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": " In this paper, we study an analytical approach to selecting expansion\nlocations for retailers selling add-on products whose demand is derived from\nthe demand of another base product. Demand for the add-on product is realized\nonly as a supplement to the demand of the base product. In our context, either\nof the two products could be subject to spatial autocorrelation where demand at\na given location is impacted by demand at other locations. Using data from an\nindustrial partner selling add-on products, we build predictive models for\nunderstanding the derived demand of the add-on product and establish an\noptimization framework for automating expansion decisions to maximize expected\nsales. Interestingly, spatial autocorrelation and the complexity of the\npredictive model impact the complexity and the structure of the prescriptive\noptimization model. Our results indicate that the models formulated are highly\neffective in predicting add-on product sales, and that using the optimization\nframework built on the predictive model can result in substantial increases in\nexpected sales over baseline policies.\n",
"title": "Predictive and Prescriptive Analytics for Location Selection of Add-on Retail Products"
} | null | null | null | null | true | null | 2601 | null | Default | null | null |
null | {
"abstract": " In this paper we study the behavior of the fractions of a factorial design\nunder permutations of the factor levels. We focus on the notion of regular\nfraction and we introduce methods to check whether a given symmetric orthogonal\narray can or can not be transformed into a regular fraction by means of\nsuitable permutations of the factor levels. The proposed techniques take\nadvantage of the complex coding of the factor levels and of some tools from\npolynomial algebra. Several examples are described, mainly involving factors\nwith five levels.\n",
"title": "Algebraic characterization of regular fractions under level permutations"
} | null | null | [
"Mathematics",
"Statistics"
]
| null | true | null | 2602 | null | Validated | null | null |
null | {
"abstract": " Biomedical events describe complex interactions between various biomedical\nentities. Event trigger is a word or a phrase which typically signifies the\noccurrence of an event. Event trigger identification is an important first step\nin all event extraction methods. However many of the current approaches either\nrely on complex hand-crafted features or consider features only within a\nwindow. In this paper we propose a method that takes the advantage of recurrent\nneural network (RNN) to extract higher level features present across the\nsentence. Thus hidden state representation of RNN along with word and entity\ntype embedding as features avoid relying on the complex hand-crafted features\ngenerated using various NLP toolkits. Our experiments have shown to achieve\nstate-of-art F1-score on Multi Level Event Extraction (MLEE) corpus. We have\nalso performed category-wise analysis of the result and discussed the\nimportance of various features in trigger identification task.\n",
"title": "Biomedical Event Trigger Identification Using Bidirectional Recurrent Neural Network Based Models"
} | null | null | null | null | true | null | 2603 | null | Default | null | null |
null | {
"abstract": " Nowadays it is quite evident that knowledge-based society necessarily\ninvolves the revaluation of human and intangible assets, as the advancement of\nlocal economies significantly depend on the qualitative and quantitative\ncharacteristics of human capital[Lundvall, 2004]. As we can instantaneously\nlink the universities as main actors in the creation of highly-qualified labour\nforce, the role of universities increases parallel to the previously mentioned\nprogresses. Universities are the general institutions of education, however i\nnthe need of adaptation to present local needs, their activities have broadened\nin the past decades [Wright et al, 2008; Etzkowitz, 2002]. Most universities\nexperienced a transition period in which next to their classic activities,\nnamely education and research, so called third mission activities also started\nto count, thus serving many purposes of economy and society.\n",
"title": "Modern-day Universities and Regional Development"
} | null | null | null | null | true | null | 2604 | null | Default | null | null |
null | {
"abstract": " Bilinear matrix inequality (BMI) problems in system and control designs are\ninvestigated in this paper. A solution method of reduction of variables (MRVs)\nis proposed. This method consists of a principle of variable classification, a\nprocedure for problem transformation, and a hybrid algorithm that combines\ndeterministic and stochastic search engines. The classification principle is\nused to classify the decision variables of a BMI problem into two categories:\n1) external and 2) internal variables. Theoretical analysis is performed to\nshow that when the classification principle is applicable, a BMI problem can be\ntransformed into an unconstrained optimization problem that has fewer decision\nvariables. Stochastic search and deterministic search are then applied to\ndetermine the decision variables of the unconstrained problem externally and\nexplore the internal problem structure, respectively. The proposed method can\naddress feasibility, single-objective, and multiobjective problems constrained\nby BMIs in a unified manner. A number of numerical examples in system and\ncontrol designs are provided to validate the proposed methodology. Simulations\nshow that the MRVs can outperform existing BMI solution methods in most\nbenchmark problems and achieve similar levels of performance in the remaining\nproblems.\n",
