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": " Let $X\\rightarrow {\\mathbb P}^1$ be an elliptically fibered $K3$ surface with\na section, admitting a sequence of Ricci-flat metrics collapsing the fibers.\nLet $\\mathcal E$ be a generic, holomoprhic $SU(n)$ bundle over $X$ such that\nthe restriction of $\\mathcal E$ to each fiber is semi-stable. Given a sequence\n$\\Xi_i$ of Hermitian-Yang-Mills connections on $\\mathcal E$ corresponding to\nthis degeneration, we prove that, if $E$ is a given fiber away from a finite\nset, the restricted sequence $\\Xi_i|_{E}$ converges to a flat connection\nuniquely determined by the holomorphic structure on $\\mathcal E$.\n",
"title": "Hermitian-Yang-Mills connections on collapsing elliptically fibered $K3$ surfaces"
} | null | null | [
"Mathematics"
]
| null | true | null | 2301 | null | Validated | null | null |
null | {
"abstract": " In this paper we propose the use of quantum genetic algorithm to optimize the\nsupport vector machine (SVM) for human action recognition. The Microsoft Kinect\nsensor can be used for skeleton tracking, which provides the joints' position\ndata. However, how to extract the motion features for representing the dynamics\nof a human skeleton is still a challenge due to the complexity of human motion.\nWe present a highly efficient features extraction method for action\nclassification, that is, using the joint angles to represent a human skeleton\nand calculating the variance of each angle during an action time window. Using\nthe proposed representation, we compared the human action classification\naccuracy of two approaches, including the optimized SVM based on quantum\ngenetic algorithm and the conventional SVM with grid search. Experimental\nresults on the MSR-12 dataset show that the conventional SVM achieved an\naccuracy of $ 93.85\\% $. The proposed approach outperforms the conventional\nmethod with an accuracy of $ 96.15\\% $.\n",
"title": "Highly Efficient Human Action Recognition with Quantum Genetic Algorithm Optimized Support Vector Machine"
} | null | null | null | null | true | null | 2302 | null | Default | null | null |
null | {
"abstract": " We present a vision-only model for gaming AI which uses a late integration\ndeep convolutional network architecture trained in a purely supervised\nimitation learning context. Although state-of-the-art deep learning models for\nvideo game tasks generally rely on more complex methods such as deep-Q\nlearning, we show that a supervised model which requires substantially fewer\nresources and training time can already perform well at human reaction speeds\non the N64 classic game Super Smash Bros. We frame our learning task as a\n30-class classification problem, and our CNN model achieves 80% top-1 and 95%\ntop-3 validation accuracy. With slight test-time fine-tuning, our model is also\ncompetitive during live simulation with the highest-level AI built into the\ngame. We will further show evidence through network visualizations that the\nnetwork is successfully leveraging temporal information during inference to aid\nin decision making. Our work demonstrates that supervised CNN models can\nprovide good performance in challenging policy prediction tasks while being\nsignificantly simpler and more lightweight than alternatives.\n",
"title": "The Game Imitation: Deep Supervised Convolutional Networks for Quick Video Game AI"
} | null | null | null | null | true | null | 2303 | null | Default | null | null |
null | {
"abstract": " Several kinds of differential relations for polynomial components of almost\nBelyi maps are presented. Saito's theory of free divisors give particularly\ninteresting (yet conjectural) logarithmic action of vector fields. The\ndifferential relations implied by Kitaev's construction of algebraic Painleve\nVI solutions through pull-back transformations are used to compute almost Belyi\nmaps for the pull-backs giving all genus 0 and 1 Painleve VI solutions in the\nLisovyy-Tykhyy classification.\n",
"title": "Differential relations for almost Belyi maps"
} | null | null | null | null | true | null | 2304 | null | Default | null | null |
null | {
"abstract": " This paper studies holomorphic semicocycles over semigroups in the unit disk,\nwhich take values in an arbitrary unital Banach algebra. We prove that every\nsuch semicocycle is a solution to a corresponding evolution problem. We then\ninvestigate the linearization problem: which semicocycles are cohomologous to\nconstant semicocycles? In contrast with the case of commutative semicocycles,\nin the non-commutative case non-linearizable semicocycles are shown to exist.\nSimple conditions for linearizability are derived and are shown to be sharp.\n",
"title": "Non-commutative holomorphic semicocycles"
} | null | null | [
"Mathematics"
]
| null | true | null | 2305 | null | Validated | null | null |
null | {
"abstract": " In this work, we plan to develop a system to compare virtual machines with\ncontainer technology. We would devise ways to measure the administrator effort\nof containers vs. Virtual Machines (VMs). Metrics that will be tested against\ninclude human efforts required, ease of migration, resource utilization and\nease of use using containers and virtual machines.\n",
"title": "Comparative Study of Virtual Machines and Containers for DevOps Developers"
} | null | null | null | null | true | null | 2306 | null | Default | null | null |
null | {
"abstract": " The emergence of intellectual property as an academic issue opens a big gate\nto a cross-disciplinary field. Different disciplines start a dialogue in the\nframework of the international multilateral treaties in the early 90's. As\nglobal economy demands new knowledge on intellectual property, Science grows at\nits own pace. However, the degree of consolidation of cross-disciplinary\nacademic communities is not clear. In order to know how closely related are\nthese communities, this paper proposes a mixed methodology to find invisible\ncolleges in the production about intellectual property. The articles examined\nin this paper were extracted from Web of Science. The analyzed period was from\n1994 to 2016, taking into account the signature of the agreement on\nTrade-Related Aspects of Intellectual Property Rights in the early 90's. A\ntotal amount of 1580 papers were processed through co-citation network\nanalysis. An especial technique, which combine algorithms of community\ndetection and defining population of articles through thresholds of shared\nreferences, was applied. In order to contrast the invisible colleges that\nemerged with the existence of formal institutional relations, it was made a\nqualitative tracking of the authors with respect to their institutional\naffiliation, lines of research and meeting places. Both methods show that the\nsubjects of interest can be grouped into 13 different issues related to\nintellectual property field. Even though most of them are related to Laws and\nEconomics, there are weak linkages between disciplines which could indicate the\nconstruction of a cross-disciplinary field.\n",
"title": "Developmental tendencies in the Academic Field of Intellectual Property through the Identification of Invisible Colleges"
} | null | null | null | null | true | null | 2307 | null | Default | null | null |
null | {
"abstract": " In this paper, we consider the cubic fourth-order nonlinear Schrödinger\nequation (4NLS) under the periodic boundary condition. We prove two results.\nOne is the local well-posedness in $H^s$ with $-1/3 \\le s < 0$ for the Cauchy\nproblem of the Wick ordered 4NLS. The other one is the non-squeezing property\nfor the flow map of 4NLS in the symplectic phase space $L^2(\\mathbb{T})$. To\nprove the former we used the ideas introduced in [Takaoka and Tsutsumi 2004]\nand [Nakanish et al 2010], and to prove the latter we used the ideas in\n[Colliander et al 2005].\n",
"title": "Periodic fourth-order cubic NLS: Local well-posedness and Non-squeezing property"
} | null | null | [
"Mathematics"
]
| null | true | null | 2308 | null | Validated | null | null |
null | {
"abstract": " In this thesis, we study the problem of feature learning on heterogeneous\nknowledge graphs. These features can be used to perform tasks such as link\nprediction, classification and clustering on graphs. Knowledge graphs provide\nrich semantics encoded in the edge and node types. Meta-paths consist of these\ntypes and abstract paths in the graph. Until now, meta-paths can only be used\nas categorical features with high redundancy and are therefore unsuitable for\nmachine learning models. We propose meta-path embeddings to solve this problem\nby learning semantical and compact vector representations of them. Current\ngraph embedding methods only embed nodes and edge types and therefore miss\nsemantics encoded in the combination of them. Our method embeds meta-paths\nusing the skipgram model with an extension to deal with the redundancy and high\namount of meta-paths in big knowledge graphs. We critically evaluate our\nembedding approach by predicting links on Wikidata. The experiments indicate\nthat we learn a sensible embedding of the meta-paths but can improve it\nfurther.\n",
"title": "Feature Learning for Meta-Paths in Knowledge Graphs"
} | null | null | null | null | true | null | 2309 | null | Default | null | null |
null | {
"abstract": " We introduce new boundary integral operators which are the exact inverses of\nthe weakly singular and hypersingular operators for the Laplacian on flat\ndisks. Moreover, we provide explicit closed forms for them and prove the\ncontinuity and ellipticity of their corresponding bilinear forms in the natural\nSobolev trace spaces. This permit us to derive new Calderón-type identities\nthat can provide the foundation for optimal operator preconditioning in\nGalerkin boundary element methods.\n",
"title": "Closed-Form Exact Inverses of the Weakly Singular and Hypersingular Operators On Disks"
} | null | null | null | null | true | null | 2310 | null | Default | null | null |
null | {
"abstract": " Granular materials are complex multi-particle ensembles in which macroscopic\nproperties are largely determined by inter-particle interactions between their\nnumerous constituents. In order to understand and to predict their macroscopic\nphysical behavior, it is necessary to analyze the composition and interactions\nat the level of individual contacts and grains. To do so requires the ability\nto image individual particles and their local configurations to high precision.\nA variety of competing and complementary imaging techniques have been developed\nfor that task. In this introductory paper accompanying the Focus Issue, we\nprovide an overview of these imaging methods and discuss their advantages and\ndrawbacks, as well as their limits of application.\n",
"title": "Focus on Imaging Methods in Granular Physics"
} | null | null | null | null | true | null | 2311 | null | Default | null | null |
null | {
"abstract": " In the stochastic matching problem, we are given a general (not necessarily\nbipartite) graph $G(V,E)$, where each edge in $E$ is realized with some\nconstant probability $p > 0$ and the goal is to compute a bounded-degree\n(bounded by a function depending only on $p$) subgraph $H$ of $G$ such that the\nexpected maximum matching size in $H$ is close to the expected maximum matching\nsize in $G$. The algorithms in this setting are considered non-adaptive as they\nhave to choose the subgraph $H$ without knowing any information about the set\nof realized edges in $G$. Originally motivated by an application to kidney\nexchange, the stochastic matching problem and its variants have received\nsignificant attention in recent years.\nThe state-of-the-art non-adaptive algorithms for stochastic matching achieve\nan approximation ratio of $\\frac{1}{2}-\\epsilon$ for any $\\epsilon > 0$,\nnaturally raising the question that if $1/2$ is the limit of what can be\nachieved with a non-adaptive algorithm. In this work, we resolve this question\nby presenting the first algorithm for stochastic matching with an approximation\nguarantee that is strictly better than $1/2$: the algorithm computes a subgraph\n$H$ of $G$ with the maximum degree $O(\\frac{\\log{(1/ p)}}{p})$ such that the\nratio of expected size of a maximum matching in realizations of $H$ and $G$ is\nat least $1/2+\\delta_0$ for some absolute constant $\\delta_0 > 0$. The degree\nbound on $H$ achieved by our algorithm is essentially the best possible (up to\nan $O(\\log{(1/p)})$ factor) for any constant factor approximation algorithm,\nsince an $\\Omega(\\frac{1}{p})$ degree in $H$ is necessary for a vertex to\nacquire at least one incident edge in a realization.\n",
"title": "The Stochastic Matching Problem: Beating Half with a Non-Adaptive Algorithm"
} | null | null | [
"Computer Science"
]
| null | true | null | 2312 | null | Validated | null | null |
null | {
"abstract": " The prevention of domestic violence (DV) have aroused serious concerns in\nTaiwan because of the disparity between the increasing amount of reported DV\ncases that doubled over the past decade and the scarcity of social workers.\nAdditionally, a large amount of data was collected when social workers use the\npredominant case management approach to document case reports information.\nHowever, these data were not properly stored or organized.\nTo improve the efficiency of DV prevention and risk management, we worked\nwith Taipei City Government and utilized the 2015 data from its DV database to\nperform a spatial pattern analysis of the reports of DV cases to build a DV\nrisk map. However, during our map building process, the issue of confounding\nbias arose because we were not able to verify if reported cases truly reflected\nreal violence occurrence or were simply false reports from potential victim's\nneighbors. Therefore, we used the random forest method to build a repeat\nvictimization risk prediction model. The accuracy and F1-measure of our model\nwere 96.3% and 62.8%. This model helped social workers differentiate the risk\nlevel of new cases, which further reduced their major workload significantly.\nTo our knowledge, this is the first project that utilized machine learning in\nDV prevention. The research approach and results of this project not only can\nimprove DV prevention process, but also be applied to other social work or\ncriminal prevention areas.\n",
"title": "Measuring the unmeasurable - a project of domestic violence risk prediction and management"
} | null | null | null | null | true | null | 2313 | null | Default | null | null |
null | {
