null | {
"abstract": " Computer science would not be the same without personal computers. In the\nWest the so called PC revolution started in the late '70s and has its roots in\nhobbyists and do-it-yourself clubs. In the following years the diffusion of\nhome and personal computers has made the discipline closer to many people. A\nbit later, to a lesser extent, yet in a similar way, the revolution took place\nalso in East European countries. Today, the scenario of personal computing has\ncompletely changed, however the computers of the '80s are still objects of\nfascination for a number of retrocomputing fans who enjoy using, programming\nand hacking the old 8-bits. The paper highlights the continuity between\nyesterday's hobbyists and today's retrocomputing enthusiasts, particularly\nfocusing on East European PCs. Besides the preservation of old hardware and\nsoftware, the community is engaged in the development of emulators and cross\ncompilers. Such tools can be used for historical investigation, for example to\ntrace the origins of the BASIC interpreters loaded in the ROMs of East European\nPCs.\n",
"title": "Discovering Eastern European PCs by hacking them. Today"
} | null | null | null | null | true | null | 19017 | null | Default | null | null |
null | {
"abstract": " Recently, some E-commerce sites launch a new interaction box called Tips on\ntheir mobile apps. Users can express their experience and feelings or provide\nsuggestions using short texts typically several words or one sentence. In\nessence, writing some tips and giving a numerical rating are two facets of a\nuser's product assessment action, expressing the user experience and feelings.\nJointly modeling these two facets is helpful for designing a better\nrecommendation system. While some existing models integrate text information\nsuch as item specifications or user reviews into user and item latent factors\nfor improving the rating prediction, no existing works consider tips for\nimproving recommendation quality. We propose a deep learning based framework\nnamed NRT which can simultaneously predict precise ratings and generate\nabstractive tips with good linguistic quality simulating user experience and\nfeelings. For abstractive tips generation, gated recurrent neural networks are\nemployed to \"translate\" user and item latent representations into a concise\nsentence. Extensive experiments on benchmark datasets from different domains\nshow that NRT achieves significant improvements over the state-of-the-art\nmethods. Moreover, the generated tips can vividly predict the user experience\nand feelings.\n",
"title": "Neural Rating Regression with Abstractive Tips Generation for Recommendation"
} | null | null | null | null | true | null | 19018 | null | Default | null | null |
null | {
"abstract": " Calibration of the Advanced LIGO detectors is the quantification of the\ndetectors' response to gravitational waves. Gravitational waves incident on the\ndetectors cause phase shifts in the interferometer laser light which are read\nout as intensity fluctuations at the detector output. Understanding this\ndetector response to gravitational waves is crucial to producing accurate and\nprecise gravitational wave strain data. Estimates of binary black hole and\nneutron star parameters and tests of general relativity require well-calibrated\ndata, as miscalibrations will lead to biased results. We describe the method of\nproducing calibration uncertainty estimates for both LIGO detectors in the\nfirst and second observing runs.\n",
"title": "Calibration Uncertainty for Advanced LIGO's First and Second Observing Runs"
} | null | null | null | null | true | null | 19019 | null | Default | null | null |
null | {
"abstract": " The probability of large deviations of the smallest Schmidt eigenvalue for\nrandom pure states of bipartite systems, denoted as $A$ and $B$, is computed\nanalytically using a Coulomb gas method. It is shown that this probability, for\nlarge $N$, goes as $\\exp[-\\beta N^2\\Phi(\\zeta)]$, where the parameter $\\beta$\nis the Dyson index of the ensemble, $\\zeta$ is the large deviation parameter\nwhile the rate function $\\Phi(\\zeta)$ is calculated exactly. Corresponding\nequilibrium Coulomb charge density is derived for its large deviations. Effects\nof the large deviations of the extreme (largest and smallest) Schmidt\neigenvalues on the bipartite entanglement are studied using the von Neumann\nentropy. Effect of these deviations is also studied on the entanglement between\nsubsystems $1$ and $2$, obtained by further partitioning the subsystem $A$,\nusing the properties of the density matrix's partial transpose\n$\\rho_{12}^\\Gamma$. The density of states of $\\rho_{12}^\\Gamma$ is found to be\nclose to the Wigner's semicircle law with these large deviations. The\nentanglement properties are captured very well by a simple random matrix model\nfor the partial transpose. The model predicts the entanglement transition\nacross a critical large deviation parameter $\\zeta$. Log negativity is used to\nquantify the entanglement between subsystems $1$ and $2$. Analytical formulas\nfor it are derived using the simple model. Numerical simulations are in\nexcellent agreement with the analytical results.\n",
"title": "Entanglement transitions induced by large deviations"
} | null | null | null | null | true | null | 19020 | null | Default | null | null |
null | {
"abstract": " Large-scale systems with arrays of solid state disks (SSDs) have become\nincreasingly common in many computing segments. To make such systems resilient,\nwe can adopt erasure coding such as Reed-Solomon (RS) code as an alternative to\nreplication because erasure coding can offer a significantly lower storage cost\nthan replication. To understand the impact of using erasure coding on system\nperformance and other system aspects such as CPU utilization and network\ntraffic, we build a storage cluster consisting of approximately one hundred\nprocessor cores with more than fifty high-performance SSDs, and evaluate the\ncluster with a popular open-source distributed parallel file system, Ceph. Then\nwe analyze behaviors of systems adopting erasure coding from the following five\nviewpoints, compared with those of systems using replication: (1) storage\nsystem I/O performance; (2) computing and software overheads; (3) I/O\namplification; (4) network traffic among storage nodes; (5) the impact of\nphysical data layout on performance of RS-coded SSD arrays. For all these\nanalyses, we examine two representative RS configurations, which are used by\nGoogle and Facebook file systems, and compare them with triple replication that\na typical parallel file system employs as a default fault tolerance mechanism.\nLastly, we collect 54 block-level traces from the cluster and make them\navailable for other researchers.\n",
"title": "Understanding System Characteristics of Online Erasure Coding on Scalable, Distributed and Large-Scale SSD Array Systems"
} | null | null | null | null | true | null | 19021 | null | Default | null | null |
null | {
"abstract": " We provide justifications for two questions on special maps on subgroups of\nthe reals. We will show that the questions can be treated from different points\nof view. We also discuss two versions of Anderson's Involution Conjecture.\n",
"title": "Three Questions on Special Homeomorphisms on Subgroups of $R$ and $R^\\infty$"
} | null | null | null | null | true | null | 19022 | null | Default | null | null |
null | {
"abstract": " As a competitive alternative to least squares regression, quantile regression\nis popular in analyzing heterogenous data. For quantile regression model\nspecified for one single quantile level $\\tau$, major difficulties of\nsemiparametric efficient estimation are the unavailability of a parametric\nefficient score and the conditional density estimation. In this paper, with the\nhelp of the least favorable submodel technique, we first derive the\nsemiparametric efficient scores for linear quantile regression models that are\nassumed for a single quantile level, multiple quantile levels and all the\nquantile levels in $(0,1)$ respectively. Our main discovery is a one-step\n(nearly) semiparametric efficient estimation for the regression coefficients of\nthe quantile regression models assumed for multiple quantile levels, which has\nseveral advantages: it could be regarded as an optimal way to pool information\nacross multiple/other quantiles for efficiency gain; it is computationally\nfeasible and easy to implement, as the initial estimator is easily available;\ndue to the nature of quantile regression models under investigation, the\nconditional density estimation is straightforward by plugging in an initial\nestimator. The resulting estimator is proved to achieve the corresponding\nsemiparametric efficiency lower bound under regularity conditions. Numerical\nstudies including simulations and an example of birth weight of children\nconfirms that the proposed estimator leads to higher efficiency compared with\nthe Koenker-Bassett quantile regression estimator for all quantiles of\ninterest.\n",
"title": "Nearly Semiparametric Efficient Estimation of Quantile Regression"
} | null | null | null | null | true | null | 19023 | null | Default | null | null |
null | {
"abstract": " The understanding of epidemics on networks has greatly benefited from the\nrecent application of message-passing approaches, which allow to derive exact\nresults for irreversible spreading (i.e. diseases with permanent acquired\nimmunity) in locally-tree like topologies. This success has suggested the\napplication of the same approach to reversible epidemics, for which an\nindividual can contract the epidemic and recover repeatedly. The underlying\nassumption is that backtracking paths (i.e. an individual is reinfected by a\nneighbor he/she previously infected) do not play a relevant role. In this paper\nwe show that this is not the case for reversible epidemics, since the neglect\nof backtracking paths leads to a formula for the epidemic threshold that is\nqualitatively incorrect in the large size limit. Moreover we define a modified\nreversible dynamics which explicitly forbids direct backtracking events and\nshow that this modification completely upsets the phenomenology.\n",
"title": "Relevance of backtracking paths in epidemic spreading on networks"
} | null | null | null | null | true | null | 19024 | null | Default | null | null |
null | {
"abstract": " Hybrid graphene photoconductor/phototransistor has achieved giant\nphotoresponsivity, but its response speed dramatically degrades as the expense\ndue to the long lifetime of trapped interfacial carriers. In this work, by\nintercalating a large-area atomically thin MoS2 film into a hybrid graphene\nphotoconductor, we have developed a prototype tunneling photoconductor, which\nexhibits a record-fast response (rising time ~17 ns) and a high responsivity\n(~$3\\times10^4$ A/W at 635 nm and 16.8 nW illumination) across the broad\nspectral range. We demonstrate that the photo-excited carriers generated in\nsilicon are transferred into graphene through a tunneling process rather than\ncarrier drift. The atomically thin MoS2 film not only serves as tunneling layer\nbut also passivates surface states, which in combination delivers a superior\nresponse speed (~3 order of magnitude improved than a device without MoS2\nlayer), while the responsivity remains high. This intriguing tunneling\nphotoconductor integrates both fast response and high responsivity and thus has\nsignificant potential in practical applications of optoelectronic devices.\n",
"title": "Hybrid graphene tunneling photoconductor with interface engineering towards fast photoresponse and high responsivity"
} | null | null | [
"Physics"
]
| null | true | null | 19025 | null | Validated | null | null |
null | {