"title": "Method of Reduction of Variables for Bilinear Matrix Inequality Problems in System and Control Designs"
} | null | null | [
"Computer Science"
]
| null | true | null | 2605 | null | Validated | null | null |
null | {
"abstract": " We have carried out the transient nonlinear transport measurements on the\nlayered cobalt oxide Ca$_3$Co$_{4}$O$_9$, in which a spin density wave (SDW)\ntransition is proposed at $T_{\\rm SDW} \\simeq 30$ K. We find that, below\n$T_{\\rm SDW}$, the electrical conductivity systematically varies with both the\napplied current and the time, indicating a close relationship between the\nobserved nonlinear conduction and the SDW order in this material. The time\ndependence of the conductivity is well analyzed by considering the dynamics of\nSDW which involves a low-field deformation and a sliding motion above a\nthreshold field. We also measure the transport properties of the isovalent\nSr-substituted systems to examine an impurity effect on the nonlinear response,\nand discuss the obtained threshold fields in terms of thermal fluctuations of\nthe SDW order parameter.\n",
"title": "Nonlinear transport associated with spin-density-wave dynamics in Ca$_3$Co$_{4}$O$_9$"
} | null | null | null | null | true | null | 2606 | null | Default | null | null |
null | {
"abstract": " Compression and computational efficiency in deep learning have become a\nproblem of great significance. In this work, we argue that the most principled\nand effective way to attack this problem is by adopting a Bayesian point of\nview, where through sparsity inducing priors we prune large parts of the\nnetwork. We introduce two novelties in this paper: 1) we use hierarchical\npriors to prune nodes instead of individual weights, and 2) we use the\nposterior uncertainties to determine the optimal fixed point precision to\nencode the weights. Both factors significantly contribute to achieving the\nstate of the art in terms of compression rates, while still staying competitive\nwith methods designed to optimize for speed or energy efficiency.\n",
"title": "Bayesian Compression for Deep Learning"
} | null | null | null | null | true | null | 2607 | null | Default | null | null |
null | {
"abstract": " In the discrete modeling approach for hybrid control systems, the continuous\nplant is reduced to a discrete event approximation, called the DES-plant, that\nis governed by a discrete event system, representing the controller. The\nobservability of the DES-plant model is crucial for the synthesis of the\ncontroller and for the proper closed loop evolution of the hybrid control\nsystem. Based on a version of the framework for hybrid control systems proposed\nby Antsaklis, the paper analysis the relation between the properties of the\ncellular space of the continuous plant and a mechanism of plant-symbols\ngeneration, on one side, and the observability of the DES-plant automaton on\nthe other side. Finally an observable discrete event abstraction of the\ncontinuous double integrator is presented.\n",
"title": "The Observability Concept in a Class of Hybrid Control systems"
} | null | null | [
"Computer Science",
"Mathematics"
]
| null | true | null | 2608 | null | Validated | null | null |
null | {
"abstract": " This paper proposes a privacy-preserving distributed recommendation\nframework, Secure Distributed Collaborative Filtering (SDCF), to preserve the\nprivacy of value, model and existence altogether. That says, not only the\nratings from the users to the items, but also the existence of the ratings as\nwell as the learned recommendation model are kept private in our framework. Our\nsolution relies on a distributed client-server architecture and a two-stage\nRandomized Response algorithm, along with an implementation on the popular\nrecommendation model, Matrix Factorization (MF). We further prove SDCF to meet\nthe guarantee of Differential Privacy so that clients are allowed to specify\narbitrary privacy levels. Experiments conducted on numerical rating prediction\nand one-class rating action prediction exhibit that SDCF does not sacrifice too\nmuch accuracy for privacy.\n",
"title": "Towards a More Reliable Privacy-preserving Recommender System"
} | null | null | null | null | true | null | 2609 | null | Default | null | null |
null | {
"abstract": " We study methods to estimate drivers' posture in vehicles using acceleration\ndata of wearable sensor and conduct field tests. To prevent fatal accidents,\ndemands for safety management of bus and taxi are high. However, acceleration\nof vehicles is added to wearable sensor in vehicles. Therefore, we study\nmethods to estimate driving posture using acceleration data acquired from shirt\ntype wearable sensor hitoe and conduct field tests.\n",
"title": "A study of posture judgement on vehicles using wearable acceleration sensor"
} | null | null | null | null | true | null | 2610 | null | Default | null | null |
null | {
"abstract": " We propose new smoothed median and the Wilcoxon's rank sum test. As is\npointed out by Maesono et al.(2016), some nonparametric discrete tests have a\nproblem with their significance probability. Because of this problem, the\nselection of the median and the Wilcoxon's test can be biased too, however, we\nshow new smoothed tests are free from the problem. Significance probabilities\nand local asymptotic powers of the new tests are studied, and we show that they\ninherit good properties of the discrete tests.\n",
"title": "Smoothed nonparametric two-sample tests"
} | null | null | null | null | true | null | 2611 | null | Default | null | null |
null | {