"abstract": " In this paper we show how to attain the capacity of discrete symmetric\nchannels with polynomial time decoding complexity by considering iterated\n$(U|U+V)$ constructions with Reed-Solomon code or algebraic geometry code\ncomponents. These codes are decoded with a recursive computation of the {\\em a\nposteriori} probabilities of the code symbols together with the Koetter-Vardy\nsoft decoder used for decoding the code components in polynomial time. We show\nthat when the number of levels of the iterated $(U|U+V)$ construction tends to\ninfinity, we attain the capacity of any discrete symmetric channel in this way.\nThis result follows from the polarization theorem together with a simple lemma\nexplaining how the Koetter-Vardy decoder behaves for Reed-Solomon codes of rate\nclose to $1$. However, even if this way of attaining the capacity of a\nsymmetric channel is essentially the Ar{\\i}kan polarization theorem, there are\nsome differences with standard polar codes.\nIndeed, with this strategy we can operate succesfully close to channel\ncapacity even with a small number of levels of the iterated $(U|U+V)$\nconstruction and the probability of error decays quasi-exponentially with the\ncodelength in such a case (i.e. exponentially if we forget about the\nlogarithmic terms in the exponent). We can even improve on this result by\nconsidering the algebraic geometry codes constructed in \\cite{TVZ82}. In such a\ncase, the probability of error decays exponentially in the codelength for any\nrate below the capacity of the channel. Moreover, when comparing this strategy\nto Reed-Solomon codes (or more generally algebraic geometry codes) decoded with\nthe Koetter-Vardy decoding algorithm, it does not only improve the noise level\nthat the code can tolerate, it also results in a significant complexity gain.\n",
"title": "Attaining Capacity with Algebraic Geometry Codes through the $(U|U+V)$ Construction and Koetter-Vardy Soft Decoding"
} | null | null | null | null | true | null | 2314 | null | Default | null | null |
null | {
"abstract": " Avionics is one kind of domain where prevention prevails. Nonetheless fails\noccur. Sometimes due to pilot misreacting, flooded in information. Sometimes\ninformation itself would be better verified than trusted. To avoid some kind of\nfailure, it has been thought to add,in midst of the ARINC664 aircraft data\nnetwork, a new kind of monitoring.\n",
"title": "Embedded real-time monitoring using SystemC in IMA network"
} | null | null | [
"Computer Science"
]
| null | true | null | 2315 | null | Validated | null | null |
null | {
"abstract": " Recent research has revealed that the output of Deep Neural Networks (DNN)\ncan be easily altered by adding relatively small perturbations to the input\nvector. In this paper, we analyze an attack in an extremely limited scenario\nwhere only one pixel can be modified. For that we propose a novel method for\ngenerating one-pixel adversarial perturbations based on differential\nevolution(DE). It requires less adversarial information(a black-box attack) and\ncan fool more types of networks due to the inherent features of DE. The results\nshow that 68.36% of the natural images in CIFAR-10 test dataset and 41.22% of\nthe ImageNet (ILSVRC 2012) validation images can be perturbed to at least one\ntarget class by modifying just one pixel with 73.22% and 5.52% confidence on\naverage. Thus, the proposed attack explores a different take on adversarial\nmachine learning in an extreme limited scenario, showing that current DNNs are\nalso vulnerable to such low dimension attacks. Besides, we also illustrate an\nimportant application of DE (or broadly speaking, evolutionary computation) in\nthe domain of adversarial machine learning: creating tools that can effectively\ngenerate low-cost adversarial attacks against neural networks for evaluating\nrobustness. The code is available on:\nthis https URL\n",
"title": "One pixel attack for fooling deep neural networks"
} | null | null | null | null | true | null | 2316 | null | Default | null | null |
null | {
"abstract": " We investigate of a special dam optimal location at the Volga river in area\nof the Akhtuba left sleeve beginning (7 \\, km to the south of the Volga\nHydroelectric Power Station dam). We claim that a new water-retaining dam can\nresolve the key problem of the Volga-Akhtuba floodplain related to insufficient\nwater amount during the spring flooding due to the overregulation of the Lower\nVolga. By using a numerical integration of Saint-Vacant equations we study the\nwater dynamics across the northern part of the Volga-Akhtuba floodplain with\ntaking into account its actual topography. As the result we found an amount of\nwater $V_A$ passing to the Akhtuba during spring period for a given water flow\nthrough the Volga Hydroelectric Power Station (so-called hydrograph which\ncharacterises the water flow per unit of time). By varying the location of the\nwater-retaining dam $ x_d, y_d $ we obtained various values of $V_A (x_d, y_d)\n$ as well as various flow spatial structure on the territory during the flood\nperiod. Gradient descent method provide us the dam coordinated with the maximum\nvalue of ${V_A}$. Such approach to the dam location choice let us to find the\nbest solution, that the value $V_A$ increases by a factor of 2. Our analysis\ndemonstrate a good potential of the numerical simulations in the field of\nhydraulic works.\n",
"title": "A computer simulation of the Volga River hydrological regime: a problem of water-retaining dam optimal location"
} | null | null | null | null | true | null | 2317 | null | Default | null | null |
null | {
"abstract": " Proton-driven plasma wakefield acceleration has been demonstrated in\nsimulations to be capable of accelerating particles to the energy frontier in a\nsingle stage, but its potential is hindered by the fact that currently\navailable proton bunches are orders of magnitude longer than the plasma\nwavelength. Fortunately, proton micro-bunching allows driving plasma waves\nresonantly. In this paper, we propose using a hollow plasma channel for\nmultiple proton bunch driven plasma wakefield acceleration and demonstrate that\nit enables the operation in the nonlinear regime and resonant excitation of\nstrong plasma waves. This new regime also involves beneficial features of\nhollow channels for the accelerated beam (such as emittance preservation and\nuniform accelerating field) and long buckets of stable deceleration for the\ndrive beam. The regime is attained at a proper ratio among plasma skin depth,\ndriver radius, hollow channel radius, and micro-bunch period.\n",
"title": "Multi-proton bunch driven hollow plasma wakefield acceleration in the nonlinear regime"
} | null | null | null | null | true | null | 2318 | null | Default | null | null |
null | {
"abstract": " Convolutional Neural Networks have been a subject of great importance over\nthe past decade and great strides have been made in their utility for producing\nstate of the art performance in many computer vision problems. However, the\nbehavior of deep networks is yet to be fully understood and is still an active\narea of research. In this work, we present an intriguing behavior: pre-trained\nCNNs can be made to improve their predictions by structurally perturbing the\ninput. We observe that these perturbations - referred as Guided Perturbations -\nenable a trained network to improve its prediction performance without any\nlearning or change in network weights. We perform various ablative experiments\nto understand how these perturbations affect the local context and feature\nrepresentations. Furthermore, we demonstrate that this idea can improve\nperformance of several existing approaches on semantic segmentation and scene\nlabeling tasks on the PASCAL VOC dataset and supervised classification tasks on\nMNIST and CIFAR10 datasets.\n",
"title": "Self corrective Perturbations for Semantic Segmentation and Classification"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2319 | null | Validated | null | null |
null | {
"abstract": " This paper is the first work to perform spatio-temporal mapping of human\nactivity using the visual content of geo-tagged videos. We utilize a recent\ndeep-learning based video analysis framework, termed hidden two-stream\nnetworks, to recognize a range of activities in YouTube videos. This framework\nis efficient and can run in real time or faster which is important for\nrecognizing events as they occur in streaming video or for reducing latency in\nanalyzing already captured video. This is, in turn, important for using video\nin smart-city applications. We perform a series of experiments to show our\napproach is able to accurately map activities both spatially and temporally. We\nalso demonstrate the advantages of using the visual content over the\ntags/titles.\n",
"title": "Large-Scale Mapping of Human Activity using Geo-Tagged Videos"
} | null | null | null | null | true | null | 2320 | null | Default | null | null |
null | {
"abstract": " This paper studies the structure of a parabolic partial differential equation\non graphs and digital n-dimensional manifolds, which are digital models of\ncontinuous n-manifolds. Conditions for the existence of solutions of equations\nare determined and investigated. Numerical solutions of the equation on a Klein\nbottle, a projective plane, a 4D sphere and a Moebius strip are presented.\n",
"title": "Structure of a Parabolic Partial Differential Equation on Graphs and Digital spaces. Solution of PDE on Digital Spaces: a Klein Bottle, a Projective Plane, a 4D Sphere and a Moebius Band"
} | null | null | null | null | true | null | 2321 | null | Default | null | null |
null | {
"abstract": " In this paper, we generally formulate the dynamics prediction problem of\nvarious network systems (e.g., the prediction of mobility, traffic and\ntopology) as the temporal link prediction task. Different from conventional\ntechniques of temporal link prediction that ignore the potential non-linear\ncharacteristics and the informative link weights in the dynamic network, we\nintroduce a novel non-linear model GCN-GAN to tackle the challenging temporal\nlink prediction task of weighted dynamic networks. The proposed model leverages\nthe benefits of the graph convolutional network (GCN), long short-term memory\n(LSTM) as well as the generative adversarial network (GAN). Thus, the dynamics,\ntopology structure and evolutionary patterns of weighted dynamic networks can\nbe fully exploited to improve the temporal link prediction performance.\nConcretely, we first utilize GCN to explore the local topological\ncharacteristics of each single snapshot and then employ LSTM to characterize\nthe evolving features of the dynamic networks. Moreover, GAN is used to enhance\nthe ability of the model to generate the next weighted network snapshot, which\ncan effectively tackle the sparsity and the wide-value-range problem of edge\nweights in real-life dynamic networks. To verify the model's effectiveness, we\nconduct extensive experiments on four datasets of different network systems and\napplication scenarios. The experimental results demonstrate that our model\nachieves impressive results compared to the state-of-the-art competitors.\n",
"title": "GCN-GAN: A Non-linear Temporal Link Prediction Model for Weighted Dynamic Networks"
} | null | null | null | null | true | null | 2322 | null | Default | null | null |
null | {
"abstract": " Graphs are an important tool to model data in different domains, including\nsocial networks, bioinformatics and the world wide web. Most of the networks\nformed in these domains are directed graphs, where all the edges have a\ndirection and they are not symmetric. Betweenness centrality is an important\nindex widely used to analyze networks. In this paper, first given a directed\nnetwork $G$ and a vertex $r \\in V(G)$, we propose a new exact algorithm to\ncompute betweenness score of $r$. Our algorithm pre-computes a set\n$\\mathcal{RV}(r)$, which is used to prune a huge amount of computations that do\nnot contribute in the betweenness score of $r$. Time complexity of our exact\nalgorithm depends on $|\\mathcal{RV}(r)|$ and it is respectively\n$\\Theta(|\\mathcal{RV}(r)|\\cdot|E(G)|)$ and\n$\\Theta(|\\mathcal{RV}(r)|\\cdot|E(G)|+|\\mathcal{RV}(r)|\\cdot|V(G)|\\log |V(G)|)$\nfor unweighted graphs and weighted graphs with positive weights.\n$|\\mathcal{RV}(r)|$ is bounded from above by $|V(G)|-1$ and in most cases, it\nis a small constant. Then, for the cases where $\\mathcal{RV}(r)$ is large, we\npresent a simple randomized algorithm that samples from $\\mathcal{RV}(r)$ and\nperforms computations for only the sampled elements. We show that this\nalgorithm provides an $(\\epsilon,\\delta)$-approximation of the betweenness\nscore of $r$. Finally, we perform extensive experiments over several real-world\ndatasets from different domains for several randomly chosen vertices as well as\nfor the vertices with the highest betweenness scores. Our experiments reveal\nthat in most cases, our algorithm significantly outperforms the most efficient\nexisting randomized algorithms, in terms of both running time and accuracy. Our\nexperiments also show that our proposed algorithm computes betweenness scores\nof all vertices in the sets of sizes 5, 10 and 15, much faster and more\naccurate than the most efficient existing algorithms.\n",
"title": "Efficient Exact and Approximate Algorithms for Computing Betweenness Centrality in Directed Graphs"
} | null | null | [
"Computer Science"
]
| null | true | null | 2323 | null | Validated | null | null |
null | {
"abstract": " Context: In the past decade, sensitive, resolved Sunyaev-Zel'dovich (SZ)\nstudies of galaxy clusters have become common. Whereas many previous SZ studies\nhave parameterized the pressure profiles of galaxy clusters, non-parametric\nreconstructions will provide insights into the thermodynamic state of the\nintracluster medium (ICM). Aims: We seek to recover the non-parametric pressure\nprofiles of the high redshift ($z=0.89$) galaxy cluster CLJ 1226.9+3332 as\ninferred from SZ data from the MUSTANG, NIKA, Bolocam, and Planck instruments,\nwhich all probe different angular scales. Methods: Our non-parametric algorithm\nmakes use of logarithmic interpolation, which under the assumption of\nellipsoidal symmetry is analytically integrable. For MUSTANG, NIKA, and Bolocam\nwe derive a non-parametric pressure profile independently and find good\nagreement among the instruments. In particular, we find that the non-parametric\nprofiles are consistent with a fitted gNFW profile. Given the ability of Planck\nto constrain the total signal, we include a prior on the integrated Compton Y\nparameter as determined by Planck. Results: For a given instrument, constraints\non the pressure profile diminish rapidly beyond the field of view. The overlap\nin spatial scales probed by these four datasets is therefore critical in\nchecking for consistency between instruments. By using multiple instruments,\nour analysis of CLJ 1226.9+3332 covers a large radial range, from the central\nregions to the cluster outskirts: $0.05 R_{500} < r < 1.1 R_{500}$. This is a\nwider range of spatial scales than is typical recovered by SZ instruments.\nSimilar analyses will be possible with the new generation of SZ instruments\nsuch as NIKA2 and MUSTANG2.\n",
"title": "A multi-instrument non-parametric reconstruction of the electron pressure profile in the galaxy cluster CLJ1226.9+3332"
} | null | null | null | null | true | null | 2324 | null | Default | null | null |
null | {
"abstract": " Disjoint-Set forests, consisting of Union-Find trees, are data structures\nhaving a widespread practical application due to their efficiency. Despite them\nbeing well-known, no exact structural characterization of these trees is known\n(such a characterization exists for Union trees which are constructed without\nusing path compression) for the case assuming union-by-rank strategy for\nmerging. In this paper we provide such a characterization by means of a simple\npush operation and show that the decision problem whether a given tree (along\nwith the rank info of its nodes) is a Union-Find tree is NP-complete,\ncomplementing our earlier similar result for the union-by-size strategy.\n",