"abstract": " This paper proposes a real-time embedded fall detection system using a\nDVS(Dynamic Vision Sensor) that has never been used for traditional fall\ndetection, a dataset for fall detection using that, and a DVS-TN(DVS-Temporal\nNetwork). The first contribution is building a DVS Falls Dataset, which made\nour network to recognize a much greater variety of falls than the existing\ndatasets that existed before and solved privacy issues using the DVS. Secondly,\nwe introduce the DVS-TN : optimized deep learning network to detect falls using\nDVS. Finally, we implemented a fall detection system which can run on\nlow-computing H/W with real-time, and tested on DVS Falls Dataset that takes\ninto account various falls situations. Our approach achieved 95.5% on the\nF1-score and operates at 31.25 FPS on NVIDIA Jetson TX1 board.\n",
"title": "Embedded Real-Time Fall Detection Using Deep Learning For Elderly Care"
} | null | null | null | null | true | null | 19026 | null | Default | null | null |
null | {
"abstract": " In this paper, we consider the problem of solving linear algebraic equations\nof the form $Ax=b$ among multi agents which seek a solution by using local\ninformation in presence of random communication topologies. The equation is\nsolved by $m$ agents where each agent only knows a subset of rows of the\npartitioned matrix $[A,b]$. We formulate the problem such that this formulation\ndoes not need the distribution of random interconnection graphs. Therefore,\nthis framework includes asynchronous updates or unreliable communication\nprotocols without B-connectivity assumption. We apply the random\nKrasnoselskii-Mann iterative algorithm which converges almost surely and in\nmean square to a solution of the problem for any matrices $A$ and $b$ and any\ninitial conditions of agents' states. We demonestrate that the limit point to\nwhich the agents' states converge is determined by the unique solution of a\nconvex optimization problem regardless of the distribution of random\ncommunication graphs. Eventually, we show by two numerical examples that the\nrate of convergence of the algorithm cannot be guaranteed.\n",
"title": "A Distributed Algorithm for Solving Linear Algebraic Equations Over Random Networks"
} | null | null | [
"Computer Science"
]
| null | true | null | 19027 | null | Validated | null | null |
null | {
"abstract": " We present the results of the investigation of magnetic properties of the\nEr$_2$Ti$_{2-x}$Sn$_x$O$_7$ series. For small doping values the ordering\ntemperature decreases linearly with $x$ while the moment configuration remains\nthe same as in the $x = 0$ parent compound. Around $x = 1.7$ doping level we\nobserve a change in the behavior, where the ordering temperature starts to\nincrease and new magnetic Bragg peaks appear. For the first time we present\nevidence of a long-range order (LRO) in Er$_2$Sn$_2$O$_7$ ($x = 2.0$) below\n$T_N = 130$ mK. It is revealed that the moment configuration corresponds to a\nPalmer-Chalker type with a value of the magnetic moment significantly\nrenormalized compared to $x = 0$. We discuss our results in the framework of a\npossible quantum phase transition occurring close to $x = 1.7$.\n",
"title": "Doping-induced quantum cross-over in Er$_2$Ti$_{2-x}$Sn$_x$O$_7$"
} | null | null | null | null | true | null | 19028 | null | Default | null | null |
null | {
"abstract": " The stochastic block model (SBM) is a probabilistic model for community\nstructure in networks. Typically, only the adjacency matrix is used to perform\nSBM parameter inference. In this paper, we consider circumstances in which\nnodes have an associated vector of continuous attributes that are also used to\nlearn the node-to-community assignments and corresponding SBM parameters. While\nthis assumption is not realistic for every application, our model assumes that\nthe attributes associated with the nodes in a network's community can be\ndescribed by a common multivariate Gaussian model. In this augmented,\nattributed SBM, the objective is to simultaneously learn the SBM connectivity\nprobabilities with the multivariate Gaussian parameters describing each\ncommunity. While there are recent examples in the literature that combine\nconnectivity and attribute information to inform community detection, our model\nis the first augmented stochastic block model to handle multiple continuous\nattributes. This provides the flexibility in biological data to, for example,\naugment connectivity information with continuous measurements from multiple\nexperimental modalities. Because the lack of labeled network data often makes\ncommunity detection results difficult to validate, we highlight the usefulness\nof our model for two network prediction tasks: link prediction and\ncollaborative filtering. As a result of fitting this attributed stochastic\nblock model, one can predict the attribute vector or connectivity patterns for\na new node in the event of the complementary source of information\n(connectivity or attributes, respectively). We also highlight two biological\nexamples where the attributed stochastic block model provides satisfactory\nperformance in the link prediction and collaborative filtering tasks.\n",
"title": "Stochastic Block Models with Multiple Continuous Attributes"
} | null | null | null | null | true | null | 19029 | null | Default | null | null |
null | {
"abstract": " It is well known that an extreme order statistic and a central order\nstatistic (os) as well as an intermediate os and a central os from a sample of\niid univariate random variables get asymptotically independent as the sample\nsize increases. We extend this result to bivariate random variables, where the\nos are taken componentwise. An explicit representation of the conditional\ndistribution of bivariate os turns out to be a powerful tool.\n",
"title": "Asymptotic Independence of Bivariate Order Statistics"
} | null | null | null | null | true | null | 19030 | null | Default | null | null |
null | {
"abstract": " Material recognition enables robots to incorporate knowledge of material\nproperties into their interactions with everyday objects. For example, material\nrecognition opens up opportunities for clearer communication with a robot, such\nas \"bring me the metal coffee mug\", and recognizing plastic versus metal is\ncrucial when using a microwave or oven. However, collecting labeled training\ndata with a robot is often more difficult than unlabeled data. We present a\nsemi-supervised learning approach for material recognition that uses generative\nadversarial networks (GANs) with haptic features such as force, temperature,\nand vibration. Our approach achieves state-of-the-art results and enables a\nrobot to estimate the material class of household objects with ~90% accuracy\nwhen 92% of the training data are unlabeled. We explore how well this approach\ncan recognize the material of new objects and we discuss challenges facing\ngeneralization. To motivate learning from unlabeled training data, we also\ncompare results against several common supervised learning classifiers. In\naddition, we have released the dataset used for this work which consists of\ntime-series haptic measurements from a robot that conducted thousands of\ninteractions with 72 household objects.\n",
"title": "Semi-Supervised Haptic Material Recognition for Robots using Generative Adversarial Networks"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 19031 | null | Validated | null | null |
null | {
"abstract": " This paper presents a method for the optimization of multi-component\nstructures comprised of two and three materials considering large motion\nsliding contact and separation along interfaces. The structural geometry is\ndefined by an explicit level set method, which allows for both shape and\ntopology changes. The mechanical model assumes finite strains, a nonlinear\nelastic material behavior, and a quasi-static response. Identification of\noverlapping surface position is handled by a coupled parametric representation\nof contact surfaces. A stabilized Lagrange method and an active set strategy\nare used to model frictionless contact and separation. The mechanical model is\ndiscretized by the extended finite element method which maintains a clear\ndefinition of geometry. Face-oriented ghost penalization and dynamic relaxation\nare implemented to improve the stability of the physical response prediction. A\nnonlinear programming scheme is used to solve the optimization problem, which\nis regularized by introducing a perimeter penalty into the objective function.\nSensitivities are determined by the adjoint method. The main characteristics of\nthe proposed method are studied by numerical examples in two dimensions. The\nnumerical results demonstrate improved design performance when compared to\nmodels optimized with a small strain assumption. Additionally, examples with\nload path dependent objectives display non-intuitive designs.\n",
"title": "Level set shape and topology optimization of finite strain bilateral contact problems"
} | null | null | null | null | true | null | 19032 | null | Default | null | null |
null | {
"abstract": " Thermal effects are already important in currently operating interferometric\ngravitational wave detectors. Planned upgrades of these detectors involve\nincreasing optical power to combat quantum shot noise. We consider the\nramifications of this increased power for one particular class of laser\nbeams--wide, flat-topped, mesa beams. In particular we model a single mesa beam\nFabry-Perot cavity having thermoelastically deformed mirrors. We calculate the\nintensity profile of the fundamental cavity eigenmode in the presence of\nthermal perturbations, and the associated changes in thermal noise. We also\noutline an idealized method of correcting for such effects. At each stage we\ncontrast our results with those of a comparable Gaussian beam cavity. Although\nwe focus on mesa beams the techniques described are applicable to any\nazimuthally symmetric system.\n",
"title": "Thermal distortions of non-Gaussian beams in Fabry-Perot cavities"
} | null | null | null | null | true | null | 19033 | null | Default | null | null |
null | {
"abstract": " We show some rigidity properties of divergence-free vector fields defined on\nhalf-spaces. As an application, we prove the existence of the classical trace\nfor a bounded, divergence-measure vector field $\\xi$ defined on the Euclidean\nplane, at almost every point of a locally oriented rectifiable set $S$, under\nthe assumption that its weak normal trace $[\\xi\\cdot \\nu_S]$ attains a local\nmaximum for the norm of $\\xi$ at the point.\n",
"title": "Rigidity and trace properties of divergence-measure vector fields"
} | null | null | null | null | true | null | 19034 | null | Default | null | null |
null | {
"abstract": " The first goal of this note is to study the Almansi property on an\nm-dimensional model in the sense of Greene and Wu and, more generally, in a\nRiemannian geometric setting. In particular, we shall prove that the only model\non which the Almansi property is verified is the Euclidean space R^m. In the\nsecond part of the paper we shall study Almansi's property and biharmonicity\nfor functions which depend on the distance from a given submanifold. Finally,\nin the last section we provide an extension to the semi-Euclidean case R^{p,q}\nwhich includes the proof of the classical Almansi property in R^m as a special\ninstance.\n",
"title": "A note on the Almansi property"
} | null | null | null | null | true | null | 19035 | null | Default | null | null |
null | {
"abstract": " We consider the quantum complexity of computing Schatten $p$-norms and\nrelated quantities, and find that the problem of estimating these quantities is\nclosely related to the one clean qubit model of computation. We show that the\nproblem of approximating $\\text{Tr}\\, (|A|^p)$ for a log-local $n$-qubit\nHamiltonian $A$ and $p=\\text{poly}(n)$, up to a suitable level of accuracy, is\ncontained in DQC1; and that approximating this quantity up to a somewhat higher\nlevel of accuracy is DQC1-hard. In some cases the level of accuracy achieved by\nthe quantum algorithm is substantially better than a natural classical\nalgorithm for the problem. The same problem can be solved for arbitrary sparse\nmatrices in BQP. One application of the algorithm is the approximate\ncomputation of the energy of a graph.\n",