"abstract": " We study the never-worse relation (NWR) for Markov decision processes with an\ninfinite-horizon reachability objective. A state q is never worse than a state\np if the maximal probability of reaching the target set of states from p is at\nmost the same value from q, regard- less of the probabilities labelling the\ntransitions. Extremal-probability states, end components, and essential states\nare all special cases of the equivalence relation induced by the NWR. Using the\nNWR, states in the same equivalence class can be collapsed. Then, actions\nleading to sub- optimal states can be removed. We show the natural decision\nproblem associated to computing the NWR is coNP-complete. Finally, we ex- tend\na previously known incomplete polynomial-time iterative algorithm to\nunder-approximate the NWR.\n",
"title": "The Complexity of Graph-Based Reductions for Reachability in Markov Decision Processes"
} | null | null | null | null | true | null | 2612 | null | Default | null | null |
null | {
"abstract": " In a network, a tunnel is a part of a path where a protocol is encapsulated\nin another one. A tunnel starts with an encapsulation and ends with the\ncorresponding decapsulation. Several tunnels can be nested at some stage,\nforming a protocol stack. Tunneling is very important nowadays and it is\ninvolved in several tasks: IPv4/IPv6 transition, VPNs, security (IPsec, onion\nrouting), etc. However, tunnel establishment is mainly performed manually or by\nscript, which present obvious scalability issues. Some works attempt to\nautomate a part of the process (e.g., TSP, ISATAP, etc.). However, the\ndetermination of the tunnel(s) endpoints is not fully automated, especially in\nthe case of an arbitrary number of nested tunnels. The lack of routing\nprotocols performing automatic tunneling is due to the unavailability of path\ncomputation algorithms taking into account encapsulations and decapsulations.\nThere is a polynomial centralized algorithm to perform the task. However, to\nthe best of our knowledge, no fully distributed path computation algorithm is\nknown. Here, we propose the first fully distributed algorithm for path\ncomputation with automatic tunneling, i.e., taking into account encapsulation,\ndecapsulation and conversion of protocols. Our algorithm is a generalization of\nthe distributed Bellman-Ford algorithm, where the distance vector is replaced\nby a protocol stack vector. This allows to know how to route a packet with some\nprotocol stack. We prove that the messages size of our algorithm is polynomial,\neven if the shortest path can be of exponential length. We also prove that the\nalgorithm converges after a polynomial number of steps in a synchronized\nsetting. We adapt our algorithm into a proto-protocol for routing with\nautomatic tunneling and we show its efficiency through simulations.\n",
"title": "A stack-vector routing protocol for automatic tunneling"
} | null | null | null | null | true | null | 2613 | null | Default | null | null |
null | {
"abstract": " The paper summarizes the development of the LVCSR system built as a part of\nthe Pashto speech-translation system at the SCALE (Summer Camp for Applied\nLanguage Exploration) 2015 workshop on \"Speech-to-text-translation for\nlow-resource languages\". The Pashto language was chosen as a good \"proxy\"\nlow-resource language, exhibiting multiple phenomena which make the\nspeech-recognition and and speech-to-text-translation systems development hard.\nEven when the amount of data is seemingly sufficient, given the fact that the\ndata originates from multiple sources, the preliminary experiments reveal that\nthere is little to no benefit in merging (concatenating) the corpora and more\nelaborate ways of making use of all of the data must be worked out.\nThis paper concentrates only on the LVCSR part and presents a range of\ndifferent techniques that were found to be useful in order to benefit from\nmultiple different corpora\n",
"title": "Using of heterogeneous corpora for training of an ASR system"
} | null | null | null | null | true | null | 2614 | null | Default | null | null |
null | {
"abstract": " To understand narrative, humans draw inferences about the underlying\nrelations between narrative events. Cognitive theories of narrative\nunderstanding define these inferences as four different types of causality,\nthat include pairs of events A, B where A physically causes B (X drop, X\nbreak), to pairs of events where A causes emotional state B (Y saw X, Y felt\nfear). Previous work on learning narrative relations from text has either\nfocused on \"strict\" physical causality, or has been vague about what relation\nis being learned. This paper learns pairs of causal events from a corpus of\nfilm scene descriptions which are action rich and tend to be told in\nchronological order. We show that event pairs induced using our methods are of\nhigh quality and are judged to have a stronger causal relation than event pairs\nfrom Rel-grams.\n",
"title": "Inferring Narrative Causality between Event Pairs in Films"
} | null | null | null | null | true | null | 2615 | null | Default | null | null |
null | {
"abstract": " We define the notion of hom-Batalin-Vilkovisky algebras and strong\ndifferential hom-Gerstenhaber algebras as a special class of hom-Gerstenhaber\nalgebras and provide canonical examples associated to some well-known\nhom-structures. Representations of a hom-Lie algebroid on a hom-bundle are\ndefined and a cohomology of a regular hom-Lie algebroid with coefficients in a\nrepresentation is studied. We discuss about relationship between these classes\nof hom-Gerstenhaber algebras and geometric structures on a vector bundle. As an\napplication, we associate a homology to a regular hom-Lie algebroid and then\ndefine a hom-Poisson homology associated to a hom-Poisson manifold.\n",