"title": "Recognizing Union-Find trees built up using union-by-rank strategy is NP-complete"
} | null | null | [
"Computer Science"
]
| null | true | null | 2325 | null | Validated | null | null |
null | {
"abstract": " In this paper, we study the problem of exploring a translating plume with a\nteam of aerial robots. The shape and the size of the plume are unknown to the\nrobots. The objective is to find a tour for each robot such that they\ncollectively explore the plume. Specifically, the tours must be such that each\npoint in the plume must be visible from the field-of-view of some robot along\nits tour. We propose a recursive Depth-First Search (DFS)-based algorithm that\nyields a constant competitive ratio for the exploration problem. The\ncompetitive ratio is\n$\\frac{2(S_r+S_p)(R+\\lfloor\\log{R}\\rfloor)}{(S_r-S_p)(1+\\lfloor\\log{R}\\rfloor)}$\nwhere $R$ is the number of robots, and $S_r$ and $S_p$ are the robot speed and\nthe plume speed, respectively. We also consider a more realistic scenario where\nthe plume shape is not restricted to grid cells but an arbitrary shape. We show\nour algorithm has\n$\\frac{2(S_r+S_p)(18R+\\lfloor\\log{R}\\rfloor)}{(S_r-S_p)(1+\\lfloor\\log{R}\\rfloor)}$\ncompetitive ratio under the fat condition. We empirically verify our algorithm\nusing simulations.\n",
"title": "A Competitive Algorithm for Online Multi-Robot Exploration of a Translating Plume"
} | null | null | [
"Computer Science"
]
| null | true | null | 2326 | null | Validated | null | null |
null | {
"abstract": " The present paper shows that warped Riemannian metrics, a class of Riemannian\nmetrics which play a prominent role in Riemannian geometry, are also of\nfundamental importance in information geometry. Precisely, the paper features a\nnew theorem, which states that the Rao-Fisher information metric of any\nlocation-scale model, defined on a Riemannian manifold, is a warped Riemannian\nmetric, whenever this model is invariant under the action of some Lie group.\nThis theorem is a valuable tool in finding the expression of the Rao-Fisher\ninformation metric of location-scale models defined on high-dimensional\nRiemannian manifolds. Indeed, a warped Riemannian metric is fully determined by\nonly two functions of a single variable, irrespective of the dimension of the\nunderlying Riemannian manifold. Starting from this theorem, several original\ncontributions are made. The expression of the Rao-Fisher information metric of\nthe Riemannian Gaussian model is provided, for the first time in the\nliterature. A generalised definition of the Mahalanobis distance is introduced,\nwhich is applicable to any location-scale model defined on a Riemannian\nmanifold. The solution of the geodesic equation is obtained, for any Rao-Fisher\ninformation metric defined in terms of warped Riemannian metrics. Finally,\nusing a mixture of analytical and numerical computations, it is shown that the\nparameter space of the von Mises-Fisher model of $n$-dimensional directional\ndata, when equipped with its Rao-Fisher information metric, becomes a Hadamard\nmanifold, a simply-connected complete Riemannian manifold of negative sectional\ncurvature, for $n = 2,\\ldots,8$. Hopefully, in upcoming work, this will be\nproved for any value of $n$.\n",
"title": "Warped Riemannian metrics for location-scale models"
} | null | null | null | null | true | null | 2327 | null | Default | null | null |
null | {
"abstract": " We look at stochastic optimization problems through the lens of statistical\ndecision theory. In particular, we address admissibility, in the statistical\ndecision theory sense, of the natural sample average estimator for a stochastic\noptimization problem (which is also known as the empirical risk minimization\n(ERM) rule in learning literature). It is well known that for general\nstochastic optimization problems, the sample average estimator may not be\nadmissible. This is known as Stein's paradox in the statistics literature. We\nshow in this paper that for optimizing stochastic linear functions over compact\nsets, the sample average estimator is admissible.\n",
"title": "Admissibility of solution estimators for stochastic optimization"
} | null | null | null | null | true | null | 2328 | null | Default | null | null |
null | {
"abstract": " We study the single machine scheduling problem with the objective to minimize\nthe total weight of late jobs. It is assumed that the processing times of jobs\nare not exactly known at the time when a complete schedule must be dispatched.\nInstead, only interval bounds for these parameters are given. In contrast to\nthe stochastic optimization approach, we consider the problem of finding a\nrobust schedule, which minimizes the maximum regret of a solution. Heuristic\nalgorithm based on mixed-integer linear programming is presented and examined\nthrough computational experiments.\n",
"title": "Min-Max Regret Scheduling To Minimize the Total Weight of Late Jobs With Interval Uncertainty"
} | null | null | null | null | true | null | 2329 | null | Default | null | null |
null | {
"abstract": " Rural building mapping is paramount to support demographic studies and plan\nactions in response to crisis that affect those areas. Rural building\nannotations exist in OpenStreetMap (OSM), but their quality and quantity are\nnot sufficient for training models that can create accurate rural building\nmaps. The problems with these annotations essentially fall into three\ncategories: (i) most commonly, many annotations are geometrically misaligned\nwith the updated imagery; (ii) some annotations do not correspond to buildings\nin the images (they are misannotations or the buildings have been destroyed);\nand (iii) some annotations are missing for buildings in the images (the\nbuildings were never annotated or were built between subsequent image\nacquisitions). First, we propose a method based on Markov Random Field (MRF) to\nalign the buildings with their annotations. The method maximizes the\ncorrelation between annotations and a building probability map while enforcing\nthat nearby buildings have similar alignment vectors. Second, the annotations\nwith no evidence in the building probability map are removed. Third, we present\na method to detect non-annotated buildings with predefined shapes and add their\nannotation. The proposed methodology shows considerable improvement in accuracy\nof the OSM annotations for two regions of Tanzania and Zimbabwe, being more\naccurate than state-of-the-art baselines.\n",
"title": "Correcting rural building annotations in OpenStreetMap using convolutional neural networks"
} | null | null | null | null | true | null | 2330 | null | Default | null | null |
null | {
"abstract": " The problem of suppressing the scattering from conductive objects is\naddressed in terms of harmonic contrast reduction. A unique compact closed-form\nsolution for a surface impedance $Z_s(m,kr)$ is found in a straightforward\nmanner and without any approximation as a function of the harmonic index $m$\n(scattering mode to suppress) and of the frequency regime $kr$ (product of\nwavenumber $k$ and radius $r$ of the cloaked system) at any frequency regime.\nIn the quasi-static limit, mantle cloaking is obtained as a particular case for\n$kr \\ll 1$ and $m=0$. In addition, beyond quasi-static regime, impedance\ncoatings for a selected dominant harmonic wave can be designed with proper\ndispersive behaviour, resulting in improved reduction levels and harmonic\nfiltering capability.\n",
"title": "Closed-form Harmonic Contrast Control with Surface Impedance Coatings for Conductive Objects"
} | null | null | null | null | true | null | 2331 | null | Default | null | null |
null | {
"abstract": " The accurate and robust simulation of transcritical real-fluid effects is\ncrucial for many engineering applications, such as fuel injection in internal\ncombustion engines, rocket engines and gas turbines. For example, in diesel\nengines, the liquid fuel is injected into the ambient gas at a pressure that\nexceeds its critical value, and the fuel jet will be heated to a supercritical\ntemperature before combustion takes place. This process is often referred to as\ntranscritical injection. The largest thermodynamic gradient in the\ntranscritical regime occurs as the fluid undergoes a liquid-like to a gas-like\ntransition when crossing the pseudo-boiling line (Yang 2000, Oschwald et al.\n2006, Banuti 2015). The complex processes during transcritical injection are\nstill not well understood. Therefore, to provide insights into high-pressure\ncombustion systems, accurate and robust numerical simulation tools are required\nfor the characterization of supercritical and transcritical flows.\n",
"title": "Numerical methods to prevent pressure oscillations in transcritical flows"
} | null | null | [
"Physics"
]
| null | true | null | 2332 | null | Validated | null | null |
null | {
"abstract": " Given a substitution tiling $T$ of the plane with subdivision operator\n$\\tau$, we study the conformal tilings $\\mathcal{T}_n$ associated with $\\tau^n\nT$. We prove that aggregate tiles within $\\mathcal{T}_n$ converge in shape as\n$n\\rightarrow \\infty$ to their associated Euclidean tiles in $T$.\n",
"title": "Shape Convergence for Aggregate Tiles in Conformal Tilings"
} | null | null | null | null | true | null | 2333 | null | Default | null | null |
null | {
"abstract": " The detection of molecular species in the atmospheres of earth-like\nexoplanets orbiting nearby stars requires an optical system that suppresses\nstarlight and maximizes the sensitivity to the weak planet signals at small\nangular separations. Achieving sufficient contrast performance on a segmented\naperture space telescope is particularly challenging due to unwanted\ndiffraction within the telescope from amplitude and phase discontinuities in\nthe pupil. Apodized vortex coronagraphs are a promising solution that\ntheoretically meet the performance needs for high contrast imaging with future\nsegmented space telescopes. We investigate the sensitivity of apodized vortex\ncoronagraphs to the expected aberrations, including segment co-phasing errors\nin piston and tip/tilt as well as other low-order and mid-spatial frequency\naberrations. Coronagraph designs and their associated telescope requirements\nare identified for conceptual HabEx and LUVOIR telescope designs.\n",
"title": "Performance and sensitivity of vortex coronagraphs on segmented space telescopes"
} | null | null | [
"Physics"
]
| null | true | null | 2334 | null | Validated | null | null |
null | {
"abstract": " I discuss several issues related to \"classical\" spacetime structure. I review\nGalilean, Newtonian, and Leibnizian spacetimes, and briefly describe more\nrecent developments. The target audience is undergraduates and early graduate\nstudents in philosophy; the presentation avoids mathematical formalism as much\nas possible.\n",
"title": "Classical Spacetime Structure"
} | null | null | null | null | true | null | 2335 | null | Default | null | null |
null | {
"abstract": " We study the convergence of the parameter family of series\n$$V_{\\alpha,\\beta}(t)=\\sum_{p}p^{-\\alpha}\\exp(2\\pi i p^{\\beta}t),\\quad\n\\alpha,\\beta \\in \\mathbb{R}_{>0},\\; t \\in [0,1)$$ defined over prime numbers\n$p$, and subsequently, their differentiability properties. The visible fractal\nnature of the graphs as a function of $\\alpha,\\beta$ is analyzed in terms of\nHölder continuity, self similarity and fractal dimension, backed with\nnumerical results. We also discuss the link of this series to random walks and\nconsequently, explore numerically its random properties.\n",
"title": "Fractal curves from prime trigonometric series"
} | null | null | null | null | true | null | 2336 | null | Default | null | null |
null | {
"abstract": " For stationary, homogeneous Markov processes (viz., Lévy processes,\nincluding Brownian motion) in dimension $d\\geq 3$, we establish an exact\nformula for the average number of $(d-1)$-dimensional facets that can be\ndefined by $d$ points on the process's path. This formula defines a\nuniversality class in that it is independent of the increments' distribution,\nand it admits a closed form when $d=3$, a case which is of particular interest\nfor applications in biophysics, chemistry and polymer science.\nWe also show that the asymptotical average number of facets behaves as\n$\\langle \\mathcal{F}_T^{(d)}\\rangle \\sim 2\\left[\\ln \\left( T/\\Delta\nt\\right)\\right]^{d-1}$, where $T$ is the total duration of the motion and\n$\\Delta t$ is the minimum time lapse separating points that define a facet.\n",
"title": "Facets on the convex hull of $d$-dimensional Brownian and Lévy motion"
} | null | null | null | null | true | null | 2337 | null | Default | null | null |
null | {
"abstract": " We consider smooth, complex quasi-projective varieties $U$ which admit a\ncompactification with a boundary which is an arrangement of smooth algebraic\nhypersurfaces. If the hypersurfaces intersect locally like hyperplanes, and the\nrelative interiors of the hypersurfaces are Stein manifolds, we prove that the\ncohomology of certain local systems on $U$ vanishes. As an application, we show\nthat complements of linear, toric, and elliptic arrangements are both duality\nand abelian duality spaces.\n",
"title": "Local systems on complements of arrangements of smooth, complex algebraic hypersurfaces"
} | null | null | null | null | true | null | 2338 | null | Default | null | null |
null | {
"abstract": " We describe a list of open problems in random matrix theory and the theory of\nintegrable systems that was presented at the conference Asymptotics in\nIntegrable Systems, Random Matrices and Random Processes and Universality,\nCentre de Recherches Mathematiques, Montreal, June 7-11, 2015. We also describe\nprogress that has been made on problems in an earlier list presented by the\nauthor on the occasion of his 60th birthday in 2005 (see [Deift P., Contemp.\nMath., Vol. 458, Amer. Math. Soc., Providence, RI, 2008, 419-430,\narXiv:0712.0849]).\n",
"title": "Some Open Problems in Random Matrix Theory and the Theory of Integrable Systems. II"
} | null | null | null | null | true | null | 2339 | null | Default | null | null |
null | {
"abstract": " We proposed a semi-parametric estimation procedure in order to estimate the\nparameters of a max-mixture model and also of a max-stable model (inverse\nmax-stable model) as an alternative to composite likelihood. A good estimation\nby the proposed estimator required the dependence measure to detect all\ndependence structures in the model, especially when dealing with the\nmax-mixture model. We overcame this challenge by using the F-madogram. The\nsemi-parametric estimation was then based on a quasi least square method, by\nminimizing the square difference between the theoretical F-madogram and an\nempirical one. We evaluated the performance of this estimator through a\nsimulation study. It was shown that on an average, the estimation is performed\nwell, although in some cases, it encountered some difficulties. We apply our\nestimation procedure to model the daily rainfalls over the East Australia.\n",
"title": "A semi-parametric estimation for max-mixture spatial processes"