"title": "The Quantum Complexity of Computing Schatten $p$-norms"
} | null | null | null | null | true | null | 19036 | null | Default | null | null |
null | {
"abstract": " We define mixed states associated with submanifolds with probability\ndensities in quantizable closed K{ä}hler manifolds. Then, we address the\nproblem of comparing two such states via their fidelity. Firstly, we estimate\nthe sub-fidelity and super-fidelity of two such states, giving lower and upper\nbounds for their fidelity, when the underlying submanifolds are two Lagrangian\nsubmanifolds intersecting transversally at a finite number of points, in the\nsemiclassical limit. Secondly, we investigate a family of examples on the\nsphere, for which we manage to obtain a better upper bound for the fidelity. We\nconclude by stating a conjecture regarding the fidelity in the general case.\n",
"title": "Bounds for fidelity of semiclassical Lagrangian states in K{ä}hler quantization"
} | null | null | [
"Mathematics"
]
| null | true | null | 19037 | null | Validated | null | null |
null | {
"abstract": " A precision measurement by AMS of the antiproton-to-proton flux ratio in\nprimary cosmic rays in the absolute rigidity range from 1 to 450 GV is\npresented based on $3.49\\times10^5$ antiproton events and $2.42\\times10^9$\nproton events. Above $\\sim60$ GV the antiproton to proton flux ratio is\nconsistent with being rigidity independent. A decreasing behaviour is expected\nfor this ratio considering the traditional models for the secondary antiproton\nflux.\n",
"title": "Precision measurement of antiproton to proton ratio with the Alpha Magnetic Spectrometer on the International Space Station"
} | null | null | null | null | true | null | 19038 | null | Default | null | null |
null | {
"abstract": " We derive analogues of the classical Rayleigh, Fjortoft and Arnold stability\nand instability theorems in the context of the 2D $\\alpha$-Euler equations.\n",
"title": "Stability criteria for the 2D $α$-Euler equations"
} | null | null | null | null | true | null | 19039 | null | Default | null | null |
null | {
"abstract": " We propose a superconducting spin-triplet valve, which consists of a\nsuperconductor and an itinerant magnetic material, with the magnet showing an\nintrinsic non-collinear order characterized by a wave vector that may be\naligned in a few equivalent preferred directions under control of a weak\nexternal magnetic field. Re-orienting the spiral direction allows one to\ncontrollably modify long-range spin-triplet superconducting correlations,\nleading to spin-valve switching behavior. Our results indicate that the\nspin-valve effect may be noticeable. This bilayer may be used as a magnetic\nmemory element for cryogenic nanoelectronics. It has the following advantages\nin comparison to superconducting spin valves proposed previously: (i) it\ncontains only one magnetic layer, which may be more easily fabricated and\ncontrolled, (ii) its ground states are separated by a potential barrier, which\nsolves the \"half-select\" problem of the addressed switch of memory elements.\n",
"title": "Superconducting spin valves controlled by spiral re-orientation in B20-family magnets"
} | null | null | null | null | true | null | 19040 | null | Default | null | null |
null | {
"abstract": " Learning has propelled the cutting edge of performance in robotic control to\nnew heights, allowing robots to operate with high performance in conditions\nthat were previously unimaginable. The majority of the work, however, assumes\nthat the unknown parts are static or slowly changing. This limits them to\nstatic or slowly changing environments. However, in the real world, a robot may\nexperience various unknown conditions. This paper presents a method to extend\nan existing single mode GP-based safe learning controller to learn an\nincreasing number of non-linear models for the robot dynamics. We show that\nthis approach enables a robot to re-use past experience from a large number of\npreviously visited operating conditions, and to safely adapt when a new and\ndistinct operating condition is encountered. This allows the robot to achieve\nsafety and high performance in an large number of operating conditions that do\nnot have to be specified ahead of time. Our approach runs independently from\nthe controller, imposing no additional computation time on the control loop\nregardless of the number of previous operating conditions considered. We\ndemonstrate the effectiveness of our approach in experiment on a 900\\,kg ground\nrobot with both physical and artificial changes to its dynamics. All of our\nexperiments are conducted using vision for localization.\n",
"title": "Experience Recommendation for Long Term Safe Learning-based Model Predictive Control in Changing Operating Conditions"
} | null | null | null | null | true | null | 19041 | null | Default | null | null |
null | {
"abstract": " We prove sharp upper and lower bounds for generalized Calderón's sums\nassociated to frames on LCA groups generated by affine actions of cocompact\nsubgroup translations and general measurable families of automorphisms. The\nproof makes use of techniques of analysis on metric spaces, and relies on a\ncounting estimate of lattice points inside metric balls. We will deduce as\nspecial cases Calderón-type inequalities for families of expanding\nautomorphisms as well as for LCA-Gabor systems.\n",
"title": "Calderón-type inequalities for affine frames"
} | null | null | null | null | true | null | 19042 | null | Default | null | null |
null | {
"abstract": " Transition metal dichalcogenides (TMDs) are interesting for understanding\nfundamental physics of two-dimensional materials (2D) as well as for many\nemerging technologies, including spin electronics. Here, we report the\ndiscovery of long-range magnetic order below TM = 40 K and 100 K in bulk\nsemiconducting TMDs 2H-MoTe2 and 2H-MoSe2, respectively, by means of muon\nspin-rotation (muSR), scanning tunneling microscopy (STM), as well as density\nfunctional theory (DFT) calculations. The muon spin rotation measurements show\nthe presence of a large and homogeneous internal magnetic fields at low\ntemperatures in both compounds indicative of long-range magnetic order. DFT\ncalculations show that this magnetism is promoted by the presence of defects in\nthe crystal. The STM measurements show that the vast majority of defects in\nthese materials are metal vacancies and chalcogen-metal antisites which are\nrandomly distributed in the lattice at the sub-percent level. DFT indicates\nthat the antisite defects are magnetic with a magnetic moment in the range of\n0.9-2.8 mu_B. Further, we find that the magnetic order stabilized in 2H-MoTe2\nand 2H-MoSe2 is highly sensitive to hydrostatic pressure. These observations\nestablish 2H-MoTe2 and 2H-MoSe2 as a new class of magnetic semiconductors and\nopens a path to studying the interplay of 2D physics and magnetism in these\ninteresting semiconductors.\n",
"title": "Magnetism in Semiconducting Molybdenum Dichalcogenides"
} | null | null | null | null | true | null | 19043 | null | Default | null | null |
null | {
"abstract": " We give a formula for computing the characteristic polynomial for certain\nhyperplane arrangements in terms of the number of bipartite graphs of given\nrank and cardinality.\n",
"title": "Characteristic Polynomial of Certain Hyperplane Arrangements through Graph Theory"
} | null | null | [
"Mathematics"
]
| null | true | null | 19044 | null | Validated | null | null |
null | {
"abstract": " The process to certify highly Automated Vehicles has not yet been defined by\nany country in the world. Currently, companies test Automated Vehicles on\npublic roads, which is time-consuming and inefficient. We proposed the\nAccelerated Evaluation concept, which uses a modified statistics of the\nsurrounding vehicles and the Importance Sampling theory to reduce the\nevaluation time by several orders of magnitude, while ensuring the evaluation\nresults are statistically accurate. In this paper, we further improve the\naccelerated evaluation concept by using Piecewise Mixture Distribution models,\ninstead of Single Parametric Distribution models. We developed and applied this\nidea to forward collision control system reacting to vehicles making cut-in\nlane changes. The behavior of the cut-in vehicles was modeled based on more\nthan 403,581 lane changes collected by the University of Michigan Safety Pilot\nModel Deployment Program. Simulation results confirm that the accuracy and\nefficiency of the Piecewise Mixture Distribution method outperformed single\nparametric distribution methods in accuracy and efficiency, and accelerated the\nevaluation process by almost four orders of magnitude.\n",
"title": "Accelerated Evaluation of Automated Vehicles Using Piecewise Mixture Models"
} | null | null | null | null | true | null | 19045 | null | Default | null | null |
null | {
"abstract": " The phase shift's influence of two strong pulsed laser waves on effective\ninteraction of electrons was studied. Considerable amplification of electrons\nrepulsion in the certain range of phase shifts and waves intensities is shown.\nThat leads to electrons scatter on greater distances than without an external\nfield. The value of the distance can be greater on 2-3 order of magnitude. Also\nconsiderable influence of the phase shift of pulses of waves on the possibility\nof effective attraction of electrons is shown.\n",
"title": "Phase shift's influence of two strong pulsed laser waves on effective interaction of electrons"
} | null | null | null | null | true | null | 19046 | null | Default | null | null |
null | {
"abstract": " We present a novel mechanism for resolving the mechanical rigidity of\nnanoscopic circular polymers that flow in a complex environment. The emergence\nof a regime of negative differential mobility induced by topological\ninteractions between the rings and the substrate is the key mechanism for\nselective sieving of circular polymers with distinct flexibilities. A simple\nmodel accurately describes the sieving process observed in molecular dynamics\nsimulations and yields experimentally verifiable analytical predictions, which\ncan be used as a reference guide for improving filtration procedures of\ncircular filaments. The topological sieving mechanism we propose ought to be\nrelevant also in probing the microscopic details of complex substrates.\n",
"title": "Topological Sieving of Rings According to their Rigidity"
} | null | null | null | null | true | null | 19047 | null | Default | null | null |
null | {
"abstract": " We present a new kind of structural Markov property for probabilistic laws on\ndecomposable graphs, which allows the explicit control of interactions between\ncliques, so is capable of encoding some interesting structure. We prove the\nequivalence of this property to an exponential family assumption, and discuss\nidentifiability, modelling, inferential and computational implications.\n",
"title": "A structural Markov property for decomposable graph laws that allows control of clique intersections"
} | null | null | null | null | true | null | 19048 | null | Default | null | null |
null | {
"abstract": " We introduce Fisher consistency in the sense of unbiasedness as a desirable\nproperty for estimators of class prior probabilities. Lack of Fisher\nconsistency could be used as a criterion to dismiss estimators that are\nunlikely to deliver precise estimates in test datasets under prior probability\nand more general dataset shift. The usefulness of this unbiasedness concept is\ndemonstrated with three examples of classifiers used for quantification:\nAdjusted Classify & Count, EM-algorithm and CDE-Iterate. We find that Adjusted\nClassify & Count and EM-algorithm are Fisher consistent. A counter-example\nshows that CDE-Iterate is not Fisher consistent and, therefore, cannot be\ntrusted to deliver reliable estimates of class probabilities.\n",