"title": "On Hom-Gerstenhaber algebras and Hom-Lie algebroids"
} | null | null | [
"Mathematics"
]
| null | true | null | 2616 | null | Validated | null | null |
null | {
"abstract": " The paper should be viewed as complement of an earlier result in [8]. In the\npaper just mentioned it is shown that 1d case of a quasilinear\nparabolic-elliptic Keller-Segel system is very special. Namely, unlike in\nhigher dimensions, there is no critical nonlinearity. Indeed, for the nonlinear\ndiffusion of the form 1/u all the solutions, independently on the magnitude of\ninitial mass, stay bounded. However, the argument presented in [8] deals with\nthe Jager-Luckhaus type system. And is very sensitive to this restriction.\nNamely, the change of variables introduced in [8], being a main step of the\nmethod, works only for the Jager-Luckhaus modification. It does not seem to be\napplicable in the usual version of the parabolic-elliptic Keller-Segel system.\nThe present paper fulfils this gap and deals with the case of the usual\nparabolic-elliptic version. To handle it we establish a new Lyapunov-like\nfunctional (it is related to what was done in [8]), which leads to global\nexistence of the initial-boundary value problem for any initial mass.\n",
"title": "Global existence in the 1D quasilinear parabolic-elliptic chemotaxis system with critical nonlinearity"
} | null | null | [
"Mathematics"
]
| null | true | null | 2617 | null | Validated | null | null |
null | {
"abstract": " We establish four supercongruences between truncated ${}_3F_2$ hypergeometric\nseries involving $p$-adic Gamma functions, which extend some of the\nRodriguez-Villegas supercongruences.\n",
"title": "Supercongruences between truncated ${}_3F_2$ hypergeometric series"
} | null | null | null | null | true | null | 2618 | null | Default | null | null |
null | {
"abstract": " A multiple classifiers fusion localization technique using received signal\nstrengths (RSSs) of visible light is proposed, in which the proposed system\ntransmits different intensity modulated sinusoidal signals by LEDs and the\nsignals received by a Photo Diode (PD) placed at various grid points. First, we\nobtain some {\\emph{approximate}} received signal strengths (RSSs) fingerprints\nby capturing the peaks of power spectral density (PSD) of the received signals\nat each given grid point. Unlike the existing RSSs based algorithms, several\nrepresentative machine learning approaches are adopted to train multiple\nclassifiers based on these RSSs fingerprints. The multiple classifiers\nlocalization estimators outperform the classical RSS-based LED localization\napproaches in accuracy and robustness. To further improve the localization\nperformance, two robust fusion localization algorithms, namely, grid\nindependent least square (GI-LS) and grid dependent least square (GD-LS), are\nproposed to combine the outputs of these classifiers. We also use a singular\nvalue decomposition (SVD) based LS (LS-SVD) method to mitigate the numerical\nstability problem when the prediction matrix is singular. Experiments conducted\non intensity modulated direct detection (IM/DD) systems have demonstrated the\neffectiveness of the proposed algorithms. The experimental results show that\nthe probability of having mean square positioning error (MSPE) of less than 5cm\nachieved by GD-LS is improved by 93.03\\% and 93.15\\%, respectively, as compared\nto those by the RSS ratio (RSSR) and RSS matching methods with the FFT length\nof 2000.\n",
"title": "Indoor Localization Using Visible Light Via Fusion Of Multiple Classifiers"
} | null | null | null | null | true | null | 2619 | null | Default | null | null |
null | {
"abstract": " Embedding graph nodes into a vector space can allow the use of machine\nlearning to e.g. predict node classes, but the study of node embedding\nalgorithms is immature compared to the natural language processing field\nbecause of a diverse nature of graphs. We examine the performance of node\nembedding algorithms with respect to graph centrality measures that\ncharacterize diverse graphs, through systematic experiments with four node\nembedding algorithms, four or five graph centralities, and six datasets.\nExperimental results give insights into the properties of node embedding\nalgorithms, which can be a basis for further research on this topic.\n",
"title": "Node Centralities and Classification Performance for Characterizing Node Embedding Algorithms"
} | null | null | null | null | true | null | 2620 | null | Default | null | null |
null | {
"abstract": " We introduce a kernel Lasso (kLasso) optimization that simultaneously\naccounts for spatial regularity and network sparsity to reconstruct spatial\ncomplex networks from data. Through a kernel function, the proposed approach\nexploits spatial embedding distances to penalize overabundance of spatially\nlong-distance connections. Examples of both synthetic and real-world spatial\nnetworks show that the proposed method improves significantly upon existing\nnetwork reconstruction techniques that mainly concerns sparsity but not spatial\nregularity. Our results highlight the promise of data fusion in the\nreconstruction of complex networks, by utilizing both microscopic node-level\ndynamics (e.g., time series data) and macroscopic network-level information\n(metadata).\n",
"title": "Data Fusion Reconstruction of Spatially Embedded Complex Networks"
} | null | null | [
"Physics",
"Statistics"
]
| null | true | null | 2621 | null | Validated | null | null |
null | {