} | null | null | null | null | true | null | 2340 | null | Default | null | null |
null | {
"abstract": " The spectra of 413 star-forming (or HII) regions in M33 (NGC 598) were\nobserved by using the multifiber spectrograph of Hectospec at the 6.5-m\nMultiple Mirror Telescope (MMT). By using this homogeneous spectra sample, we\nmeasured the intensities of emission lines and some physical parameters, such\nas electron temperatures, electron densities, and metallicities. Oxygen\nabundances were derived via the direct method (when available) and two\nempirical strong-line methods, namely, O3N2 and N2. In the high-metallicity\nend, oxygen abundances derived from O3N2 calibration were higher than those\nderived from N2 index, indicating an inconsistency between O3N2 and N2\ncalibrations. We presented a detailed analysis of the spatial distribution of\ngas-phase oxygen abundances in M33 and confirmed the existence of the\naxisymmetric global metallicity distribution widely assumed in literature.\nLocal variations were also observed and subsequently associated with spiral\nstructures to provide evidence of radial migration driven by arms. Our O/H\ngradient fitted out to 1.1 $R_{25}$ resulted in slopes of $-0.17\\pm0.03$,\n$-0.19\\pm0.01$, and $-0.16\\pm0.17$ dex $R_{25}^{-1}$ utilizing abundances from\nO3N2, N2 diagnostics, and direct method, respectively.\n",
"title": "Spectroscopic Observation and Analysis of HII regions in M33 with MMT: Temperatures and Oxygen Abundances"
} | null | null | null | null | true | null | 2341 | null | Default | null | null |
null | {
"abstract": " The problem of output-only parameter identification for nonlinear oscillators\nforced by colored noise is considered. In this context, it is often assumed\nthat the forcing noise is white, since its actual spectral content is unknown.\nThe impact of this white noise forcing assumption upon parameter identification\nis quantitatively analyzed. First, a Van der Pol oscillator forced by an\nOrnstein-Uhlenbeck process is considered. Second, the practical case of\nthermoacoustic limit cycles in combustion chambers with turbulence-induced\nforcing is investigated. It is shown that in both cases, the system parameters\nare accurately identified if time signals are appropriately band-pass filtered\naround the oscillator eigenfrequency.\n",
"title": "Output-only parameter identification of a colored-noise-driven Van der Pol oscillator -- Thermoacoustic instabilities as an example"
} | null | null | null | null | true | null | 2342 | null | Default | null | null |
null | {
"abstract": " Over half a million individuals are diagnosed with head and neck cancer each\nyear worldwide. Radiotherapy is an important curative treatment for this\ndisease, but it requires manually intensive delineation of radiosensitive\norgans at risk (OARs). This planning process can delay treatment commencement.\nWhile auto-segmentation algorithms offer a potentially time-saving solution,\nthe challenges in defining, quantifying and achieving expert performance\nremain. Adopting a deep learning approach, we demonstrate a 3D U-Net\narchitecture that achieves performance similar to experts in delineating a wide\nrange of head and neck OARs. The model was trained on a dataset of 663\ndeidentified computed tomography (CT) scans acquired in routine clinical\npractice and segmented according to consensus OAR definitions. We demonstrate\nits generalisability through application to an independent test set of 24 CT\nscans available from The Cancer Imaging Archive collected at multiple\ninternational sites previously unseen to the model, each segmented by two\nindependent experts and consisting of 21 OARs commonly segmented in clinical\npractice. With appropriate validation studies and regulatory approvals, this\nsystem could improve the effectiveness of radiotherapy pathways.\n",
"title": "Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapy"
} | null | null | null | null | true | null | 2343 | null | Default | null | null |
null | {
"abstract": " We consider the problem of detecting a few targets among a large number of\nhierarchical data streams. The data streams are modeled as random processes\nwith unknown and potentially heavy-tailed distributions. The objective is an\nactive inference strategy that determines, sequentially, which data stream to\ncollect samples from in order to minimize the sample complexity under a\nreliability constraint. We propose an active inference strategy that induces a\nbiased random walk on the tree-structured hierarchy based on confidence bounds\nof sample statistics. We then establish its order optimality in terms of both\nthe size of the search space (i.e., the number of data streams) and the\nreliability requirement. The results find applications in hierarchical heavy\nhitter detection, noisy group testing, and adaptive sampling for active\nlearning, classification, and stochastic root finding.\n",
"title": "Anomaly Detection in Hierarchical Data Streams under Unknown Models"
} | null | null | null | null | true | null | 2344 | null | Default | null | null |
null | {
"abstract": " A number of image-processing problems can be formulated as optimization\nproblems. The objective function typically contains several terms specifically\ndesigned for different purposes. Parameters in front of these terms are used to\ncontrol the relative weights among them. It is of critical importance to tune\nthese parameters, as quality of the solution depends on their values. Tuning\nparameter is a relatively straightforward task for a human, as one can\nintelligently determine the direction of parameter adjustment based on the\nsolution quality. Yet manual parameter tuning is not only tedious in many\ncases, but becomes impractical when a number of parameters exist in a problem.\nAiming at solving this problem, this paper proposes an approach that employs\ndeep reinforcement learning to train a system that can automatically adjust\nparameters in a human-like manner. We demonstrate our idea in an example\nproblem of optimization-based iterative CT reconstruction with a pixel-wise\ntotal-variation regularization term. We set up a parameter tuning policy\nnetwork (PTPN), which maps an CT image patch to an output that specifies the\ndirection and amplitude by which the parameter at the patch center is adjusted.\nWe train the PTPN via an end-to-end reinforcement learning procedure. We\ndemonstrate that under the guidance of the trained PTPN for parameter tuning at\neach pixel, reconstructed CT images attain quality similar or better than in\nthose reconstructed with manually tuned parameters.\n",
"title": "Intelligent Parameter Tuning in Optimization-based Iterative CT Reconstruction via Deep Reinforcement Learning"
} | null | null | null | null | true | null | 2345 | null | Default | null | null |
null | {
"abstract": " An adversarial example is an example that has been adjusted to produce a\nwrong label when presented to a system at test time. To date, adversarial\nexample constructions have been demonstrated for classifiers, but not for\ndetectors. If adversarial examples that could fool a detector exist, they could\nbe used to (for example) maliciously create security hazards on roads populated\nwith smart vehicles. In this paper, we demonstrate a construction that\nsuccessfully fools two standard detectors, Faster RCNN and YOLO. The existence\nof such examples is surprising, as attacking a classifier is very different\nfrom attacking a detector, and that the structure of detectors - which must\nsearch for their own bounding box, and which cannot estimate that box very\naccurately - makes it quite likely that adversarial patterns are strongly\ndisrupted. We show that our construction produces adversarial examples that\ngeneralize well across sequences digitally, even though large perturbations are\nneeded. We also show that our construction yields physical objects that are\nadversarial.\n",
"title": "Adversarial Examples that Fool Detectors"
} | null | null | [
"Computer Science"
]
| null | true | null | 2346 | null | Validated | null | null |
null | {
"abstract": " We propose a homotopy continuation method called FLUX for approximating\ncomplicated probability density functions. It is based on progressive\nprocessing for smoothly morphing a given density into the desired one.\nDistributed ordinary differential equations (DODEs) with an artificial time\n$\\gamma \\in [0,1]$ are derived for describing the evolution from the initial\ndensity to the desired final density. For a finite-dimensional parametrization,\nthe DODEs are converted to a system of ordinary differential equations (SODEs),\nwhich are solved for $\\gamma \\in [0,1]$ and return the desired result for\n$\\gamma=1$. This includes parametric representations such as Gaussians or\nGaussian mixtures and nonparametric setups such as sample sets. In the latter\ncase, we obtain a particle flow between the two densities along the artificial\ntime.\nFLUX is applied to state estimation in stochastic nonlinear dynamic systems\nby gradual inclusion of measurement information. The proposed approximation\nmethod (1) is fast, (2) can be applied to arbitrary nonlinear systems and is\nnot limited to additive noise, (3) allows for target densities that are only\nknown at certain points, (4) does not require optimization, (5) does not\nrequire the solution of partial differential equations, and (6) works with\nstandard procedures for solving SODEs. This manuscript is limited to the\none-dimensional case and a fixed number of parameters during the progression.\nFuture extensions will include consideration of higher dimensions and on the\nfly adaption of the number of parameters.\n",
"title": "FLUX: Progressive State Estimation Based on Zakai-type Distributed Ordinary Differential Equations"
} | null | null | null | null | true | null | 2347 | null | Default | null | null |
null | {
"abstract": " User participation is considered an effective way to conduct requirements\nengineering, but user-developer perception gaps in requirements understanding\noccur frequently. Since user participation in practice is not as active as we\nexpect and the requirements perception gap has been recognized as a risk that\nnegatively affects projects, exploring whether user-developer perception gaps\nin requirements understanding will hinder user participation is worthwhile.\nThis will help develop a greater comprehension of the intertwined relationship\nbetween user participation and perception gap, a topic that has not yet been\nextensively examined. This study investigates the direct and mediating\ninfluences of user-developer requirements perception gaps on user participation\nby integrating requirements uncertainty and top management support. Survey data\ncollected from 140 subjects were examined and analyzed using structural\nequation modeling. The results indicate that perception gaps have a direct\nnegative effect on user participation and negate completely the positive effect\nof top management support on user participation. Additionally, perception gaps\ndo not have a mediating effect between requirements uncertainty and user\nparticipation because requirements uncertainty does not significantly and\ndirectly affect user participation, but requirements uncertainty indirectly\ninfluences user participation due to its significant direct effect on\nperception gaps. The theoretical and practical implications are discussed, and\nlimitations and possible future research areas are identified.\n",
"title": "Direct and mediating influences of user-developer perception gaps in requirements understanding on user participation"
} | null | null | null | null | true | null | 2348 | null | Default | null | null |
null | {
"abstract": " In this article, we consider the equivariant Schrödinger map from $\\Bbb\nH^2$ to $\\Bbb S^2$ which converges to the north pole of $\\Bbb S^2$ at the\norigin and spatial infinity of the hyperbolic space. If the energy of the data\nis less than $4\\pi$, we show that the local existence of Schrödinger map.\nFurthermore, if the energy of the data sufficiently small, we prove the\nsolutions are global in time.\n",
"title": "Equivariant Schrödinger maps from two dimensional hyperbolic space"
} | null | null | null | null | true | null | 2349 | null | Default | null | null |
null | {
"abstract": " A collection of arbitrarily-shaped solid objects, each moving at a constant\nspeed, can be used to mix or stir ideal fluid, and can give rise to interesting\nflow patterns. Assuming these systems of fluid stirrers are two-dimensional,\nthe mathematical problem of resolving the flow field - given a particular\ndistribution of any finite number of stirrers of specified shape and speed -\ncan be formulated as a Riemann-Hilbert problem. We show that this\nRiemann-Hilbert problem can be solved numerically using a fast and accurate\nalgorithm for any finite number of stirrers based around a boundary integral\nequation with the generalized Neumann kernel. Various systems of fluid stirrers\nare considered, and our numerical scheme is shown to handle highly multiply\nconnected domains (i.e. systems of many fluid stirrers) with minimal\ncomputational expense.\n",
"title": "A fast numerical method for ideal fluid flow in domains with multiple stirrers"
} | null | null | [
"Mathematics"
]
| null | true | null | 2350 | null | Validated | null | null |
null | {
"abstract": " A reinforcement learning agent tries to maximize its cumulative payoff by\ninteracting in an unknown environment. It is important for the agent to explore\nsuboptimal actions as well as to pick actions with highest known rewards. Yet,\nin sensitive domains, collecting more data with exploration is not always\npossible, but it is important to find a policy with a certain performance\nguaranty. In this paper, we present a brief survey of methods available in the\nliterature for balancing exploration-exploitation trade off and computing\nrobust solutions from fixed samples in reinforcement learning.\n",
"title": "A Short Survey on Probabilistic Reinforcement Learning"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2351 | null | Validated | null | null |
null | {
"abstract": " The structure, composition and electrophysical characteristics of\nlow-temperature silicon dioxide films under influence of various technology\nfactors, such as ion implantation, laser irradiation, thermal and photonic\nannealing, have been studied. Silicon dioxide films have been obtained by\nmonosilane oxidation using plasma chemical method, reactive cathode sputtering,\nand tetraethoxysilane pyrolysis. In the capacity of substrates, germanium,\nsilicon, gallium arsenide and gallium nitride were used. Structure and\ncomposition of the dielectric films were analyzed by methods of infrared\ntransmission spectroscopy and frustrated internal reflectance spectroscopy.\nAnalysis of modification efficiency of low-temperature silicon dioxide films\nhas been made depending on the substrate type, structure and properties of the\nfilms, their moisture permeability, dielectric deposition technique, type and\ndose of implantation ions, temperature and kind of annealing.\n",
"title": "Modification of low-temperature silicon dioxide films under the influence of technology factors"
} | null | null | null | null | true | null | 2352 | null | Default | null | null |
null | {