"title": "Fisher consistency for prior probability shift"
} | null | null | null | null | true | null | 19049 | null | Default | null | null |
null | {
"abstract": " The notion of disentangled autoencoders was proposed as an extension to the\nvariational autoencoder by introducing a disentanglement parameter $\\beta$,\ncontrolling the learning pressure put on the possible underlying latent\nrepresentations. For certain values of $\\beta$ this kind of autoencoders is\ncapable of encoding independent input generative factors in separate elements\nof the code, leading to a more interpretable and predictable model behaviour.\nIn this paper we quantify the effects of the parameter $\\beta$ on the model\nperformance and disentanglement. After training multiple models with the same\nvalue of $\\beta$, we establish the existence of consistent variance in one of\nthe disentanglement measures, proposed in literature. The negative consequences\nof the disentanglement to the autoencoder's discriminative ability are also\nasserted while varying the amount of examples available during training.\n",
"title": "Quantifying the Effects of Enforcing Disentanglement on Variational Autoencoders"
} | null | null | null | null | true | null | 19050 | null | Default | null | null |
null | {
"abstract": " Global hypothesis tests are a useful tool in the context of, e.g, clinical\ntrials, genetic studies or meta analyses, when researchers are not interested\nin testing individual hypotheses, but in testing whether none of the hypotheses\nis false. There are several possibilities how to test the global null\nhypothesis when the individual null hypotheses are independent. If it is\nassumed that many of the individual null hypotheses are false, combinations\ntests have been recommended to maximise power. If, however, it is assumed that\nonly one or a few null hypotheses are false, global tests based on individual\ntest statistics are more powerful (e.g., Bonferroni or Simes test). However,\nusually there is no a-priori knowledge on the number of false individual null\nhypotheses. We therefore propose an omnibus test based on the combination of\np-values. We show that this test yields an impressive overall performance. The\nproposed method is implemented in the R-package omnibus.\n",
"title": "An omnibus test for the global null hypothesis"
} | null | null | null | null | true | null | 19051 | null | Default | null | null |
null | {
"abstract": " We prove that the ordinary least-squares (OLS) estimator attains nearly\nminimax optimal performance for the identification of linear dynamical systems\nfrom a single observed trajectory. Our upper bound relies on a generalization\nof Mendelson's small-ball method to dependent data, eschewing the use of\nstandard mixing-time arguments. Our lower bounds reveal that these upper bounds\nmatch up to logarithmic factors. In particular, we capture the correct\nsignal-to-noise behavior of the problem, showing that more unstable linear\nsystems are easier to estimate. This behavior is qualitatively different from\narguments which rely on mixing-time calculations that suggest that unstable\nsystems are more difficult to estimate. We generalize our technique to provide\nbounds for a more general class of linear response time-series.\n",
"title": "Learning Without Mixing: Towards A Sharp Analysis of Linear System Identification"
} | null | null | [
"Statistics"
]
| null | true | null | 19052 | null | Validated | null | null |
null | {
"abstract": " Motivated by applications in protein function prediction, we consider a\nchallenging supervised classification setting in which positive labels are\nscarce and there are no explicit negative labels. The learning algorithm must\nthus select which unlabeled examples to use as negative training points,\npossibly ending up with an unbalanced learning problem. We address these issues\nby proposing an algorithm that combines active learning (for selecting negative\nexamples) with imbalance-aware learning (for mitigating the label imbalance).\nIn our experiments we observe that these two techniques operate\nsynergistically, outperforming state-of-the-art methods on standard protein\nfunction prediction benchmarks.\n",
"title": "Positive and Unlabeled Learning through Negative Selection and Imbalance-aware Classification"
} | null | null | null | null | true | null | 19053 | null | Default | null | null |
null | {
"abstract": " We study properties of quantim wires with spin-orbit coupling and time\nreversal symmetry breaking, in normal and superconducting states. Electronic\nband structures are classified according to quasi-one-dimensional magnetic\npoint groups, or magnetic classes. The latter belong to one of three distinct\ntypes, depending on the way the time reversal operation appears in the group\nelements. The superconducting gap functions are constructed using antiunitary\noperations and have different symmetry properties depending on the type of the\nmagnetic point group. We obtain the spectrum of the Andreev boundary modes near\nthe end of the wire in a model-independent way, using the semiclassical\napproach with the boundary conditions described by a phenomenological\nscattering matrix. Explicit expressions for the bulk topological invariants\ncontrolling the number of the boundary zero modes are presented in the general\nmultiband case for two types of the magnetic point groups, corresponding to\nDIII and BDI symmetry classes.\n",
"title": "Superconductivity in quantum wires: A symmetry analysis"
} | null | null | null | null | true | null | 19054 | null | Default | null | null |
null | {
"abstract": " The paper presents a new state estimation algorithm for a bilinear equation\nrepresenting the Fourier- Galerkin (FG) approximation of the Navier-Stokes (NS)\nequations on a torus in R2. This state equation is subject to uncertain but\nbounded noise in the input (Kolmogorov forcing) and initial conditions, and its\noutput is incomplete and contains bounded noise. The algorithm designs a\ntime-dependent gain such that the estimation error converges to zero\nexponentially. The sufficient condition for the existence of the gain are\nformulated in the form of algebraic Riccati equations. To demonstrate the\nresults we apply the proposed algorithm to the reconstruction a chaotic fluid\nflow from incomplete and noisy data.\n",
"title": "Exponentially convergent data assimilation algorithm for Navier-Stokes equations"
} | null | null | null | null | true | null | 19055 | null | Default | null | null |
null | {
"abstract": " The group affect or emotion in an image of people can be inferred by\nextracting features about both the people in the picture and the overall makeup\nof the scene. The state-of-the-art on this problem investigates a combination\nof facial features, scene extraction and even audio tonality. This paper\ncombines three additional modalities, namely, human pose, text-based tagging\nand CNN extracted features / predictions. To the best of our knowledge, this is\nthe first time all of the modalities were extracted using deep neural networks.\nWe evaluate the performance of our approach against baselines and identify\ninsights throughout this paper.\n",
"title": "A Multi-Modal Approach to Infer Image Affect"
} | null | null | null | null | true | null | 19056 | null | Default | null | null |
null | {
"abstract": " 3D beamforming is a promising approach for interference coordination in\ncellular networks which brings significant improvements in comparison with\nconventional 2D beamforming techniques. This paper investigates the problem of\njoint beamforming design and tilt angle adaptation of the BS antenna array for\nmaximizing energy efficiency (EE) in downlink of multi-cell multi-user\ncoordinated cellular networks. An iterative algorithm based on fractional\nprogramming approach is introduced to solve the resulting non-convex\noptimization problem. In each iteration, users are clustered based on their\nelevation angle. Then, optimization of the tilt angle is carried out through a\nreduced complexity greedy search to find the best tilt angle for a given\nplacement of the users. Numerical results show that the proposed algorithm\nachieves higher EE compared to the 2D beamforming techniques.\n",
"title": "Joint Tilt Angle Adaptation and Beamforming in Multicell Multiuser Cellular Networks"
} | null | null | null | null | true | null | 19057 | null | Default | null | null |
null | {
"abstract": " We study sharp detection thresholds for degree corrections in Stochastic\nBlock Models in the context of a goodness of fit problem. When degree\ncorrections are relatively dense, a simple test based on the total number of\nedges is asymptotically optimal. For sparse degree corrections in non-dense\ngraphs, simple degree based Higher Criticism Test (Mukherjee, Mukherjee, Sen\n2016) is optimal with sharp constants. In contrast, for dense graphs, the\noptimal procedure runs in two stages. It involves running a suitable community\nrecovery algorithm in stage 1, followed by a Higher Criticism Test based on a\nlinear combination of within and across (estimated) community degrees in stage\n2. The necessity of the two step procedure is demonstrated by the failure of\nthe ordinary Maximum Degree Test in achieving sharp constants. As necessary\ntools we also derive asymptotic distribution of the Maximum Degree in\nStochastic Block Models along with moderate deviation and local central limit\ntype asymptotics of positive linear combinations of independent Binomial random\nvariables.\n",
"title": "Testing Degree Corrections in Stochastic Block Models"
} | null | null | [
"Mathematics",
"Statistics"
]
| null | true | null | 19058 | null | Validated | null | null |
null | {
"abstract": " Driven by the visions of Internet of Things and 5G communications, recent\nyears have seen a paradigm shift in mobile computing, from the centralized\nMobile Cloud Computing towards Mobile Edge Computing (MEC). The main feature of\nMEC is to push mobile computing, network control and storage to the network\nedges (e.g., base stations and access points) so as to enable\ncomputation-intensive and latency-critical applications at the resource-limited\nmobile devices. MEC promises dramatic reduction in latency and mobile energy\nconsumption, tackling the key challenges for materializing 5G vision. The\npromised gains of MEC have motivated extensive efforts in both academia and\nindustry on developing the technology. A main thrust of MEC research is to\nseamlessly merge the two disciplines of wireless communications and mobile\ncomputing, resulting in a wide-range of new designs ranging from techniques for\ncomputation offloading to network architectures. This paper provides a\ncomprehensive survey of the state-of-the-art MEC research with a focus on joint\nradio-and-computational resource management. We also present a research outlook\nconsisting of a set of promising directions for MEC research, including MEC\nsystem deployment, cache-enabled MEC, mobility management for MEC, green MEC,\nas well as privacy-aware MEC. Advancements in these directions will facilitate\nthe transformation of MEC from theory to practice. Finally, we introduce recent\nstandardization efforts on MEC as well as some typical MEC application\nscenarios.\n",
"title": "A Survey on Mobile Edge Computing: The Communication Perspective"
} | null | null | null | null | true | null | 19059 | null | Default | null | null |
null | {