"abstract": " We consider the problem of reconstructing signals and images from periodic\nnonlinearities. For such problems, we design a measurement scheme that supports\nefficient reconstruction; moreover, our method can be adapted to extend to\ncompressive sensing-based signal and image acquisition systems. Our techniques\ncan be potentially useful for reducing the measurement complexity of high\ndynamic range (HDR) imaging systems, with little loss in reconstruction\nquality. Several numerical experiments on real data demonstrate the\neffectiveness of our approach.\n",
"title": "Reconstruction from Periodic Nonlinearities, With Applications to HDR Imaging"
} | null | null | [
"Statistics"
]
| null | true | null | 2622 | null | Validated | null | null |
null | {
"abstract": " We identify multirole logic as a new form of logic in which\nconjunction/disjunction is interpreted as an ultrafilter on the power set of\nsome underlying set (of roles) and the notion of negation is generalized to\nendomorphisms on this underlying set. We formalize both multirole logic (MRL)\nand linear multirole logic (LMRL) as natural generalizations of classical logic\n(CL) and classical linear logic (CLL), respectively, and also present a\nfilter-based interpretation for intuitionism in multirole logic. Among various\nmeta-properties established for MRL and LMRL, we obtain one named multiparty\ncut-elimination stating that every cut involving one or more sequents (as a\ngeneralization of a (binary) cut involving exactly two sequents) can be\neliminated, thus extending the celebrated result of cut-elimination by Gentzen.\n",
"title": "Multirole Logic (Extended Abstract)"
} | null | null | null | null | true | null | 2623 | null | Default | null | null |
null | {
"abstract": " In this work we present the novel ASTRID method for investigating which\nattribute interactions classifiers exploit when making predictions. Attribute\ninteractions in classification tasks mean that two or more attributes together\nprovide stronger evidence for a particular class label. Knowledge of such\ninteractions makes models more interpretable by revealing associations between\nattributes. This has applications, e.g., in pharmacovigilance to identify\ninteractions between drugs or in bioinformatics to investigate associations\nbetween single nucleotide polymorphisms. We also show how the found attribute\npartitioning is related to a factorisation of the data generating distribution\nand empirically demonstrate the utility of the proposed method.\n",
"title": "Interpreting Classifiers through Attribute Interactions in Datasets"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2624 | null | Validated | null | null |
null | {
"abstract": " In this paper, we propose a modified Levy jump diffusion model with market\nsentiment memory for stock prices, where the market sentiment comes from data\nmining implementation using Tweets on Twitter. We take the market sentiment\nprocess, which has memory, as the signal of Levy jumps in the stock price. An\nonline learning and optimization algorithm with the Unscented Kalman filter\n(UKF) is then proposed to learn the memory and to predict possible price jumps.\nExperiments show that the algorithm provides a relatively good performance in\nidentifying asset return trends.\n",
"title": "A Modified Levy Jump-Diffusion Model Based on Market Sentiment Memory for Online Jump Prediction"
} | null | null | null | null | true | null | 2625 | null | Default | null | null |
null | {
"abstract": " We present a test to quantify how well some approximate methods, designed to\nreproduce the mildly non-linear evolution of perturbations, are able to\nreproduce the clustering of DM halos once the grouping of particles into halos\nis defined and kept fixed. The following methods have been considered:\nLagrangian Perturbation Theory (LPT) up to third order, Truncated LPT,\nAugmented LPT, MUSCLE and COLA. The test runs as follows: halos are defined by\napplying a friends-of-friends (FoF) halo finder to the output of an N-body\nsimulation. The approximate methods are then applied to the same initial\nconditions of the simulation, producing for all particles displacements from\ntheir starting position and velocities. The position and velocity of each halo\nare computed by averaging over the particles that belong to that halo,\naccording to the FoF halo finder. This procedure allows us to perform a\nwell-posed test of how clustering of the matter density and halo density fields\nare recovered, without asking to the approximate method an accurate\nreconstruction of halos. We have considered the results at $z=0,0.5,1$, and we\nhave analysed power spectrum in real and redshift space, object-by-object\ndifference in position and velocity, density Probability Distribution Function\n(PDF) and its moments, phase difference of Fourier modes. We find that higher\nLPT orders are generally able to better reproduce the clustering of halos,\nwhile little or no improvement is found for the matter density field when going\nto 2LPT and 3LPT. Augmentation provides some improvement when coupled with\n2LPT, while its effect is limited when coupled with 3LPT. Little improvement is\nbrought by MUSCLE with respect to Augmentation. The more expensive\nparticle-mesh code COLA outperforms all LPT methods [abridged]\n",
"title": "Testing approximate predictions of displacements of cosmological dark matter halos"
} | null | null | [
"Physics"
]
| null | true | null | 2626 | null | Validated | null | null |
null | {