"abstract": " This work describes the development of a high-resolution tactile-sensing\nfinger for robot grasping. This finger, inspired by previous GelSight sensing\ntechniques, features an integration that is slimmer, more robust, and with more\nhomogeneous output than previous vision-based tactile sensors. To achieve a\ncompact integration, we redesign the optical path from illumination source to\ncamera by combining light guides and an arrangement of mirror reflections. We\nparameterize the optical path with geometric design variables and describe the\ntradeoffs between the finger thickness, the depth of field of the camera, and\nthe size of the tactile sensing area. The sensor sustains the wear from\ncontinuous use -- and abuse -- in grasping tasks by combining tougher materials\nfor the compliant soft gel, a textured fabric skin, a structurally rigid body,\nand a calibration process that maintains homogeneous illumination and contrast\nof the tactile images during use. Finally, we evaluate the sensor's durability\nalong four metrics that track the signal quality during more than 3000 grasping\nexperiments.\n",
"title": "GelSlim: A High-Resolution, Compact, Robust, and Calibrated Tactile-sensing Finger"
} | null | null | null | null | true | null | 2353 | null | Default | null | null |
null | {
"abstract": " The upcoming Fermilab E989 experiment will measure the muon anomalous\nmagnetic moment $a_{\\mu}$ . This measurement is motivated by the previous\nmeasurement performed in 2001 by the BNL E821 experiment that reported a 3-4\nstandard deviation discrepancy between the measured value and the Standard\nModel prediction. The new measurement at Fermilab aims to improve the precision\nby a factor of four reducing the total uncertainty from 540 parts per billion\n(BNL E821) to 140 parts per billion (Fermilab E989). This paper gives the\nstatus of the experiment.\n",
"title": "The Muon g-2 experiment at Fermilab"
} | null | null | null | null | true | null | 2354 | null | Default | null | null |
null | {
"abstract": " In their previous work, S. Koenig, S. Ovsienko and the second author showed\nthat every quasi-hereditary algebra is Morita equivalent to the right algebra,\ni.e. the opposite algebra of the left dual, of a coring. Let $A$ be an\nassociative algebra and $V$ an $A$-coring whose right algebra $R$ is\nquasi-hereditary. In this paper, we give a combinatorial description of an\nassociative algebra $B$ and a $B$-coring $W$ whose right algebra is the Ringel\ndual of $R$. We apply our results in small examples to obtain restrictions on\nthe $A_\\infty$-structure of the $\\textrm{Ext}$-algebra of standard modules over\na class of quasi-hereditary algebras related to birational morphisms of smooth\nsurfaces.\n",
"title": "Ringel duality as an instance of Koszul duality"
} | null | null | null | null | true | null | 2355 | null | Default | null | null |
null | {
"abstract": " Telecom companies are severely damaged by bypass fraud or SIM boxing.\nHowever, there is a shortage of published research to tackle this problem. The\ntraditional method of Test Call Generating is easily overcome by fraudsters and\nthe need for more sophisticated ways is inevitable. In this work, we are\ndeveloping intelligent algorithms that mine a huge amount of mobile operator's\ndata and detect the SIMs that are used to bypass international calls. This\nmethod will make it hard for fraudsters to generate revenue and hinder their\nwork. Also by reducing fraudulent activities, quality of service can be\nincreased as well as customer satisfaction. Our technique has been evaluated\nand tested on real world mobile operator data, and proved to be very efficient.\n",
"title": "Bypass Fraud Detection: Artificial Intelligence Approach"
} | null | null | null | null | true | null | 2356 | null | Default | null | null |
null | {
"abstract": " It is observed that many thin superconducting films with not too high\ndisorder level (generally R$_N/\\Box \\leq 2000 \\Omega$) placed in magnetic field\nshow an anomalous metallic phase where the resistance is low but still finite\nas temperature goes to zero. Here we report in weakly disordered amorphous\nInO$_x$ thin films, that this \"Bose metal\" metal phase possesses no cyclotron\nresonance and hence non-Drude electrodynamics. Its microwave dynamical\nconductivity shows signatures of remaining short-range superconducting\ncorrelations and strong phase fluctuations through the whole anomalous regime.\nThe absence of a finite frequency resonant mode can be associated with a\nvanishing downstream component of the vortex current parallel to the\nsupercurrent and an emergent particle-hole symmetry of this anomalous metal,\nwhich establishes its non-Fermi liquid character.\n",
"title": "Absence of cyclotron resonance in the anomalous metallic phase in InO$_x$"
} | null | null | null | null | true | null | 2357 | null | Default | null | null |
null | {
"abstract": " The goal of scenario reduction is to approximate a given discrete\ndistribution with another discrete distribution that has fewer atoms. We\ndistinguish continuous scenario reduction, where the new atoms may be chosen\nfreely, and discrete scenario reduction, where the new atoms must be chosen\nfrom among the existing ones. Using the Wasserstein distance as measure of\nproximity between distributions, we identify those $n$-point distributions on\nthe unit ball that are least susceptible to scenario reduction, i.e., that have\nmaximum Wasserstein distance to their closest $m$-point distributions for some\nprescribed $m<n$. We also provide sharp bounds on the added benefit of\ncontinuous over discrete scenario reduction. Finally, to our best knowledge, we\npropose the first polynomial-time constant-factor approximations for both\ndiscrete and continuous scenario reduction as well as the first exact\nexponential-time algorithms for continuous scenario reduction.\n",
"title": "Scenario Reduction Revisited: Fundamental Limits and Guarantees"
} | null | null | null | null | true | null | 2358 | null | Default | null | null |
null | {
"abstract": " The relationship of scientific knowledge development to technological\ndevelopment is widely recognized as one of the most important and complex\naspects of technological evolution. This paper adds to our understanding of the\nrelationship through use of a more rigorous structure for differentiating among\ntechnologies based upon technological domains (defined as consisting of the\nartifacts over time that fulfill a specific generic function using a specific\nbody of technical knowledge).\n",
"title": "Testing the science/technology relationship by analysis of patent citations of scientific papers after decomposition of both science and technology"
} | null | null | null | null | true | null | 2359 | null | Default | null | null |
null | {
"abstract": " The learning of mixture models can be viewed as a clustering problem. Indeed,\ngiven data samples independently generated from a mixture of distributions, we\noften would like to find the correct target clustering of the samples according\nto which component distribution they were generated from. For a clustering\nproblem, practitioners often choose to use the simple k-means algorithm.\nk-means attempts to find an optimal clustering which minimizes the\nsum-of-squared distance between each point and its cluster center. In this\npaper, we provide sufficient conditions for the closeness of any optimal\nclustering and the correct target clustering assuming that the data samples are\ngenerated from a mixture of log-concave distributions. Moreover, we show that\nunder similar or even weaker conditions on the mixture model, any optimal\nclustering for the samples with reduced dimensionality is also close to the\ncorrect target clustering. These results provide intuition for the\ninformativeness of k-means (with and without dimensionality reduction) as an\nalgorithm for learning mixture models. We verify the correctness of our\ntheorems using numerical experiments and demonstrate using datasets with\nreduced dimensionality significant speed ups for the time required to perform\nclustering.\n",
"title": "The Informativeness of $k$-Means and Dimensionality Reduction for Learning Mixture Models"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2360 | null | Validated | null | null |
null | {
"abstract": " Resolving abstract anaphora is an important, but difficult task for text\nunderstanding. Yet, with recent advances in representation learning this task\nbecomes a more tangible aim. A central property of abstract anaphora is that it\nestablishes a relation between the anaphor embedded in the anaphoric sentence\nand its (typically non-nominal) antecedent. We propose a mention-ranking model\nthat learns how abstract anaphors relate to their antecedents with an\nLSTM-Siamese Net. We overcome the lack of training data by generating\nartificial anaphoric sentence--antecedent pairs. Our model outperforms\nstate-of-the-art results on shell noun resolution. We also report first\nbenchmark results on an abstract anaphora subset of the ARRAU corpus. This\ncorpus presents a greater challenge due to a mixture of nominal and pronominal\nanaphors and a greater range of confounders. We found model variants that\noutperform the baselines for nominal anaphors, without training on individual\nanaphor data, but still lag behind for pronominal anaphors. Our model selects\nsyntactically plausible candidates and -- if disregarding syntax --\ndiscriminates candidates using deeper features.\n",
"title": "A Mention-Ranking Model for Abstract Anaphora Resolution"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2361 | null | Validated | null | null |
null | {
"abstract": " We present an effective harmonic density interpolation method for the\nnumerical evaluation of singular and nearly singular Laplace boundary integral\noperators and layer potentials in two and three spatial dimensions. The method\nrelies on the use of Green's third identity and local Taylor-like\ninterpolations of density functions in terms of harmonic polynomials. The\nproposed technique effectively regularizes the singularities present in\nboundary integral operators and layer potentials, and recasts the latter in\nterms of integrands that are bounded or even more regular, depending on the\norder of the density interpolation. The resulting boundary integrals can then\nbe easily, accurately, and inexpensively evaluated by means of standard\nquadrature rules. A variety of numerical examples demonstrate the effectiveness\nof the technique when used in conjunction with the classical trapezoidal rule\n(to integrate over smooth curves) in two-dimensions, and with a Chebyshev-type\nquadrature rule (to integrate over surfaces given as unions of non-overlapping\nquadrilateral patches) in three-dimensions.\n",
"title": "Harmonic density interpolation methods for high-order evaluation of Laplace layer potentials in 2D and 3D"
} | null | null | null | null | true | null | 2362 | null | Default | null | null |
null | {
"abstract": " Many problems in computer vision and recommender systems involve low-rank\nmatrices. In this work, we study the problem of finding the maximum entry of a\nstochastic low-rank matrix from sequential observations. At each step, a\nlearning agent chooses pairs of row and column arms, and receives the noisy\nproduct of their latent values as a reward. The main challenge is that the\nlatent values are unobserved. We identify a class of non-negative matrices\nwhose maximum entry can be found statistically efficiently and propose an\nalgorithm for finding them, which we call LowRankElim. We derive a\n$\\DeclareMathOperator{\\poly}{poly} O((K + L) \\poly(d) \\Delta^{-1} \\log n)$\nupper bound on its $n$-step regret, where $K$ is the number of rows, $L$ is the\nnumber of columns, $d$ is the rank of the matrix, and $\\Delta$ is the minimum\ngap. The bound depends on other problem-specific constants that clearly do not\ndepend $K L$. To the best of our knowledge, this is the first such result in\nthe literature.\n",
"title": "Stochastic Low-Rank Bandits"
} | null | null | null | null | true | null | 2363 | null | Default | null | null |
null | {
"abstract": " Classical novae show a rapid rise in optical brightness over a few hours.\nUntil recently the rise phase, particularly the phenomenon of a pre-maximum\nhalt, was observed sporadically. Solar observation satellites observing Coronal\nMass Ejections enable us to observe the pre-maximum phase in unprecedented\ntemporal resolution. We present observations of V5589 Sgr with STEREO HI-1B at\na cadence of 40 min, the highest to date. We temporally resolve a pre-maximum\nhalt for the first time, with two examples each rising over 40 min then\ndeclining within 80 min. Comparison with a grid of outburst models suggests\nthis double peak, and the overall rise timescale, are consistent with a white\ndwarf mass, central temperature and accretion rate close to 1.0 solar mass,\n5x10^7 K and 10^-10 solar masses per year respectively. The modelling formally\npredicts mass loss onset at JD 2456038.2391+/-0.0139, 12 hrs before optical\nmaximum. The model assumes a main-sequence donor. Observational evidence is for\na subgiant companion; meaning the accretion rate is under-estimated.\nPost-maximum we see erratic variations commonly associated with much slower\nnovae. Estimating the decline rate difficult, but we place the time to decline\ntwo magnitudes as 2.1 < t_2(days) < 3.9 making V5589 Sgr a \"very fast\" nova.\nThe brightest point defines \"day 0\" as JD 2456038.8224+/-0.0139, although at\nthis high cadence the meaning of the observed maximum becomes difficult to\ndefine. We suggest that such erratic variability normally goes undetected in\nfaster novae due to the low cadence of typical observations; implying erratic\nbehaviour is not necessarily related to the rate of decline.\n",
"title": "Temporal resolution of a pre-maximum halt in a Classical Nova: V5589 Sgr observed with STEREO HI-1B"
} | null | null | null | null | true | null | 2364 | null | Default | null | null |
null | {
"abstract": " In this work, we compare the thermophysical properties and particle sizes\nderived from the Mars Science Laboratory (MSL) rover's Ground Temperature\nSensor (GTS) of the Bagnold dunes, specifically Namib dune, to those derived\norbitally from Thermal Emission Imaging System (THEMIS), ultimately linking\nthese measurements to ground-truth particle sizes determined from Mars Hand\nLens Imager (MAHLI) images. In general, we find that all three datasets report\nconsistent particle sizes for the Bagnold dunes (~110-350 microns, and are\nwithin measurement and model uncertainties), indicating that particle sizes of\nhomogeneous materials determined from orbit are reliable. Furthermore, we\nexamine the effects of two physical characteristics that could influence the\nmodeled thermal inertia and particle sizes, including: 1) fine-scale (cm-m\nscale) ripples, and 2) thin layering of indurated/armored materials. To first\norder, we find small scale ripples and thin (approximately centimeter scale)\nlayers do not significantly affect the determination of bulk thermal inertia\nfrom orbital thermal data determined from a single nighttime temperature.\nModeling of a layer of coarse or indurated material reveals that a thin layer\n(< ~5 mm; similar to what was observed by the Curiosity rover) would not\nsignificantly change the observed thermal properties of the surface and would\nbe dominated by the properties of the underlying material. Thermal inertia and\ngrain sizes of relatively homogeneous materials derived from nighttime orbital\ndata should be considered as reliable, as long as there are not significant\nsub-pixel anisothermality effects (e.g. lateral mixing of multiple\nthermophysically distinct materials).\n",