"abstract": " Neural networks offer high-accuracy solutions to a range of problems, but are\ncostly to run in production systems because of computational and memory\nrequirements during a forward pass. Given a trained network, we propose a\ntechique called Deep Learning Approximation to build a faster network in a tiny\nfraction of the time required for training by only manipulating the network\nstructure and coefficients without requiring re-training or access to the\ntraining data. Speedup is achieved by by applying a sequential series of\nindependent optimizations that reduce the floating-point operations (FLOPs)\nrequired to perform a forward pass. First, lossless optimizations are applied,\nfollowed by lossy approximations using singular value decomposition (SVD) and\nlow-rank matrix decomposition. The optimal approximation is chosen by weighing\nthe relative accuracy loss and FLOP reduction according to a single parameter\nspecified by the user. On PASCAL VOC 2007 with the YOLO network, we show an\nend-to-end 2x speedup in a network forward pass with a 5% drop in mAP that can\nbe re-gained by finetuning.\n",
"title": "Deep Learning Approximation: Zero-Shot Neural Network Speedup"
} | null | null | [
"Statistics"
]
| null | true | null | 19060 | null | Validated | null | null |
null | {
"abstract": " Multirate digital signal processing and model reduction applications require\ncomputation of the frequency truncated norm of a discrete-time system. This\npaper explains how to compute the frequency truncated norm of a discrete-time\nsystem. To this end, a much-generalized problem of integrating a transfer\nfunction of a discrete-time system given in the descriptor form over an\ninterval of limited frequencies is also discussed along with its computation.\n",
"title": "Frequency truncated discrete-time system norm"
} | null | null | null | null | true | null | 19061 | null | Default | null | null |
null | {
"abstract": " Recent work has shown that state-of-the-art classifiers are quite brittle, in\nthe sense that a small adversarial change of an originally with high confidence\ncorrectly classified input leads to a wrong classification again with high\nconfidence. This raises concerns that such classifiers are vulnerable to\nattacks and calls into question their usage in safety-critical systems. We show\nin this paper for the first time formal guarantees on the robustness of a\nclassifier by giving instance-specific lower bounds on the norm of the input\nmanipulation required to change the classifier decision. Based on this analysis\nwe propose the Cross-Lipschitz regularization functional. We show that using\nthis form of regularization in kernel methods resp. neural networks improves\nthe robustness of the classifier without any loss in prediction performance.\n",
"title": "Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulation"
} | null | null | null | null | true | null | 19062 | null | Default | null | null |
null | {
"abstract": " For nanotechnology, the semiconductor device is scaled down dramatically with\nadditional strain engineering for device enhancement, the overall device\ncharacteristic is no longer dominated by the device size but also circuit\nlayout. The higher order layout effects, such as well proximity effect (WPE),\noxide spacing effect (OSE) and poly spacing effect (PSE), play an important\nrole for the device performance, it is critical to understand Design for\nManufacturability (DFM) impacts with various layout topology toward the overall\ncircuit performance. Currently, the layout effects (WPE, OSE and PSE) are\nvalidated through digital standard cell and analog differential pair test\nstructure. However, two analog layout structures: the guard ring and dummy fill\nimpact are not well studied yet, then, this paper describes the current mirror\ntest circuit to examine the guard ring and dummy fills DFM impacts using TSMC\n28nm HPM process.\n",
"title": "Complete DFM Model for High-Performance Computing SoCs with Guard Ring and Dummy Fill Effect"
} | null | null | null | null | true | null | 19063 | null | Default | null | null |
null | {
"abstract": " Location-based social network data offers the promise of collecting the data\nfrom a large base of users over a longer span of time at negligible cost. While\nseveral studies have applied social network data to activity and mobility\nanalysis, a comparison with travel diaries and general statistics has been\nlacking. In this paper, we analysed geo-referenced Twitter activities from a\nlarge number of users in Singapore and neighbouring countries. By combining\nthis data, population statistics and travel diaries and applying clustering\ntechniques, we addressed detection of activity locations, as well as spatial\nseparation and transitions between these locations. Kernel density estimation\nperforms best to detect activity locations due to the scattered nature of the\ntwitter data; more activity locations are detected per user than reported in\nthe travel survey. The descriptive analysis shows that determining home\nlocations is more difficult than detecting work locations for most planning\nzones. Spatial separations between detected activity locations from Twitter\ndata - as reported in a travel survey and captured by public transport smart\ncard data - are mostly similarly distributed, but also show relevant\ndifferences for very short and very long distances. This also holds for the\ntransitions between zones. Whether the differences between Twitter data and\nother data sources stem from differences in the population sub-sample,\nclustering methodology, or whether social networks are being used significantly\nmore at specific locations must be determined by further research. Despite\nthese shortcomings, location-based social network data offers a promising data\nsource for insights into activity locations and mobility patterns, especially\nfor regions where travel survey data is not readily available.\n",
"title": "Investigating the potential of social network data for transport demand models"
} | null | null | null | null | true | null | 19064 | null | Default | null | null |
null | {
"abstract": " Representing large-scale motions and topological changes in the finite volume\n(FV) framework, while at the same time preserving the accuracy of the numerical\nsolution, is difficult. In this paper, we present a robust, highly efficient\nmethod designed to achieve this capability. The proposed approach conceptually\nshares many of the characteristics of the cut-cell interface tracking method,\nbut without the need for complex cell splitting/merging operations. The heart\nof the new technique is to align existing mesh facets with the geometry to be\nrepresented. We then modify the matrix contributions from these facets such\nthat they are represented in an identical fashion to traditional boundary\nconditions. The collection of such faces is named a Generalised Internal\nBoundary (GIB). In order to introduce motion into the system, we rely on the\nclassical ALE (Arbitrary Lagrangian-Eulerian) approach, but with the caveat\nthat the non-time-dependent motion of elements instantaneously crossing the\ninterface is handled separately from the time dependent component. The new\nmethodology is validated through comparison with: a) a body-fitted grid\nsimulation of an oscillating two dimensional cylinder and b) experimental\nresults of a butterfly valve.\n",
"title": "Generalized Internal Boundaries (GIB)"
} | null | null | null | null | true | null | 19065 | null | Default | null | null |
null | {
"abstract": " A strong certification process is required to insure the safety of airplanes,\nand more specifically the robustness of avionics applications. To implement\nthis process, the development of avionics software must follow long and costly\nprocedures. Most of these procedures have to be reexecuted each time the\nsoftware is modified. In this paper, we propose a framework to reduce the cost\nand time impact of a software modification. With this new approach, the piece\nof software likely to change is isolated from the rest of the application, so\nit can be certified independently. This helps the system integrator to adapt an\navionics application to the specificities of the target airplane, without the\nneed for a new certification of the application.\n",
"title": "Reducing Certification Granularity to Increase Adaptability of Avionics Software"
} | null | null | null | null | true | null | 19066 | null | Default | null | null |
null | {
"abstract": " Evolutionary game dynamics in structured populations has been extensively\nexplored in past decades. However, most previous studies assume that payoffs of\nindividuals are fully determined by the strategic behaviors of interacting\nparties and social ties between them only serve as the indicator of the\nexistence of interactions. This assumption neglects important information\ncarried by inter-personal social ties such as genetic similarity, geographic\nproximity, and social closeness, which may crucially affect the outcome of\ninteractions. To model these situations, we present a framework of evolutionary\nmultiplayer games on graphs with edge diversity, where different types of edges\ndescribe diverse social ties. Strategic behaviors together with social ties\ndetermine the resulting payoffs of interactants. Under weak selection, we\nprovide a general formula to predict the success of one behavior over the\nother. We apply this formula to various examples which cannot be dealt with\nusing previous models, including the division of labor and relationship- or\nedge-dependent games. We find that labor division facilitates collective\ncooperation by decomposing a many-player game into several games of smaller\nsizes. The evolutionary process based on relationship-dependent games can be\napproximated by interactions under a transformed and unified game. Our work\nstresses the importance of social ties and provides effective methods to reduce\nthe calculating complexity in analyzing the evolution of realistic systems.\n",
"title": "Evolutionary multiplayer games on graphs with edge diversity"
} | null | null | null | null | true | null | 19067 | null | Default | null | null |
null | {
"abstract": " In this paper, we study the robust consensus problem for a set of\ndiscrete-time linear agents to coordinate over an uncertain communication\nnetwork, which is to achieve consensus against the transmission errors and\nnoises resulted from the information exchange between the agents. We model the\nnetwork by means of communication links subject to multiplicative stochastic\nuncertainties, which are susceptible to describing packet dropout, random\ndelay, and fading phenomena. Different communication topologies, such as\nundirected graphs and leader-follower graphs, are considered. We derive\nsufficient conditions for robust consensus in the mean square sense. This\nresults unveil intrinsic constraints on consensus attainment imposed by the\nnetwork synchronizability, the unstable agent dynamics, and the channel\nuncertainty variances. Consensus protocols are designed based on the state\ninformation transmitted over the uncertain channels, by solving a modified\nalgebraic Riccati equation.\n",
"title": "Robust Consensus for Multi-Agent Systems Communicating over Stochastic Uncertain Networks"
} | null | null | null | null | true | null | 19068 | null | Default | null | null |
null | {
"abstract": " We analyze the \"higher rank\" gauge theories, that capture some of the\nphenomenology of the Fracton order. It is shown that these theories loose gauge\ninvariance when arbitrarily weak and smooth curvature is introduced. We propose\na resolution to this problem by introducing a theory invariant under\narea-preserving diffeomorphisms, which reduce to the \"higher rank\" gauge\ntransformations upon linearization around a flat background. The proposed\ntheory is \\emph{geometric} in nature and is interpreted as a theory of\n\\emph{fractional topological elasticity}. This theory exhibits the Fracton\nphenomenology. We explore the conservation laws, topological excitations,\nlinear response, various kinematical constraints, and canonical structure of\nthe theory. Finally, we emphasize that the very structure of Riemann-Cartan\ngeometry, which we use to formulate the theory, encodes the Fracton\nphenomenology, suggesting that the Fracton order itself is \\emph{geometric} in\nnature.\n",
"title": "Fractional Topological Elasticity and Fracton Order"