"abstract": " Advances in Wireless Sensor Network (WSN) have provided the availability of\nsmall and low-cost sensors with the capability of sensing various types of\nphysical and environmental conditions, data processing, and wireless\ncommunication. Since WSN protocols are application specific, the focus has been\ngiven to the routing protocols that might differ depending on the application\nand network architecture. In this work, novel routing protocols have been\nproposed which is a cluster-based security protocol is named as Efficient and\nSecure Routing Protocol (ESRP) for WSN. The goal of ESRP is to provide an\nenergy efficient routing solution with dynamic security features for clustered\nWSN. During the network formation, a node which is connected to a Personal\nComputer (PC) has been selected as a sink node. Once the sensor nodes were\ndeployed, the sink node logically segregates the other nodes in a cluster\nstructure and subsequently creates a WSN. This centralized cluster formation\nmethod is used to reduce the node level processing burden and avoid multiple\ncommunications. In order to ensure reliable data delivery, various security\nfeatures have been incorporated in the proposed protocol such as Modified\nZero-Knowledge Protocol (MZKP), Promiscuous hearing method, Trapping of\nadversaries and Mine detection. One of the unique features of this ESRP is that\nit can dynamically decide about the selection of these security methods, based\non the residual energy of nodes.\n",
"title": "Efficient and Secure Routing Protocol for WSN-A Thesis"
} | null | null | null | null | true | null | 2627 | null | Default | null | null |
null | {
"abstract": " Samples with a common mean but possibly different, ordered variances arise in\nvarious fields such as interlaboratory experiments, field studies or the\nanalysis of sensor data. Estimators for the common mean under ordered variances\ntypically employ random weights, which depend on the sample means and the\nunbiased variance estimators. They take different forms when the sample\nestimators are in agreement with the order constraints or not, which\ncomplicates even basic analyses such as estimating their variance. We propose\nto use the jackknife, whose consistency is established for general smooth\ntwo--sample statistics induced by continuously Gâteux or Fréchet\ndifferentiable functionals, and, more generally, asymptotically linear\ntwo--sample statistics, allowing us to study a large class of common mean\nestimators. Further, it is shown that the common mean estimators under\nconsideration satisfy a central limit theorem (CLT). We investigate the\naccuracy of the resulting confidence intervals by simulations and illustrate\nthe approach by analyzing several data sets.\n",
"title": "Jackknife variance estimation for common mean estimators under ordered variances and general two-sample statistics"
} | null | null | null | null | true | null | 2628 | null | Default | null | null |
null | {
"abstract": " We report the discovery and constrain the physical conditions of the\ninterstellar medium of the highest-redshift millimeter-selected dusty\nstar-forming galaxy (DSFG) to date, SPT-S J031132-5823.4 (hereafter\nSPT0311-58), at $z=6.900 +/- 0.002$. SPT0311-58 was discovered via its 1.4mm\nthermal dust continuum emission in the South Pole Telescope (SPT)-SZ survey.\nThe spectroscopic redshift was determined through an ALMA 3mm frequency scan\nthat detected CO(6-5), CO(7-6) and [CI](2-1), and subsequently confirmed by\ndetections of CO(3-2) with ATCA and [CII] with APEX. We constrain the\nproperties of the ISM in SPT0311-58 with a radiative transfer analysis of the\ndust continuum photometry and the CO and [CI] line emission. This allows us to\ndetermine the gas content without ad hoc assumptions about gas mass scaling\nfactors. SPT0311-58 is extremely massive, with an intrinsic gas mass of $M_{\\rm\ngas} = 3.3 \\pm 1.9 \\times10^{11}\\,M_{\\odot}$. Its large mass and intense star\nformation is very rare for a source well into the Epoch of Reionization.\n",
"title": "ISM properties of a Massive Dusty Star-Forming Galaxy discovered at z ~ 7"
} | null | null | null | null | true | null | 2629 | null | Default | null | null |
null | {
"abstract": " This article proposes a numerical scheme for computing the evolution of\nvehicular traffic on a road network over a finite time horizon. The traffic\ndynamics on each link is modeled by the Hamilton-Jacobi (HJ) partial\ndifferential equation (PDE), which is an equivalent form of the\nLighthill-Whitham-Richards PDE. The main contribution of this article is the\nconstruction of a single convex optimization program which computes the traffic\nflow at a junction over a finite time horizon and decouples the PDEs on\nconnecting links. Compared to discretization schemes which require the\ncomputation of all traffic states on a time-space grid, the proposed convex\noptimization approach computes the boundary flows at the junction using only\nthe initial condition on links and the boundary conditions of the network. The\ncomputed boundary flows at the junction specify the boundary condition for the\nHJ PDE on connecting links, which then can be separately solved using an\nexisting semi-explicit scheme for single link HJ PDE. As demonstrated in a\nnumerical example of ramp metering control, the proposed convex optimization\napproach also provides a natural framework for optimal traffic control\napplications.\n",
"title": "A convex formulation of traffic dynamics on transportation networks"
} | null | null | null | null | true | null | 2630 | null | Default | null | null |
null | {