"title": "The Thermophysical Properties of the Bagnold Dunes, Mars: Ground-truthing Orbital Data"
} | null | null | null | null | true | null | 2365 | null | Default | null | null |
null | {
"abstract": " Recovering low-rank structures via eigenvector perturbation analysis is a\ncommon problem in statistical machine learning, such as in factor analysis,\ncommunity detection, ranking, matrix completion, among others. While a large\nvariety of results provide tight bounds on the average errors between empirical\nand population statistics of eigenvectors, fewer results are tight for\nentrywise analyses, which are critical for a number of problems such as\ncommunity detection and ranking.\nThis paper investigates the entrywise perturbation analysis for a large class\nof random matrices whose expectations are low-rank, including community\ndetection, synchronization ($\\mathbb{Z}_2$-spiked Wigner model) and matrix\ncompletion models. Denoting by $\\{u_k\\}$, respectively $\\{u_k^*\\}$, the\neigenvectors of a random matrix $A$, respectively $\\mathbb{E} A$, the paper\ncharacterizes cases for which $$u_k \\approx \\frac{A u_k^*}{\\lambda_k^*}$$\nserves as a first-order approximation under the $\\ell_\\infty$ norm. The fact\nthat the approximation is both tight and linear in the random matrix $A$ allows\nfor sharp comparisons of $u_k$ and $u_k^*$. In particular, it allows to compare\nthe signs of $u_k$ and $u_k^*$ even when $\\| u_k - u_k^*\\|_{\\infty}$ is large,\nwhich in turn allows to settle the conjecture in Abbe et al. (2016) that the\nspectral algorithm achieves exact recovery in the stochastic block model\nwithout any trimming or cleaning steps. The results are further extended to the\nperturbation of eigenspaces, providing new bounds for $\\ell_\\infty$-type errors\nin noisy matrix completion.\n",
"title": "Entrywise Eigenvector Analysis of Random Matrices with Low Expected Rank"
} | null | null | null | null | true | null | 2366 | null | Default | null | null |
null | {
"abstract": " We present the use of the fitted Q iteration in algorithmic trading. We show\nthat the fitted Q iteration helps alleviate the dimension problem that the\nbasic Q-learning algorithm faces in application to trading. Furthermore, we\nintroduce a procedure including model fitting and data simulation to enrich\ntraining data as the lack of data is often a problem in realistic application.\nWe experiment our method on both simulated environment that permits arbitrage\nopportunity and real-world environment by using prices of 450 stocks. In the\nformer environment, the method performs well, implying that our method works in\ntheory. To perform well in the real-world environment, the agents trained might\nrequire more training (iteration) and more meaningful variables with predictive\nvalue.\n",
"title": "Algorithmic Trading with Fitted Q Iteration and Heston Model"
} | null | null | null | null | true | null | 2367 | null | Default | null | null |
null | {
"abstract": " Finite rank median spaces are a simultaneous generalisation of finite\ndimensional CAT(0) cube complexes and real trees. If $\\Gamma$ is an irreducible\nlattice in a product of rank one simple Lie groups, we show that every action\nof $\\Gamma$ on a complete, finite rank median space has a global fixed point.\nThis is in sharp contrast with the behaviour of actions on infinite rank median\nspaces.\nThe fixed point property is obtained as corollary to a superrigidity result;\nthe latter holds for irreducible lattices in arbitrary products of compactly\ngenerated groups.\nIn previous work, we introduced \"Roller compactifications\" of median spaces;\nthese generalise a well-known construction in the case of cube complexes. We\nprovide a reduced $1$-cohomology class that detects group actions with a finite\norbit in the Roller compactification. Even for CAT(0) cube complexes, only\nsecond bounded cohomology classes were known with this property, due to\nChatterji-Fernós-Iozzi. As a corollary, we observe that, in Gromov's density\nmodel, random groups at low density do not have Shalom's property $H_{FD}$.\n",
"title": "Superrigidity of actions on finite rank median spaces"
} | null | null | null | null | true | null | 2368 | null | Default | null | null |
null | {
"abstract": " Explaining underlying causes or effects about events is a challenging but\nvaluable task. We define a novel problem of generating explanations of a time\nseries event by (1) searching cause and effect relationships of the time series\nwith textual data and (2) constructing a connecting chain between them to\ngenerate an explanation. To detect causal features from text, we propose a\nnovel method based on the Granger causality of time series between features\nextracted from text such as N-grams, topics, sentiments, and their composition.\nThe generation of the sequence of causal entities requires a commonsense\ncausative knowledge base with efficient reasoning. To ensure good\ninterpretability and appropriate lexical usage we combine symbolic and neural\nrepresentations, using a neural reasoning algorithm trained on commonsense\ncausal tuples to predict the next cause step. Our quantitative and human\nanalysis show empirical evidence that our method successfully extracts\nmeaningful causality relationships between time series with textual features\nand generates appropriate explanation between them.\n",
"title": "Detecting and Explaining Causes From Text For a Time Series Event"
} | null | null | null | null | true | null | 2369 | null | Default | null | null |
null | {
"abstract": " We propose a type system for reasoning on protocol conformance and deadlock\nfreedom in networks of processes that communicate through unordered mailboxes.\nWe model these networks in the mailbox calculus, a mild extension of the\nasynchronous {\\pi}-calculus with first-class mailboxes and selective input. The\ncalculus subsumes the actor model and allows us to analyze networks with\ndynamic topologies and varying number of processes possibly mixing different\nconcurrency abstractions. Well-typed processes are deadlock free and never fail\nbecause of unexpected messages. For a non-trivial class of them, junk freedom\nis also guaranteed. We illustrate the expressiveness of the calculus and of the\ntype system by encoding instances of non-uniform, concurrent objects, binary\nsessions extended with joins and forks, and some known actor benchmarks.\n",
"title": "Mailbox Types for Unordered Interactions"
} | null | null | null | null | true | null | 2370 | null | Default | null | null |
null | {
"abstract": " A homomorphism from a graph G to a graph H is a function from the vertices of\nG to the vertices of H that preserves edges. A homomorphism is surjective if it\nuses all of the vertices of H and it is a compaction if it uses all of the\nvertices of H and all of the non-loop edges of H. Hell and Nesetril gave a\ncomplete characterisation of the complexity of deciding whether there is a\nhomomorphism from an input graph G to a fixed graph H. A complete\ncharacterisation is not known for surjective homomorphisms or for compactions,\nthough there are many interesting results. Dyer and Greenhill gave a complete\ncharacterisation of the complexity of counting homomorphisms from an input\ngraph G to a fixed graph H. In this paper, we give a complete characterisation\nof the complexity of counting surjective homomorphisms from an input graph G to\na fixed graph H and we also give a complete characterisation of the complexity\nof counting compactions from an input graph G to a fixed graph H. In an\naddendum we use our characterisations to point out a dichotomy for the\ncomplexity of the respective approximate counting problems (in the connected\ncase).\n",
"title": "The Complexity of Counting Surjective Homomorphisms and Compactions"
} | null | null | null | null | true | null | 2371 | null | Default | null | null |
null | {
"abstract": " We consider the spatially homogeneous Boltzmann equation for hard potentials\nwith angular cutoff. This equation has a unique conservative weak solution\n$(f_t)_{t\\geq 0}$, once the initial condition $f_0$ with finite mass and energy\nis fixed. Taking advantage of the energy conservation, we propose a recursive\nalgorithm that produces a $(0,\\infty)\\times\\mathbb{R}^3$ random variable\n$(M_t,V_t)$ such that $E[M_t {\\bf 1}_{\\{V_t \\in \\cdot\\}}]=f_t$. We also write\ndown a series expansion of $f_t$. Although both the algorithm and the series\nexpansion might be theoretically interesting in that they explicitly express\n$f_t$ in terms of $f_0$, we believe that the algorithm is not very efficient in\npractice and that the series expansion is rather intractable. This is a tedious\nextension to non-Maxwellian molecules of Wild's sum and of its interpretation\nby McKean.\n",
"title": "A recursive algorithm and a series expansion related to the homogeneous Boltzmann equation for hard potentials with angular cutoff"
} | null | null | null | null | true | null | 2372 | null | Default | null | null |
null | {
"abstract": " Materials science has adopted the term of auxetic behavior for structural\ndeformations where stretching in some direction entails lateral widening,\nrather than lateral shrinking. Most studies, in the last three decades, have\nexplored repetitive or cellular structures and used the notion of negative\nPoisson's ratio as the hallmark of auxetic behavior. However, no general\nauxetic principle has been established from this perspective. In the present\npaper, we show that a purely geometric approach to periodic auxetics is apt to\nidentify essential characteristics of frameworks with auxetic deformations and\ncan generate a systematic and endless series of periodic auxetic designs. The\ncritical features refer to convexity properties expressed through families of\nhomothetic ellipsoids.\n",
"title": "Periodic auxetics: Structure and design"
} | null | null | null | null | true | null | 2373 | null | Default | null | null |
null | {
"abstract": " Asymptotics of maximum likelihood estimation for $\\alpha$-stable law are\nanalytically investigated with $(M)$ parameterization. The consistency and\nasymptotic normality are shown on the interior of the whole parameter space.\nAlthough these asymptotics have been proved with $(B)$ parameterization, there\nare several gaps between. Especially in the latter, the density, so that scores\nand their derivatives are discontinuous at $\\alpha=1$ for $\\beta\\neq 0$ and\nusual asymptotics are impossible, whereas in $(M)$ form these quantities are\nshown to be continuous on the interior of the parameter space. We fill these\ngaps and provide a convenient theory for applied people. We numerically\napproximate the Fisher information matrix around the Cauchy law\n$(\\alpha,\\beta)=(1,0)$. The results exhibit continuity at $\\alpha=1,\\,\\beta\\neq\n0$ and this secures the accuracy of our calculations.\n",
"title": "Asymptotics of maximum likelihood estimation for stable law with $(M)$ parameterization"
} | null | null | null | null | true | null | 2374 | null | Default | null | null |
null | {
"abstract": " Deep Neural Networks (DNNs) have emerged as a core tool for machine learning.\nThe computations performed during DNN training and inference are dominated by\noperations on the weight matrices describing the DNN. As DNNs incorporate more\nstages and more nodes per stage, these weight matrices may be required to be\nsparse because of memory limitations. The GraphBLAS.org math library standard\nwas developed to provide high performance manipulation of sparse weight\nmatrices and input/output vectors. For sufficiently sparse matrices, a sparse\nmatrix library requires significantly less memory than the corresponding dense\nmatrix implementation. This paper provides a brief description of the\nmathematics underlying the GraphBLAS. In addition, the equations of a typical\nDNN are rewritten in a form designed to use the GraphBLAS. An implementation of\nthe DNN is given using a preliminary GraphBLAS C library. The performance of\nthe GraphBLAS implementation is measured relative to a standard dense linear\nalgebra library implementation. For various sizes of DNN weight matrices, it is\nshown that the GraphBLAS sparse implementation outperforms a BLAS dense\nimplementation as the weight matrix becomes sparser.\n",
"title": "Enabling Massive Deep Neural Networks with the GraphBLAS"
} | null | null | null | null | true | null | 2375 | null | Default | null | null |
null | {
"abstract": " Marshall and Olkin (1997, Biometrika, 84, 641 - 652) introduced a very\npowerful method to introduce an additional parameter to a class of continuous\ndistribution functions and hence it brings more flexibility to the model. They\nhave demonstrated their method for the exponential and Weibull classes. In the\nsame paper they have briefly indicated regarding its bivariate extension. The\nmain aim of this paper is to introduce the same method, for the first time, to\nthe class of discrete generalized exponential distributions both for the\nunivariate and bivariate cases. We investigate several properties of the\nproposed univariate and bivariate classes. The univariate class has three\nparameters, whereas the bivariate class has five parameters. It is observed\nthat depending on the parameter values the univariate class can be both zero\ninflated as well as heavy tailed. We propose to use EM algorithm to estimate\nthe unknown parameters. Small simulation experiments have been performed to see\nthe effectiveness of the proposed EM algorithm, and a bivariate data set has\nbeen analyzed and it is observed that the proposed models and the EM algorithm\nwork quite well in practice.\n",
"title": "Univariate and Bivariate Geometric Discrete Generalized Exponential Distributions"
} | null | null | null | null | true | null | 2376 | null | Default | null | null |
null | {
"abstract": " Rapid improvements in machine learning over the past decade are beginning to\nhave far-reaching effects. For communications, engineers with limited domain\nexpertise can now use off-the-shelf learning packages to design\nhigh-performance systems based on simulations. Prior to the current revolution\nin machine learning, the majority of communication engineers were quite aware\nthat system parameters (such as filter coefficients) could be learned using\nstochastic gradient descent. It was not at all clear, however, that more\ncomplicated parts of the system architecture could be learned as well. In this\npaper, we discuss the application of machine-learning techniques to two\ncommunications problems and focus on what can be learned from the resulting\nsystems. We were pleasantly surprised that the observed gains in one example\nhave a simple explanation that only became clear in hindsight. In essence, deep\nlearning discovered a simple and effective strategy that had not been\nconsidered earlier.\n",
"title": "What Can Machine Learning Teach Us about Communications?"