} | null | null | null | null | true | null | 19069 | null | Default | null | null |
null | {
"abstract": " Let $K$ be a local field whose residue field has characteristic $p$ and let\n$L/K$ be a finite separable totally ramified extension. Let $\\pi_L$ be a\nuniformizer for $L$ and let $f(X)$ be the minimum polynomial for $\\pi_L$ over\n$K$. Suppose $\\tilde{\\pi}_L$ is another uniformizer for $L$ such that\n$\\tilde{\\pi}_L\\equiv\\pi_L+r\\pi_L^{\\ell+1} \\pmod{\\pi_L^{\\ell+2}}$ for some\n$\\ell\\ge1$ and $r\\in O_K$. Let $\\tilde{f}(X)$ be the minimum polynomial for\n$\\tilde{\\pi}_L$ over $K$. In this paper we give congruences for the\ncoefficients of $\\tilde{f}(X)$ in terms of $r$ and the coefficients of $f(X)$.\nThese congruences improve and extend work of Krasner.\n",
"title": "Perturbing Eisenstein polynomials over local fields"
} | null | null | null | null | true | null | 19070 | null | Default | null | null |
null | {
"abstract": " Galactic orbits have been constructed over long time intervals for ten\nglobular clusters located near the Galactic center. A model with an axially\nsymmetric gravitational potential for the Galaxy was initially applied, after\nwhich a non-axially symmetric potential corresponding to the central bar was\nadded. Variations in the trajectories of all these globular clusters in the XY\nplane due to the influence of the bar were detected. These were greatest for\nthe cluster Terzan 4 in the meridional (RZ) plane. The globular clusters Terzan\n1, Terzan 2, Terzan 4, Terzan 9, NGC 6522, and NGC 6558 always remained within\nthe Galactic bulge, no farther than 4 kpc from the Galactic center.\n",
"title": "Galactic Orbits of Globular Clusters in the Region of the Galactic Bulge"
} | null | null | null | null | true | null | 19071 | null | Default | null | null |
null | {
"abstract": " Motion planning for underwater vehicles must consider the effect of ocean\ncurrents. We present an efficient method to compute reachability and cost\nbetween sample points in sampling-based motion planning that supports\nlong-range planning over hundreds of kilometres in complicated flows. The idea\nis to search a reduced space of control inputs that consists of stream\nfunctions whose level sets, or streamlines, optimally connect two given points.\nSuch stream functions are generated by superimposing a control input onto the\nunderlying current flow. A streamline represents the resulting path that a\nvehicle would follow as it is carried along by the current given that control\ninput. We provide rigorous analysis that shows how our method avoids exhaustive\nsearch of the control space, and demonstrate simulated examples in complicated\nflows including a traversal along the east coast of Australia, using actual\ncurrent predictions, between Sydney and Brisbane.\n",
"title": "Streamlines for Motion Planning in Underwater Currents"
} | null | null | [
"Computer Science"
]
| null | true | null | 19072 | null | Validated | null | null |
null | {
"abstract": " Hot, Dust-Obscured Galaxies, or \"Hot DOGs\", are a rare, dusty, hyperluminous\ngalaxy population discovered by the WISE mission. Predominantly at redshifts\n2-3, they include the most luminous known galaxies in the universe. Their high\nluminosities likely come from accretion onto highly obscured super massive\nblack holes (SMBHs). We have conducted a pilot survey to measure the SMBH\nmasses of five z~2 Hot DOGs via broad H_alpha emission lines, using\nKeck/MOSFIRE and Gemini/FLAMINGOS-2. We detect broad H_alpha emission in all\nfive Hot DOGs. We find substantial corresponding SMBH masses for these Hot DOGs\n(~ 10^{9} M_sun), and their derived Eddington ratios are close to unity. These\nz~2 Hot DOGs are the most luminous AGNs at given BH masses, suggesting they are\naccreting at the maximum rates for their BHs. A similar property is found for\nknown z~6 quasars. Our results are consistent with scenarios in which Hot DOGs\nrepresent a transitional, high-accretion phase between obscured and unobscured\nquasars. Hot DOGs may mark a special evolutionary stage before the red quasar\nand optical quasar phases, and they may be present at other cosmic epochs.\n",
"title": "Eddington-Limited Accretion in z~2 WISE-selected Hot, Dust-Obscured Galaxies"
} | null | null | null | null | true | null | 19073 | null | Default | null | null |
null | {
"abstract": " Surface scattering is the key limiting factor to thermal transport in\ndielectric crystals as the length scales are reduced or when temperature is\nlowered. To explain this phenomenon, it is commonly assumed that the mean free\npaths of heat carriers are bound by the crystal size and that thermal\nconductivity is reduced in a manner proportional to such mean free paths. We\nshow here that these conclusions rely on simplifying assumptions and\napproximated transport models. Instead, starting from the linearized Boltzmann\ntransport equation in the relaxon basis, we show how the problem can be reduced\nto a set of decoupled linear differential equations. Then, the heat flow can be\ninterpreted as a hydrodynamic phenomenon, with the relaxon gas being slowed\ndown in proximity of a surface by friction effects, similar to the flux of a\nviscous fluid in a pipe. As an example, we study a ribbon and a trench of\nmonolayer molybdenum disulphide, describing the procedure to reconstruct the\ntemperature and thermal conductivity profile in the sample interior and showing\nhow to estimate the effect of nanostructuring. The approach is general and\ncould be extended to other transport carriers, such as electrons, or extended\nto materials of higher dimensionality and to different geometries, such as thin\nfilms.\n",
"title": "Boltzmann Transport in Nanostructures as a Friction Effect"
} | null | null | null | null | true | null | 19074 | null | Default | null | null |
null | {
"abstract": " Systems Engineering (SE) is the set of processes and documentation required\nfor successfully realising large-scale engineering projects, but the classical\napproach is not a good fit for software-intensive projects, especially when the\nneeds of the different stakeholders are not fully known from the beginning, and\nrequirement priorities might change. The SKA is the ultimate software-enabled\ntelescope, with enormous amounts of computing hardware and software required to\nperform its data reduction. We give an overview of the system and software\nengineering processes in the SKA1 development, and the tension between\nclassical and agile SE.\n",
"title": "Agile Software Engineering and Systems Engineering at SKA Scale"
} | null | null | null | null | true | null | 19075 | null | Default | null | null |
null | {
"abstract": " We reduce the exponent in the error term of the prime geodesic theorem for\ncompact Riemann surfaces from $\\frac{3}{4}$ to $\\frac{7}{10}$ outside a set of\nfinite logarithmic measure.\n",
"title": "Prime geodesic theorem of Gallagher type"
} | null | null | null | null | true | null | 19076 | null | Default | null | null |
null | {
"abstract": " This paper introduces PriMaL, a general PRIvacy-preserving MAchine-Learning\nmethod for reducing the privacy cost of information transmitted through a\nnetwork. Distributed sensor networks are often used for automated\nclassification and detection of abnormal events in high-stakes situations, e.g.\nfire in buildings, earthquakes, or crowd disasters. Such networks might\ntransmit privacy-sensitive information, e.g. GPS location of smartphones, which\nmight be disclosed if the network is compromised. Privacy concerns might slow\ndown the adoption of the technology, in particular in the scenario of social\nsensing where participation is voluntary, thus solutions are needed which\nimprove privacy without compromising on the event detection accuracy. PriMaL is\nimplemented as a machine-learning layer that works on top of an existing event\ndetection algorithm. Experiments are run in a general simulation framework, for\nseveral network topologies and parameter values. The privacy footprint of\nstate-of-the-art event detection algorithms is compared within the proposed\nframework. Results show that PriMaL is able to reduce the privacy cost of a\ndistributed event detection algorithm below that of the corresponding\ncentralized algorithm, within the bounds of some assumptions about the\nprotocol. Moreover the performance of the distributed algorithm is not\nstatistically worse than that of the centralized algorithm.\n",
"title": "PriMaL: A Privacy-Preserving Machine Learning Method for Event Detection in Distributed Sensor Networks"
} | null | null | [
"Computer Science"
]
| null | true | null | 19077 | null | Validated | null | null |
null | {
"abstract": " The online problem of computing the top eigenvector is fundamental to machine\nlearning. In both adversarial and stochastic settings, previous results (such\nas matrix multiplicative weight update, follow the regularized leader, follow\nthe compressed leader, block power method) either achieve optimal regret but\nrun slow, or run fast at the expense of loosing a $\\sqrt{d}$ factor in total\nregret where $d$ is the matrix dimension.\nWe propose a $\\textit{follow-the-compressed-leader (FTCL)}$ framework which\nachieves optimal regret without sacrificing the running time. Our idea is to\n\"compress\" the matrix strategy to dimension 3 in the adversarial setting, or\ndimension 1 in the stochastic setting. These respectively resolve two open\nquestions regarding the design of optimal and efficient algorithms for the\nonline eigenvector problem.\n",
"title": "Follow the Compressed Leader: Faster Online Learning of Eigenvectors and Faster MMWU"
} | null | null | null | null | true | null | 19078 | null | Default | null | null |
null | {
"abstract": " Using Maple, we implement a SAT solver based on the principle of\ninclusion-exclusion and the Bonferroni inequalities. Using randomly generated\ninput, we investigate the performance of our solver as a function of the number\nof variables and number of clauses. We also test it against Maple's built-in\ntautology procedure. Finally, we implement the Lovász local lemma with Maple\nand discuss its applicability to SAT.\n",
"title": "Solving satisfiability using inclusion-exclusion"
} | null | null | [
"Computer Science"
]
| null | true | null | 19079 | null | Validated | null | null |
null | {
"abstract": " We consider submanifolds into Riemannian manifold with metallic structures.\nWe obtain some new results for hypersurfaces in these spaces and we express the\nfundamental theorem of submanifolds into products spaces in terms of metallic\nstructures. Moreover, we define new structures called complex metallic\nstructures. We show that these structures are linked with complex structures.\nThen, we consider submanifolds into Riemannian manifold with such structures\nwith a focus on invariant submanifolds and hypersurfaces. We also express in\nparticular the fundamental theorem of submanifolds of complex space form in\nterms of complex metallic structures.\n",
"title": "Isometric immersions into manifolds with metallic structures"
} | null | null | [
"Mathematics"
]
| null | true | null | 19080 | null | Validated | null | null |
null | {
"abstract": " Superconducting electronic devices have re-emerged as contenders for both\nclassical and quantum computing due to their fast operation speeds, low\ndissipation and long coherence times. An ultimate demonstration of coherence is\nlasing. We use one of the fundamental aspects of superconductivity, the ac\nJosephson effect, to demonstrate a laser made from a Josephson junction\nstrongly coupled to a multi-mode superconducting cavity. A dc voltage bias to\nthe junction provides a source of microwave photons, while the circuit's\nnonlinearity allows for efficient down-conversion of higher order Josephson\nfrequencies down to the cavity's fundamental mode. The simple fabrication and\noperation allows for easy integration with a range of quantum devices, allowing\nfor efficient on-chip generation of coherent microwave photons at low\ntemperatures.\n",