"abstract": " The reproducibility of scientific research has become a point of critical\nconcern. We argue that openness and transparency are critical for\nreproducibility, and we outline an ecosystem for open and transparent science\nthat has emerged within the human neuroimaging community. We discuss the range\nof open data sharing resources that have been developed for neuroimaging data,\nand the role of data standards (particularly the Brain Imaging Data Structure)\nin enabling the automated sharing, processing, and reuse of large neuroimaging\ndatasets. We outline how the open-source Python language has provided the basis\nfor a data science platform that enables reproducible data analysis and\nvisualization. We also discuss how new advances in software engineering, such\nas containerization, provide the basis for greater reproducibility in data\nanalysis. The emergence of this new ecosystem provides an example for many\nareas of science that are currently struggling with reproducibility.\n",
"title": "Computational and informatics advances for reproducible data analysis in neuroimaging"
} | null | null | null | null | true | null | 2631 | null | Default | null | null |
null | {
"abstract": " We prove that the Tate, Beilinson and Parshin conjectures are invariant under\nHomological Projective Duality (=HPD). As an application, we obtain a proof of\nthese celebrated conjectures (as well as of the strong form of the Tate\nconjecture) in the new cases of linear sections of determinantal varieties and\ncomplete intersections of quadrics. Furthermore, we extend the original\nconjectures of Tate, Beilinson and Parshin from schemes to stacks and prove\nthese extended conjectures for certain low-dimensional global orbifolds.\n",
"title": "HPD-invariance of the Tate, Beilinson and Parshin conjectures"
} | null | null | null | null | true | null | 2632 | null | Default | null | null |
null | {
"abstract": " The dueling bandits problem is an online learning framework for learning from\npairwise preference feedback, and is particularly well-suited for modeling\nsettings that elicit subjective or implicit human feedback. In this paper, we\nstudy the problem of multi-dueling bandits with dependent arms, which extends\nthe original dueling bandits setting by simultaneously dueling multiple arms as\nwell as modeling dependencies between arms. These extensions capture key\ncharacteristics found in many real-world applications, and allow for the\nopportunity to develop significantly more efficient algorithms than were\npossible in the original setting. We propose the \\selfsparring algorithm, which\nreduces the multi-dueling bandits problem to a conventional bandit setting that\ncan be solved using a stochastic bandit algorithm such as Thompson Sampling,\nand can naturally model dependencies using a Gaussian process prior. We present\na no-regret analysis for multi-dueling setting, and demonstrate the\neffectiveness of our algorithm empirically on a wide range of simulation\nsettings.\n",
"title": "Multi-dueling Bandits with Dependent Arms"
} | null | null | null | null | true | null | 2633 | null | Default | null | null |
null | {
"abstract": " The presence of dusty debris around main sequence stars denotes the existence\nof planetary systems. Such debris disks are often identified by the presence of\nexcess continuum emission at infrared and (sub-)millimetre wavelengths, with\nmeasurements at longer wavelengths tracing larger and cooler dust grains. The\nexponent of the slope of the disk emission at sub-millimetre wavelengths, `q',\ndefines the size distribution of dust grains in the disk. This size\ndistribution is a function of the rigid strength of the dust producing parent\nplanetesimals. As part of the survey `PLAnetesimals around TYpical Pre-main\nseqUence Stars' (PLATYPUS) we observed six debris disks at 9-mm using the\nAustralian Telescope Compact Array. We obtain marginal (~3-\\sigma) detections\nof three targets: HD 105, HD 61005, and HD 131835. Upper limits for the three\nremaining disks, HD20807, HD109573, and HD109085, provide further constraint of\nthe (sub-)millimetre slope of their spectral energy distributions. The values\nof q (or their limits) derived from our observations are all smaller than the\noft-assumed steady state collisional cascade model (q = 3.5), but lie well\nwithin the theoretically expected range for debris disks q ~ 3 to 4. The\nmeasured q values for our targets are all < 3.3, consistent with both\ncollisional modelling results and theoretical predictions for parent\nplanetesimal bodies being `rubble piles' held together loosely by their\nself-gravity.\n",
"title": "New constraints on the millimetre emission of six debris disks"
} | null | null | null | null | true | null | 2634 | null | Default | null | null |
null | {
"abstract": " Based on a quasi-one-dimensional limit of quantum Hall states on a thin\ntorus, we construct a model of interaction-induced topological pumping which\nmimics the Hall response of the bosonic integer quantum Hall (BIQH) state. The\nquasi-one-dimensional counterpart of the BIQH state is identified as the\nHaldane phase composed of two-component bosons which form effective spin-$1$\ndegrees of freedom. An adiabatic change between the Haldane phase and trivial\nMott insulators constitute {\\it off-diagonal} topological pumping in which the\ntranslation of the lattice potential for one component induces a current in the\nother. The mechanism of this pumping is interpreted in terms of changes in\npolarizations between symmetry-protected quantized values.\n",
"title": "Bosonic integer quantum Hall effect as topological pumping"
} | null | null | [
"Physics"
]
| null | true | null | 2635 | null | Validated | null | null |
null | {