} | null | null | null | null | true | null | 2377 | null | Default | null | null |
null | {
"abstract": " This paper studies the dynamics of a network-based SIRS epidemic model with\nvaccination and a nonmonotone incidence rate. This type of nonlinear incidence\ncan be used to describe the psychological or inhibitory effect from the\nbehavioral change of the susceptible individuals when the number of infective\nindividuals on heterogeneous networks is getting larger. Using the analytical\nmethod, epidemic threshold $R_0$ is obtained. When $R_0$ is less than one, we\nprove the disease-free equilibrium is globally asymptotically stable and the\ndisease dies out, while $R_0$ is greater than one, there exists a unique\nendemic equilibrium. By constructing a suitable Lyapunov function, we also\nprove the endemic equilibrium is globally asymptotically stable if the\ninhibitory factor $\\alpha$ is sufficiently large. Numerical experiments are\nalso given to support the theoretical results. It is shown both theoretically\nand numerically a larger $\\alpha$ can accelerate the extinction of the disease\nand reduce the level of disease.\n",
"title": "Global stability of a network-based SIRS epidemic model with nonmonotone incidence rate"
} | null | null | null | null | true | null | 2378 | null | Default | null | null |
null | {
"abstract": " For every $q\\in \\mathbb N$ let $\\textrm{FO}_q$ denote the class of sentences\nof first-order logic FO of quantifier rank at most $q$. If a graph property can\nbe defined in $\\textrm{FO}_q$, then it can be decided in time $O(n^q)$. Thus,\nminimizing $q$ has favorable algorithmic consequences. Many graph properties\namount to the existence of a certain set of vertices of size $k$. Usually this\ncan only be expressed by a sentence of quantifier rank at least $k$. We use the\ncolor-coding method to demonstrate that some (hyper)graph problems can be\ndefined in $\\textrm{FO}_q$ where $q$ is independent of $k$. This property of a\ngraph problem is equivalent to the question of whether the corresponding\nparameterized problem is in the class $\\textrm{para-AC}^0$.\nIt is crucial for our results that the FO-sentences have access to built-in\naddition and multiplication. It is known that then FO corresponds to the\ncircuit complexity class uniform $\\textrm{AC}^0$. We explore the connection\nbetween the quantifier rank of FO-sentences and the depth of\n$\\textrm{AC}^0$-circuits, and prove that $\\textrm{FO}_q \\subsetneq\n\\textrm{FO}_{q+1}$ for structures with built-in addition and multiplication.\n",
"title": "Slicewise definability in first-order logic with bounded quantifier rank"
} | null | null | null | null | true | null | 2379 | null | Default | null | null |
null | {
"abstract": " Community analysis is an important way to ascertain whether or not a complex\nsystem consists of sub-structures with different properties. In this paper, we\ngive a two level community structure analysis for the SSCI journal system by\nmost similar co-citation pattern. Five different strategies for the selection\nof most similar node (journal) pairs are introduced. The efficiency is checked\nby the normalized mutual information technique. Statistical properties and\ncomparisons of the community results show that both of the two level detection\ncould give instructional information for the community structure of complex\nsystems. Further comparisons of the five strategies indicates that, the most\nefficient strategy is to assign nodes with maximum similarity into the same\ncommunity whether the similarity information is complete or not, while random\nselection generates small world local community with no inside order. These\nresults give valuable indication for efficient community detection by most\nsimilar node pairs.\n",
"title": "The efficiency of community detection by most similar node pairs"
} | null | null | [
"Computer Science"
]
| null | true | null | 2380 | null | Validated | null | null |
null | {
"abstract": " Composite adaptive control schemes, which use both the system tracking errors\nand the prediction error to drive the update laws, have become widespread in\nachieving an improvement of system performance. However, a strong\npersistent-excitation (PE) condition should be satisfied to guarantee the\nparameter convergence. This paper proposes a novel composite adaptive control\nto guarantee parameter convergence without PE condition for nonlinear\nteleoperation systems with dynamic uncertainties and time-varying communication\ndelays. The stability criteria of the closed-loop teleoperation system are\ngiven in terms of linear matrix inequalities. New tracking performance measures\nare proposed to evaluate the position tracking between the master and the\nslave. Simulation studies are given to show the effectiveness of the proposed\nmethod.\n",
"title": "Composite Adaptive Control for Bilateral Teleoperation Systems without Persistency of Excitation"
} | null | null | null | null | true | null | 2381 | null | Default | null | null |
null | {
"abstract": " We introduce a new critical value $c_\\infty(L)$ for Tonelli Lagrangians $L$\non the tangent bundle of the 2-sphere without minimizing measures supported on\na point. We show that $c_\\infty(L)$ is strictly larger than the Mañé\ncritical value $c(L)$, and on every energy level $e\\in(c(L),c_\\infty(L))$ there\nexist infinitely many periodic orbits of the Lagrangian system of $L$, one of\nwhich is a local minimizer of the free-period action functional. This has\napplications to Finsler metrics of Randers type on the 2-sphere. We show that,\nunder a suitable criticality assumption on a given Randers metric, after\nrescaling its magnetic part with a sufficiently large multiplicative constant,\nthe new metric admits infinitely many closed geodesics, one of which is a\nwaist. Examples of critical Randers metrics include the celebrated Katok\nmetric.\n",
"title": "Infinitely many periodic orbits just above the Mañé critical value on the 2-sphere"
} | null | null | null | null | true | null | 2382 | null | Default | null | null |
null | {
"abstract": " We analyze a decentralized random walk-based algorithm for data collection at\nthe sink in a multi-hop sensor network. Our algorithm, Random-Collect, which\ninvolves data packets being passed to random neighbors in the network according\nto a random walk mechanism, requires no configuration and incurs no routing\noverhead. To analyze this method, we model the data generation process as\nindependent Bernoulli arrivals at the source nodes. We analyze both latency and\nthroughput in this setting, providing a theoretical lower bound for the\nthroughput and a theoretical upper bound for the latency. The main contribution\nof our paper, however, is the throughput result: we present a general lower\nbound on the throughput achieved by our data collection method in terms of the\nunderlying network parameters. In particular, we show that the rate at which\nour algorithm can collect data depends on the spectral gap of the given random\nwalk's transition matrix and if the random walk is simple then it also depends\non the maximum and minimum degrees of the graph modeling the network. For\nlatency, we show that the time taken to collect data not only depends on the\nworst-case hitting time of the given random walk but also depends on the data\narrival rate. In fact, our latency bound reflects the data rate-latency\ntrade-off i.e., in order to achieve a higher data rate we need to compromise on\nlatency and vice-versa. We also discuss some examples that demonstrate that our\nlower bound on the data rate is optimal up to constant factors, i.e., there\nexists a network topology and sink placement for which the maximum stable data\nrate is just a constant factor above our lower bound.\n",
"title": "Decentralized Random Walk-Based Data Collection in Networks"
} | null | null | null | null | true | null | 2383 | null | Default | null | null |
null | {
"abstract": " The Fan Region is one of the dominant features in the polarized radio sky,\nlong thought to be a local (distance < 500 pc) synchrotron feature. We present\n1.3-1.8 GHz polarized radio continuum observations of the region from the\nGlobal Magneto-Ionic Medium Survey (GMIMS) and compare them to maps of Halpha\nand polarized radio continuum intensity from 0.408-353 GHz. The high-frequency\n(> 1 GHz) and low-frequency (< 600 MHz) emission have different morphologies,\nsuggesting a different physical origin. Portions of the 1.5 GHz Fan Region\nemission are depolarized by about 30% by ionized gas structures in the Perseus\nArm, indicating that this fraction of the emission originates >2 kpc away. We\nargue for the same conclusion based on the high polarization fraction at 1.5\nGHz (about 40%). The Fan Region is offset with respect to the Galactic plane,\ncovering -5° < b < +10°; we attribute this offset to the warp in the\nouter Galaxy. We discuss origins of the polarized emission, including the\nspiral Galactic magnetic field. This idea is a plausible contributing factor\nalthough no model to date readily reproduces all of the observations. We\nconclude that models of the Galactic magnetic field should account for the > 1\nGHz emission from the Fan Region as a Galactic-scale, not purely local,\nfeature.\n",
"title": "The Fan Region at 1.5 GHz. I: Polarized synchrotron emission extending beyond the Perseus Arm"
} | null | null | null | null | true | null | 2384 | null | Default | null | null |
null | {
"abstract": " Determining the redshift distribution $n(z)$ of galaxy samples is essential\nfor several cosmological probes including weak lensing. For imaging surveys,\nthis is usually done using photometric redshifts estimated on an\nobject-by-object basis. We present a new approach for directly measuring the\nglobal $n(z)$ of cosmological galaxy samples, including uncertainties, using\nforward modeling. Our method relies on image simulations produced using UFig\n(Ultra Fast Image Generator) and on ABC (Approximate Bayesian Computation)\nwithin the $MCCL$ (Monte-Carlo Control Loops) framework. The galaxy population\nis modeled using parametric forms for the luminosity functions, spectral energy\ndistributions, sizes and radial profiles of both blue and red galaxies. We\napply exactly the same analysis to the real data and to the simulated images,\nwhich also include instrumental and observational effects. By adjusting the\nparameters of the simulations, we derive a set of acceptable models that are\nstatistically consistent with the data. We then apply the same cuts to the\nsimulations that were used to construct the target galaxy sample in the real\ndata. The redshifts of the galaxies in the resulting simulated samples yield a\nset of $n(z)$ distributions for the acceptable models. We demonstrate the\nmethod by determining $n(z)$ for a cosmic shear like galaxy sample from the\n4-band Subaru Suprime-Cam data in the COSMOS field. We also complement this\nimaging data with a spectroscopic calibration sample from the VVDS survey. We\ncompare our resulting posterior $n(z)$ distributions to the one derived from\nphotometric redshifts estimated using 36 photometric bands in COSMOS and find\ngood agreement. This offers good prospects for applying our approach to current\nand future large imaging surveys.\n",
"title": "The redshift distribution of cosmological samples: a forward modeling approach"
} | null | null | null | null | true | null | 2385 | null | Default | null | null |
null | {
"abstract": " In this paper we develop a novel computational sensing framework for sensing\nand recovering structured signals. When trained on a set of representative\nsignals, our framework learns to take undersampled measurements and recover\nsignals from them using a deep convolutional neural network. In other words, it\nlearns a transformation from the original signals to a near-optimal number of\nundersampled measurements and the inverse transformation from measurements to\nsignals. This is in contrast to traditional compressive sensing (CS) systems\nthat use random linear measurements and convex optimization or iterative\nalgorithms for signal recovery. We compare our new framework with\n$\\ell_1$-minimization from the phase transition point of view and demonstrate\nthat it outperforms $\\ell_1$-minimization in the regions of phase transition\nplot where $\\ell_1$-minimization cannot recover the exact solution. In\naddition, we experimentally demonstrate how learning measurements enhances the\noverall recovery performance, speeds up training of recovery framework, and\nleads to having fewer parameters to learn.\n",
"title": "DeepCodec: Adaptive Sensing and Recovery via Deep Convolutional Neural Networks"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2386 | null | Validated | null | null |
null | {
"abstract": " Nowadays, the construction of a complex robotic system requires a high level\nof specialization in a large number of diverse scientific areas. It is\nreasonable that a single researcher cannot create from scratch the entirety of\nthis system, as it is impossible for him to have the necessary skills in the\nnecessary fields. This obstacle is being surpassed with the existent robotic\nframeworks. This paper tries to give an extensive review of the most famous\nrobotic frameworks and middleware, as well as to provide the means to\neffortlessly compare them. Additionally, we try to investigate the differences\nbetween the definitions of a robotic framework, a robotic middleware and a\nrobotic architecture.\n",
"title": "Robotic frameworks, architectures and middleware comparison"
} | null | null | null | null | true | null | 2387 | null | Default | null | null |
null | {
"abstract": " The joint sparse recovery problem is a generalization of the single\nmeasurement vector problem which is widely studied in Compressed Sensing and it\naims to recovery a set of jointly sparse vectors. i.e. have nonzero entries\nconcentrated at common location. Meanwhile l_p-minimization subject to matrices\nis widely used in a large number of algorithms designed for this problem.\nTherefore the main contribution in this paper is two theoretical results about\nthis technique. The first one is to prove that in every multiple systems of\nlinear equation, there exists a constant p* such that the original unique\nsparse solution also can be recovered from a minimization in l_p quasi-norm\nsubject to matrices whenever 0< p<p*. The other one is to show an analysis\nexpression of such p*. Finally, we display the results of one example to\nconfirm the validity of our conclusions.\n",
"title": "Analysis of equivalence relation in joint sparse recovery"
} | null | null | [
"Computer Science",
"Mathematics"
]
| null | true | null | 2388 | null | Validated | null | null |
null | {
"abstract": " The dynamics of infectious diseases spread is crucial in determining their\nrisk and offering ways to contain them. We study sequential vaccination of\nindividuals in networks. In the original (deterministic) version of the\nFirefighter problem, a fire breaks out at some node of a given graph. At each\ntime step, b nodes can be protected by a firefighter and then the fire spreads\nto all unprotected neighbors of the nodes on fire. The process ends when the\nfire can no longer spread. We extend the Firefighter problem to a probabilistic\nsetting, where the infection is stochastic. We devise a simple policy that only\nvaccinates neighbors of infected nodes and is optimal on regular trees and on\ngeneral graphs for a sufficiently large budget. We derive methods for\ncalculating upper and lower bounds of the expected number of infected\nindividuals, as well as provide estimates on the budget needed for containment\nin expectation. We calculate these explicitly on trees, d-dimensional grids,\nand Erdős Rényi graphs. Finally, we construct a state-dependent budget\nallocation strategy and demonstrate its superiority over constant budget\nallocation on real networks following a first order acquaintance vaccination\npolicy.\n",