"title": "Demonstration of an ac Josephson junction laser"
} | null | null | null | null | true | null | 19081 | null | Default | null | null |
null | {
"abstract": " These notes correspond to a mini-course given at the Poisson 2016 conference\nin Geneva. Starting from classical integrable systems in the sense of\nLiouville, we explore the notion of non-degenerate singularity and expose\nrecent research in connection with semi-toric systems. The quantum and\nsemiclassical counterpart will also be presented, in the viewpoint of the\ninverse question: from the quantum mechanical spectrum, can you recover the\nclassical system?\n",
"title": "Integrable systems, symmetries and quantization"
} | null | null | null | null | true | null | 19082 | null | Default | null | null |
null | {
"abstract": " Investigating Friedel oscillations in ultracold gases would complement the\nstudies performed on solid state samples with scanning-tunneling microscopes.\nIn atomic quantum gases interactions and external potentials can be tuned\nfreely and the inherently slower dynamics allow to access non-equilibrium\ndynamics following a potential or interaction quench. Here, we examine how\nFriedel oscillations can be observed in current ultracold gas experiments under\nrealistic conditions. To this aim we numerically calculate the amplitude of the\nFriedel oscillations which a potential barrier provokes in a 1D Fermi gas and\ncompare it to the expected atomic and photonic shot noise in a density\nmeasurement. We find that to detect Friedel oscillations the signal from\nseveral thousand one-dimensional systems has to be averaged. However, as up to\n100 parallel one-dimensional systems can be prepared in a single run with\npresent experiments, averaging over about 100 images is sufficient.\n",
"title": "Detecting Friedel oscillations in ultracold Fermi gases"
} | null | null | null | null | true | null | 19083 | null | Default | null | null |
null | {
"abstract": " While variational methods have been among the most powerful tools for solving\nlinear inverse problems in imaging, deep (convolutional) neural networks have\nrecently taken the lead in many challenging benchmarks. A remaining drawback of\ndeep learning approaches is their requirement for an expensive retraining\nwhenever the specific problem, the noise level, noise type, or desired measure\nof fidelity changes. On the contrary, variational methods have a plug-and-play\nnature as they usually consist of separate data fidelity and regularization\nterms.\nIn this paper we study the possibility of replacing the proximal operator of\nthe regularization used in many convex energy minimization algorithms by a\ndenoising neural network. The latter therefore serves as an implicit natural\nimage prior, while the data term can still be chosen independently. Using a\nfixed denoising neural network in exemplary problems of image deconvolution\nwith different blur kernels and image demosaicking, we obtain state-of-the-art\nreconstruction results. These indicate the high generalizability of our\napproach and a reduction of the need for problem-specific training.\nAdditionally, we discuss novel results on the analysis of possible optimization\nalgorithms to incorporate the network into, as well as the choices of algorithm\nparameters and their relation to the noise level the neural network is trained\non.\n",
"title": "Learning Proximal Operators: Using Denoising Networks for Regularizing Inverse Imaging Problems"
} | null | null | null | null | true | null | 19084 | null | Default | null | null |
null | {
"abstract": " Metric learning aims at learning a distance which is consistent with the\nsemantic meaning of the samples. The problem is generally solved by learning an\nembedding for each sample such that the embeddings of samples of the same\ncategory are compact while the embeddings of samples of different categories\nare spread-out in the feature space. We study the features extracted from the\nsecond last layer of a deep neural network based classifier trained with the\ncross entropy loss on top of the softmax layer. We show that training\nclassifiers with different temperature values of softmax function leads to\nfeatures with different levels of compactness. Leveraging these insights, we\npropose a \"heating-up\" strategy to train a classifier with increasing\ntemperatures, leading the corresponding embeddings to achieve state-of-the-art\nperformance on a variety of metric learning benchmarks.\n",
"title": "Heated-Up Softmax Embedding"
} | null | null | null | null | true | null | 19085 | null | Default | null | null |
null | {
"abstract": " This paper introduces a fast algorithm for simultaneous inversion and\ndeterminant computation of small sized matrices in the context of fully\nPolarimetric Synthetic Aperture Radar (PolSAR) image processing and analysis.\nThe proposed fast algorithm is based on the computation of the adjoint matrix\nand the symmetry of the input matrix. The algorithm is implemented in a general\npurpose graphical processing unit (GPGPU) and compared to the usual approach\nbased on Cholesky factorization. The assessment with simulated observations and\ndata from an actual PolSAR sensor show a speedup factor of about two when\ncompared to the usual Cholesky factorization. Moreover, the expressions\nprovided here can be implemented in any platform.\n",
"title": "Fast Matrix Inversion and Determinant Computation for Polarimetric Synthetic Aperture Radar"
} | null | null | null | null | true | null | 19086 | null | Default | null | null |
null | {
"abstract": " (abridged) We investigate the signatures left by the cosmic neutrino\nbackground on the clustering of matter, CDM+baryons and halos in redshift-space\nusing a set of more than 1000 N-body and hydrodynamical simulations with\nmassless and massive neutrinos. We find that the effect neutrinos induce on the\nclustering of CDM+baryons in redshift-space on small scales is almost entirely\ndue to the change in $\\sigma_8$. Neutrinos imprint a characteristic signature\nin the quadrupole of the matter (CDM+baryons+neutrinos) field on small scales,\nthat can be used to disentangle the effect of $\\sigma_8$ and $M_\\nu$. We show\nthat the effect of neutrinos on the clustering of halos is very different, on\nall scales, to the one induced by $\\sigma_8$. We find that the effects of\nneutrinos of the growth rate of CDM+baryons ranges from $\\sim0.3\\%$ to $2\\%$ on\nscales $k\\in[0.01, 0.5]~h{\\rm Mpc}^{-1}$ for neutrinos with masses $M_\\nu\n\\leqslant 0.15$ eV. We compute the bias between the momentum of halos and the\nmomentum of CDM+baryon and find it to be 1 on large scales for all models with\nmassless and massive neutrinos considered. This point towards a velocity bias\nbetween halos and total matter on large scales that it is important to account\nfor in order to extract unbiased neutrino information from velocity/momentum\nsurveys such as kSZ observations. We show that baryonic effects can affect the\nclustering of matter and CDM+baryons in redshift-space by up to a few percent\ndown to $k=0.5~h{\\rm Mpc}^{-1}$. We find that hydrodynamics and astrophysical\nprocesses, as implemented in our simulations, only distort the relative effect\nthat neutrinos induce on the anisotropic clustering of matter, CDM+baryons and\nhalos in redshift-space by less than $1\\%$. Thus, the effect of neutrinos in\nthe fully non-linear regime can be written as a transfer function with very\nweak dependence on astrophysics.\n",
"title": "The imprint of neutrinos on clustering in redshift-space"
} | null | null | null | null | true | null | 19087 | null | Default | null | null |
null | {
"abstract": " Data analytics (such as association rule mining and decision tree mining) can\ndiscover useful statistical knowledge from a big data set. But protecting the\nprivacy of the data provider and the data user in the process of analytics is a\nserious issue. Usually, the privacy of both parties cannot be fully protected\nsimultaneously by a classical algorithm. In this paper, we present a quantum\nprotocol for data mining that can much better protect privacy than the known\nclassical algorithms: (1) if both the data provider and the data user are\nhonest, the data user can know nothing about the database except the\nstatistical results, and the data provider can get nearly no information about\nthe results mined by the data user; (2) if the data user is dishonest and tries\nto disclose private information of the other, she/he will be detected with a\nhigh probability; (3) if the data provider tries to disclose the privacy of the\ndata user, she/he cannot get any useful information since the data user hides\nhis privacy among noises.\n",
"title": "Quantum Privacy-Preserving Data Analytics"
} | null | null | null | null | true | null | 19088 | null | Default | null | null |
null | {
"abstract": " Using the first-principles and Monte Carlo methods, here we systematically\nstudy magnetic properties of monolayer octagonal-tetragonal phosphorus with 3d\ntransition-metal (TM) adatoms. Different from the puckered hexagonal black\nphosphorus monolayer (phosphorene or $\\alpha$-P), the octagonal-tetragonal\nphase of 2D phosphorus (named as ot-P or $\\epsilon$-P in this article) is\nbuckled with octagon-tetragon structure. Our calculations show that all TMs,\nexcept the closed-shell Zn atom, are able to strongly bind onto monolayer\n$ot$-P with significant binding energies. Local magnetic moments (up to 6\n$\\mu$B) on adatoms of Sc, Ti, V, Cr, Mn, Fe and Co originate from the exchange\nand crystal-field splitting of TM 3d orbitals. The magnetic coupling between\nlocalized magnetic states of adatoms is dependent on adatomic distances and\ndirections. Lastly, the uniformly magnetic order is investigated to screening\ntwo-dimensional dilute ferromagnets with high Curie temperature for\napplications of spintronics. It is found that ot-P with V atoms homogeneously\nadsorbed at the centre of octagons with a concentration of 5% has the most\nstable ferromagnetic ground state. Its Curie temperature is estimated to be 173\nK using the Monte Carlo method.\n",
"title": "Magnetic Properties of Transition-Metal Adsorbed ot-Phosphorus Monolayer: A First-principles and Monte Carlo Study"
} | null | null | null | null | true | null | 19089 | null | Default | null | null |
null | {
"abstract": " In this note we present a fast algorithm that finds for any $r$ the number\n$N_r$ of $\\mathbb{F}_{q^r}$ rational points on a smooth absolutely irreducible\ncurve $C$ defined over $\\mathbb{F}_{q}$ assuming that we know $N_1,\\cdots,N_g$,\nwhere $g$ is the genus of $C$. The proof of its validity is given in detail and\nits working are illustrated with several examples. In an Appendix we list the\nPython function in which we have implemented the algorithm together with other\nroutines used in the examples.\n",
"title": "A bootstrap for the number of $\\mathbb{F}_{q^r}$-rational points on a curve over $\\mathbb{F}_q$"
} | null | null | null | null | true | null | 19090 | null | Default | null | null |
null | {
"abstract": " We investigate the role of transition metal atoms of group V-b (V, Nb, and\nTa) and VI-b (Cr, Mo, and W) as n- or p-type dopants in anatase TiO2 using\nthermodynamic principles and density functional theory with the\nHeyd-Scuseria-Ernzerhof HSE06 hybrid functional. The HSE06 functional provides\na realistic value for the band gap, which ensures a correct classification of\ndopants as shallow or deep donors or acceptors. Defect formation energies and\nthermodynamic transition levels are calculated taking into account the\nconstraints imposed by the stability of TiO2 and the solubility limit of the\nimpurities. Nb, Ta, W and Mo are identified as shallow donors. Although W\nprovides two electrons, Nb and Ta show a considerably lower formation energy,\nin particular under O-poor conditions. Mo donates in principle one electron,\nbut under specific conditions can turn into a double donor. V impurities are\ndeep donors and Cr shows up as an amphoteric defect, thereby acting as an\nelectron trapping center in n-type TiO2 especially under O-rich conditions. A\ncomparison with the available experimental data yields excellent agreement.\n",
"title": "Doping anatase TiO2 with group V-b and VI-b transition metal atoms: a hybrid functional first-principles study"
} | null | null | null | null | true | null | 19091 | null | Default | null | null |
null | {
"abstract": " We introduce an algorithm to generate (not solve) spin-glass instances with\nplanted solutions of arbitrary size and structure. First, a set of small\nproblem patches with open boundaries is solved either exactly or with a\nheuristic, and then the individual patches are stitched together to create a\nlarge problem with a known planted solution. Because in these problems\nfrustration is typically smaller than in random problems, we first assess the\ntypical computational complexity of the individual patches using population\nannealing Monte Carlo, and introduce an approach that allows one to fine-tune\nthe typical computational complexity of the patch-planted system. The scaling\nof the typical computational complexity of these planted instances with various\nnumbers of patches and patch sizes is investigated and compared to random\ninstances.\n",
"title": "Patch-planting spin-glass solution for benchmarking"
} | null | null | null | null | true | null | 19092 | null | Default | null | null |
null | {
"abstract": " The processes that led to the formation of the planetary bodies in the Solar\nSystem are still not fully understood. Using the results obtained with the\ncomprehensive suite of instruments on-board ESA's Rosetta mission, we present\nevidence that comet 67P/Churyumov-Gerasimenko likely formed through the gentle\ngravitational collapse of a bound clump of mm-sized dust aggregates\n(\"pebbles\"), intermixed with microscopic ice particles. This formation scenario\nleads to a cometary make-up that is simultaneously compatible with the global\nporosity, homogeneity, tensile strength, thermal inertia, vertical temperature\nprofiles, sizes and porosities of emitted dust, and the steep increase in\nwater-vapour production rate with decreasing heliocentric distance, measured by\nthe instruments on-board the Rosetta spacecraft and the Philae lander. Our\nfindings suggest that the pebbles observed to be abundant in protoplanetary\ndiscs around young stars provide the building material for comets and other\nminor bodies.\n",
"title": "Evidence for the formation of comet 67P/Churyumov-Gerasimenko through gravitational collapse of a bound clump of pebbles"
} | null | null | null | null | true | null | 19093 | null | Default | null | null |
null | {
"abstract": " Galaxy clusters are thought to grow by accreting mass through large-scale,\nstrong, yet elusive, virial shocks. Such a shock is expected to accelerate\nrelativistic electrons, thus generating a spectrally-flat leptonic virial-ring.\nHowever, until now, only the nearby Coma cluster has shown evidence for a\n$\\gamma$-ray virial ring. We stack Fermi-LAT data for the 112 most massive,\nhigh latitude, extended clusters, enhancing the ring sensitivity by rescaling\nclusters to their virial radii and utilizing the expected flat energy spectrum.\nIn addition to a central unresolved, hard signal (detected at the $\\sim\n5.8\\sigma$ confidence level), probably dominated by AGN, we identify (at the\n$5.8\\sigma$ confidence level) a bright, spectrally-flat $\\gamma$-ray ring at\nthe expected virial shock position. The ring signal implies that the shock\ndeposits $\\sim 0.6\\%$ (with an interpretation uncertainty factor $\\sim2$) of\nthe thermal energy in relativistic electrons over a Hubble time. This result,\nconsistent with the Coma signal, validates and calibrates the virial shock\nmodel, and indicates that the cumulative emission from such shocks\nsignificantly contributes to the diffuse extragalactic $\\gamma$-ray and\nlow-frequency radio backgrounds.\n",
"title": "Detection of virial shocks in stacked Fermi-LAT clusters"
} | null | null | null | null | true | null | 19094 | null | Default | null | null |
null | {
"abstract": " A metamaterial made by stacked hole-array layers known as a fishnet\nmetamaterial behaves as a hyperbolic metamaterial at wavelength much longer\nthan hole-array period. However, the analytical formulae of effective\nparameters of a fishnet metamaterial have not been reported hindering the\ndesign of deep-subwavelength imaging devices using this structure. We report\nthe new closed-form formulae of effective parameters comprising anisotropic\ndispersion relation of a fishnet metamaterial working at terahertz or microwave\nfrequency. These effective parameters of a fishnet metamaterial are consistent\nwith those obtained by quasi-full solutions using known effective parameters of\na hole-array layer working at frequency below its spoof plasma frequency with\nthe superlattice period much smaller than the hole-array period. We also\ntheoretically demonstrate the deep-subwavelength focusing at {\\lambda}/83 using\nthe composite structure of a slit-array layer and a fishnet metamaterial. It is\nfound that the focused intensity inside a fishnet metamaterial is several times\nlarger than that without the fishnet metamaterial, but the transmitted\nintensity is still restricted by large-wavevector difference in air and a\nfishnet metamaterial. Our effective parameters may aid the next-generation\ndeep-subwavelength imaging devices working at terahertz or microwave radiation.\n",
"title": "Closed-form formulae of hyperbolic metamaterial made by stacked hole-array layers working at terahertz or microwave radiation"
} | null | null | null | null | true | null | 19095 | null | Default | null | null |
null | {
"abstract": " We develop a novel policy synthesis algorithm, RMPflow, based on\ngeometrically consistent transformations of Riemannian Motion Policies (RMPs).\nRMPs are a class of reactive motion policies designed to parameterize\nnon-Euclidean behaviors as dynamical systems in intrinsically nonlinear task\nspaces. Given a set of RMPs designed for individual tasks, RMPflow can\nconsistently combine these local policies to generate an expressive global\npolicy, while simultaneously exploiting sparse structure for computational\nefficiency. We study the geometric properties of RMPflow and provide sufficient\nconditions for stability. Finally, we experimentally demonstrate that\naccounting for the geometry of task policies can simplify classically difficult\nproblems, such as planning through clutter on high-DOF manipulation systems.\n",
"title": "RMPflow: A Computational Graph for Automatic Motion Policy Generation"
} | null | null | null | null | true | null | 19096 | null | Default | null | null |
null | {
"abstract": " Reduced motor control is one of the most frequent features associated with\naging and disease. Nonlinear and fractal analyses have proved to be useful in\ninvestigating human physiological alterations with age and disease. Similar\nfindings have not been established for any of the model organisms typically\nstudied by biologists, though. If the physiology of a simpler model organism\ndisplays the same characteristics, this fact would open a new research window\non the control mechanisms that organisms use to regulate physiological\nprocesses during aging and stress. Here, we use a recently introduced animal\ntracking technology to simultaneously follow tens of Caenorhabdits elegans for\nseveral hours and use tools from fractal physiology to quantitatively evaluate\nthe effects of aging and temperature stress on nematode motility. Similarly to\nhuman physiological signals, scaling analysis reveals long-range correlations\nin numerous motility variables, fractal properties in behavioral shifts, and\nfluctuation dynamics over a wide range of timescales. These properties change\nas a result of a superposition of age and stress-related adaptive mechanisms\nthat regulate motility.\n",
"title": "Long-range correlations and fractal dynamics in C. elegans: changes with aging and stress"
} | null | null | null | null | true | null | 19097 | null | Default | null | null |
null | {
"abstract": " Recent work of M.D. Johnston et al. has produced sufficient conditions on the\nstructure of a chemical reaction network which guarantee that the corresponding\ndiscrete state space system exhibits an extinction event. The conditions\nconsist of a series of systems of equalities and inequalities on the edges of a\nmodified reaction network called a domination-expanded reaction network. In\nthis paper, we present a computational implementation of these conditions\nwritten in Python and apply the program on examples drawn from the biochemical\nliterature, including a model of polyamine metabolism in mammals and a model of\nthe pentose phosphate pathway in Trypanosoma brucei. We also run the program on\n458 models from the European Bioinformatics Institute's BioModels Database and\nreport our results.\n",
"title": "A Computational Approach to Extinction Events in Chemical Reaction Networks with Discrete State Spaces"
} | null | null | null | null | true | null | 19098 | null | Default | null | null |
null | {
"abstract": " A person's weight status can have profound implications on their life,\nranging from mental health, to longevity, to financial income. At the societal\nlevel, \"fat shaming\" and other forms of \"sizeism\" are a growing concern, while\nincreasing obesity rates are linked to ever raising healthcare costs. For these\nreasons, researchers from a variety of backgrounds are interested in studying\nobesity from all angles. To obtain data, traditionally, a person would have to\naccurately self-report their body-mass index (BMI) or would have to see a\ndoctor to have it measured. In this paper, we show how computer vision can be\nused to infer a person's BMI from social media images. We hope that our tool,\nwhich we release, helps to advance the study of social aspects related to body\nweight.\n",
"title": "Face-to-BMI: Using Computer Vision to Infer Body Mass Index on Social Media"
} | null | null | [
"Computer Science"
]
| null | true | null | 19099 | null | Validated | null | null |
null | {
"abstract": " This paper proposes a new method which builds a simplex based approximation\nof a $d-1$-dimensional manifold $M$ separating a $d$-dimensional compact set\ninto two parts, and an efficient algorithm classifying points according to this\napproximation. In a first variant, the approximation is made of simplices that\nare defined in the cubes of a regular grid covering the compact set, from\nboundary points that approximate the intersection between $M$ and the edges of\nthe cubes. All the simplices defined in a cube share the barycentre of the\nboundary points located in the cube and include simplices similarly defined in\ncube facets, and so on recursively. In a second variant, the Kuhn triangulation\nis used to break the cubes into simplices and the approximation is defined in\nthese simplices from the boundary points computed on their edges, with the same\nprinciple. Both the approximation in cubes and in simplices define a separating\nsurface on the whole grid and classifying a point on one side or the other of\nthis surface requires only a small number (at most $d$) of simple tests. Under\nsome conditions on the definition of the boundary points and on the reach of\n$M$, for both variants the Hausdorff distance between $M$ and its approximation\ndecreases like $\\mathcal{O}(d n_G^{-2})$, where $n_G$ is the number of points\non each axis of the grid. The approximation in cubes requires computing less\nboundary points than the approximation in simplices but the latter is always a\nmanifold and is more accurate for a given value of $n_G$. The paper reports\ntests of the method when varying $n_G$ and the dimensionality of the space (up\nto 9).\n",
"title": "Recursive simplex stars"
} | null | null | [
"Computer Science"
]
| null | true | null | 19100 | null | Validated | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.