"title": "The Stochastic Firefighter Problem"
} | null | null | null | null | true | null | 2389 | null | Default | null | null |
null | {
"abstract": " We investigate the dynamical complexity of Cournot oligopoly dynamics of\nthree firms by using the qualitative methods of dynamical systems to study the\nphase structure of this model. The phase space is organized with\none-dimensional and two-dimensional invariant submanifolds (for the monopoly\nand duopoly) and unique stable node (global attractor) in the positive quadrant\nof the phase space (Cournot equilibrium). We also study the integrability of\nthe system. We demonstrate the effectiveness of the method of the Darboux\npolynomials in searching for first integrals of the oligopoly. The general\nmethod as well as examples of adopting this method are presented. We study\nDarboux non-integrability of the oligopoly for linear demand functions and find\nfirst integrals of this system for special classes of the system, in\nparticular, rational integrals can be found for a quite general set of model\nparameters. We show how first integral can be useful in lowering the dimension\nof the system using the example of $n$ almost identical firms. This first\nintegral also gives information about the structure of the phase space and the\nbehaviour of trajectories in the neighbourhood of a Nash equilibrium\n",
"title": "The phase space structure of the oligopoly dynamical system by means of Darboux integrability"
} | null | null | null | null | true | null | 2390 | null | Default | null | null |
null | {
"abstract": " Mathematical modelers have long known of a \"rule of thumb\" referred to as the\nLinear Chain Trick (LCT; aka the Gamma Chain Trick): a technique used to\nconstruct mean field ODE models from continuous-time stochastic state\ntransition models where the time an individual spends in a given state (i.e.,\nthe dwell time) is Erlang distributed (i.e., gamma distributed with integer\nshape parameter). Despite the LCT's widespread use, we lack general theory to\nfacilitate the easy application of this technique, especially for complex\nmodels. This has forced modelers to choose between constructing ODE models\nusing heuristics with oversimplified dwell time assumptions, using time\nconsuming derivations from first principles, or to instead use non-ODE models\n(like integro-differential equations or delay differential equations) which can\nbe cumbersome to derive and analyze. Here, we provide analytical results that\nenable modelers to more efficiently construct ODE models using the LCT or\nrelated extensions. Specifically, we 1) provide novel extensions of the LCT to\nvarious scenarios found in applications; 2) provide formulations of the LCT and\nit's extensions that bypass the need to derive ODEs from integral or stochastic\nmodel equations; and 3) introduce a novel Generalized Linear Chain Trick (GLCT)\nframework that extends the LCT to a much broader family of distributions,\nincluding the flexible phase-type distributions which can approximate\ndistributions on $\\mathbb{R}^+$ and be fit to data. These results give modelers\nmore flexibility to incorporate appropriate dwell time assumptions into mean\nfield ODEs, including conditional dwell time distributions, and these results\nhelp clarify connections between individual-level stochastic model assumptions\nand the structure of corresponding mean field ODEs.\n",
"title": "Generalizations of the 'Linear Chain Trick': Incorporating more flexible dwell time distributions into mean field ODE models"
} | null | null | null | null | true | null | 2391 | null | Default | null | null |
null | {
"abstract": " Egbert Brieskorn died on July 11, 2013, a few days after his 77th birthday.\nHe was an impressive personality who has left a lasting impression on all who\nknew him, whether inside or outside of mathematics. Brieskorn was a great\nmathematician, but his interests, his knowledge, and activities ranged far\nbeyond mathematics. In this contribution, which is strongly influenced by many\nyears of personal connectedness of the authors with Brieskorn, we try to give a\ndeeper insight into the life and work of Brieskorn. We illuminate both his\npersonal commitment to peace and the environment as well as his long-term study\nof the life and work of Felix Hausdorff and the publication of Hausdorff's\ncollected works. However, the main focus of the article is on the presentation\nof his remarkable and influential mathematical work.\n",
"title": "Life and work of Egbert Brieskorn (1936 - 2013)"
} | null | null | [
"Mathematics"
]
| null | true | null | 2392 | null | Validated | null | null |
null | {
"abstract": " Exploiting full-duplex (FD) technology on base stations (BSs) is a promising\nsolution to enhancing the system performance. Motivated by this, we revisit a\nfull-duplex base station (FD-BS) aided OFDMA system, which consists of one BS,\nseveral uplink/downlink users and multiple subcarriers. A joint 3-dimensional\n(3D) mapping scheme among subcarriers, down-link users (DUEs), uplink users\n(UUEs) is considered as well as an associated power allocation optimization. In\ndetail, we first decompose the complex 3D mapping problem into three\n2-dimensional sub ones and solve them by using the iterative Hungarian method,\nrespectively. Then based on the Lagrange dual method, we sequentially solve the\npower allocation and 3- dimensional mapping problem by fixing a dual point.\nFinally, the optimal solution can be obtained by utilizing the sub-gradient\nmethod. Unlike existing work that only solves either 3D mapping or power\nallocation problem but with a high computation complexity, we tackle both of\nthem and have successfully reduced computation complexity from exponential to\npolynomial order. Numerical simulations are conducted to verify the proposed\nscheme.\n",
"title": "Resource Allocation for a Full-Duplex Base Station Aided OFDMA System"
} | null | null | [
"Computer Science"
]
| null | true | null | 2393 | null | Validated | null | null |
null | {
"abstract": " We consider task and motion planning in complex dynamic environments for\nproblems expressed in terms of a set of Linear Temporal Logic (LTL)\nconstraints, and a reward function. We propose a methodology based on\nreinforcement learning that employs deep neural networks to learn low-level\ncontrol policies as well as task-level option policies. A major challenge in\nthis setting, both for neural network approaches and classical planning, is the\nneed to explore future worlds of a complex and interactive environment. To this\nend, we integrate Monte Carlo Tree Search with hierarchical neural net control\npolicies trained on expressive LTL specifications. This paper investigates the\nability of neural networks to learn both LTL constraints and control policies\nin order to generate task plans in complex environments. We demonstrate our\napproach in a simulated autonomous driving setting, where a vehicle must drive\ndown a road in traffic, avoid collisions, and navigate an intersection, all\nwhile obeying given rules of the road.\n",
"title": "Combining Neural Networks and Tree Search for Task and Motion Planning in Challenging Environments"
} | null | null | null | null | true | null | 2394 | null | Default | null | null |
null | {
"abstract": " Recently, deep neural networks (DNNs) have been regarded as the\nstate-of-the-art classification methods in a wide range of applications,\nespecially in image classification. Despite the success, the huge number of\nparameters blocks its deployment to situations with light computing resources.\nResearchers resort to the redundancy in the weights of DNNs and attempt to find\nhow fewer parameters can be chosen while preserving the accuracy at the same\ntime. Although several promising results have been shown along this research\nline, most existing methods either fail to significantly compress a\nwell-trained deep network or require a heavy fine-tuning process for the\ncompressed network to regain the original performance. In this paper, we\npropose the \\textit{Block Term} networks (BT-nets) in which the commonly used\nfully-connected layers (FC-layers) are replaced with block term layers\n(BT-layers). In BT-layers, the inputs and the outputs are reshaped into two\nlow-dimensional high-order tensors, then block-term decomposition is applied as\ntensor operators to connect them. We conduct extensive experiments on benchmark\ndatasets to demonstrate that BT-layers can achieve a very large compression\nratio on the number of parameters while preserving the representation power of\nthe original FC-layers as much as possible. Specifically, we can get a higher\nperformance while requiring fewer parameters compared with the tensor train\nmethod.\n",
"title": "BT-Nets: Simplifying Deep Neural Networks via Block Term Decomposition"
} | null | null | null | null | true | null | 2395 | null | Default | null | null |
null | {
"abstract": " Sequential Monte Carlo (SMC) methods are a class of Monte Carlo methods that\nare used to obtain random samples of a high dimensional random variable in a\nsequential fashion. Many problems encountered in applications often involve\ndifferent types of constraints. These constraints can make the problem much\nmore challenging. In this paper, we formulate a general framework of using SMC\nfor constrained sampling problems based on forward and backward pilot\nresampling strategies. We review some existing methods under the framework and\ndevelop several new algorithms. It is noted that all information observed or\nimposed on the underlying system can be viewed as constraints. Hence the\napproach outlined in this paper can be useful in many applications.\n",
"title": "Resampling Strategy in Sequential Monte Carlo for Constrained Sampling Problems"
} | null | null | null | null | true | null | 2396 | null | Default | null | null |
null | {
"abstract": " An incoming electron is reflected back as a hole at a\nnormal-metal-superconductor interface, a process known as Andreev reflection.\nWe predict that there exists a universal transverse shift in this process due\nto the effect of spin-orbit coupling in the normal metal. Particularly, using\nboth the scattering approach and the argument of angular momentum conservation,\nwe demonstrate that the shifts are pronounced for lightly-doped Weyl\nsemimetals, and are opposite for incoming electrons with different chirality,\ngenerating a chirality-dependent Hall effect for the reflected holes. The\npredicted shift is not limited to Weyl systems, but exists for a general\nthree-dimensional spin-orbit- coupled metal interfaced with a superconductor.\n",
"title": "Transverse Shift in Andreev Reflection"
} | null | null | null | null | true | null | 2397 | null | Default | null | null |
null | {
"abstract": " We examine the dynamics of entanglement entropy of all parts in an open\nsystem consisting of a two-level dimer interacting with an environment of\noscillators. The dimer-environment interaction is almost energy conserving. We\nfind the precise link between decoherence and production of entanglement\nentropy. We show that not all environment oscillators carry significant\nentanglement entropy and we identify the oscillator frequency regions which\ncontribute to the production of entanglement entropy. Our results hold for\narbitrary strengths of the dimer-environment interaction, and they are\nmathematically rigorous.\n",
"title": "Production of Entanglement Entropy by Decoherence"
} | null | null | null | null | true | null | 2398 | null | Default | null | null |
null | {
"abstract": " Aggressive incentive schemes that allow individuals to impose economic\npunishment on themselves if they fail to meet health goals present a promising\napproach for encouraging healthier behavior. However, the element of choice\ninherent in these schemes introduces concerns that only non-representative\nsectors of the population will select aggressive incentives, leaving value on\nthe table for those who don't opt in. In a field experiment conducted over a 29\nweek period on individuals wearing Fitbit activity trackers, we find modest and\nshort lived increases in physical activity for those provided the choice of\naggressive incentives. In contrast, we find significant and persistent\nincreases for those assigned (oftentimes against their stated preference) to\nthe same aggressive incentives. The modest benefits for those provided a choice\nseems to emerge because those who benefited most from the aggressive incentives\nwere the least likely to choose them, and it was those who did not need them\nwho opted in. These results are confirmed in a follow up lab experiment. We\nalso find that benefits to individuals assigned to aggressive incentives were\npronounced if they also updated their step target in the Fitbit mobile\napplication to match the new activity goal we provided them. Our findings have\nimportant implications for incentive based interventions to improve health\nbehavior. For firms and policy makers, our results suggest that one effective\nstrategy for encouraging sustained healthy behavior combines exposure to\naggressive incentive schemes to jolt individuals out of their comfort zones\nwith technology decision aids that help individuals sustain this behavior after\nincentives end.\n",
"title": "Aggressive Economic Incentives and Physical Activity: The Role of Choice and Technology Decision Aids"
} | null | null | null | null | true | null | 2399 | null | Default | null | null |
null | {
"abstract": " The nearby space surrounding the Earth is densely populated by artificial\nsatellites and instruments, whose orbits are distributed within the\nLow-Earth-Orbit region (LEO), ranging between 90 and 2 000 $km$ of altitude. As\na consequence of collisions and fragmentations, many space debris of different\nsizes are left in the LEO region. Given the threat raised by the possible\ndamages which a collision of debris can provoke with operational or manned\nsatellites, the study of their dynamics is nowadays mandatory. This work is\nfocused on the existence of equilibria and the dynamics of resonances in LEO.\nWe base our results on a simplified model which includes the geopotential and\nthe atmospheric drag. Using such model, we make a qualitative study of the\nresonances and the equilibrium positions, including their location and\nstability. The dissipative effect due to the atmosphere provokes a tidal decay,\nbut we give examples of different behaviors, precisely a straightforward\npassage through the resonance or rather a temporary capture. We also\ninvestigate the effect of the solar cycle which is responsible of fluctuations\nof the atmospheric density and we analyze the influence of Sun and Moon on LEO\nobjects.\n",
"title": "Dynamics of resonances and equilibria of Low Earth Objects"
} | null | null | null | null | true | null | 2400 | null | Default | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.