"abstract": " With onboard operating systems becoming increasingly common in vehicles, the\nreal-time broadband infotainment and Intelligent Transportation System (ITS)\nservice applications in fast-motion vehicles become ever demanding, which are\nhighly expected to significantly improve the efficiency and safety of our daily\non-road lives. The emerging ITS and vehicular applications, e.g., trip\nplanning, however, require substantial efforts on the real-time pervasive\ninformation collection and big data processing so as to provide quick decision\nmaking and feedbacks to the fast moving vehicles, which thus impose the\nsignificant challenges on the development of an efficient vehicular\ncommunication platform. In this article, we present TrasoNET, an integrated\nnetwork framework to provide realtime intelligent transportation services to\nconnected vehicles by exploring the data analytics and networking techniques.\nTrasoNET is built upon two key components. The first one guides vehicles to the\nappropriate access networks by exploring the information of realtime traffic\nstatus, specific user preferences, service applications and network conditions.\nThe second component mainly involves a distributed automatic access engine,\nwhich enables individual vehicles to make distributed access decisions based on\naccess recommender, local observation and historic information. We showcase the\napplication of TrasoNET in a case study on real-time traffic sensing based on\nreal traces of taxis.\n",
"title": "Connected Vehicular Transportation: Data Analytics and Traffic-dependent Networking"
} | null | null | null | null | true | null | 2636 | null | Default | null | null |
null | {
"abstract": " We obtain a spectral gap characterization of strongly ergodic equivalence\nrelations on standard measure spaces. We use our spectral gap criterion to\nprove that a large class of skew-product equivalence relations arising from\nmeasurable $1$-cocycles with values into locally compact abelian groups are\nstrongly ergodic. By analogy with the work of Connes on full factors, we\nintroduce the Sd and $\\tau$ invariants for type ${\\rm III}$ strongly ergodic\nequivalence relations. As a corollary to our main results, we show that for any\ntype ${\\rm III_1}$ ergodic equivalence relation $\\mathcal R$, the Maharam\nextension $\\mathord{\\text {c}}(\\mathcal R)$ is strongly ergodic if and only if\n$\\mathcal R$ is strongly ergodic and the invariant $\\tau(\\mathcal R)$ is the\nusual topology on $\\mathbf R$. We also obtain a structure theorem for almost\nperiodic strongly ergodic equivalence relations analogous to Connes' structure\ntheorem for almost periodic full factors. Finally, we prove that for arbitrary\nstrongly ergodic free actions of bi-exact groups (e.g. hyperbolic groups), the\nSd and $\\tau$ invariants of the orbit equivalence relation and of the\nassociated group measure space von Neumann factor coincide.\n",
"title": "Strongly ergodic equivalence relations: spectral gap and type III invariants"
} | null | null | null | null | true | null | 2637 | null | Default | null | null |
null | {
"abstract": " Dynamic neural network toolkits such as PyTorch, DyNet, and Chainer offer\nmore flexibility for implementing models that cope with data of varying\ndimensions and structure, relative to toolkits that operate on statically\ndeclared computations (e.g., TensorFlow, CNTK, and Theano). However, existing\ntoolkits - both static and dynamic - require that the developer organize the\ncomputations into the batches necessary for exploiting high-performance\nalgorithms and hardware. This batching task is generally difficult, but it\nbecomes a major hurdle as architectures become complex. In this paper, we\npresent an algorithm, and its implementation in the DyNet toolkit, for\nautomatically batching operations. Developers simply write minibatch\ncomputations as aggregations of single instance computations, and the batching\nalgorithm seamlessly executes them, on the fly, using computationally efficient\nbatched operations. On a variety of tasks, we obtain throughput similar to that\nobtained with manual batches, as well as comparable speedups over\nsingle-instance learning on architectures that are impractical to batch\nmanually.\n",
"title": "On-the-fly Operation Batching in Dynamic Computation Graphs"
} | null | null | [
"Computer Science",
"Statistics"
]
| null | true | null | 2638 | null | Validated | null | null |
null | {
"abstract": " Clustering is the process of finding and analyzing underlying group structure\nin data. In recent years, data as become increasingly higher dimensional and,\ntherefore, an increased need has arisen for dimension reduction techniques for\nclustering. Although such techniques are firmly established in the literature\nfor multivariate data, there is a relative paucity in the area of matrix\nvariate or three way data. Furthermore, the few methods that are available all\nassume matrix variate normality, which is not always sensible if cluster\nskewness or excess kurtosis is present. Mixtures of bilinear factor analyzers\nmodels using skewed matrix variate distributions are proposed. In all, four\nsuch mixture models are presented, based on matrix variate skew-t, generalized\nhyperbolic, variance gamma and normal inverse Gaussian distributions,\nrespectively.\n",
"title": "Mixtures of Skewed Matrix Variate Bilinear Factor Analyzers"
} | null | null | null | null | true | null | 2639 | null | Default | null | null |
null | {
"abstract": " Deep learning models require extensive architecture design exploration and\nhyperparameter optimization to perform well on a given task. The exploration of\nthe model design space is often made by a human expert, and optimized using a\ncombination of grid search and search heuristics over a large space of possible\nchoices. Neural Architecture Search (NAS) is a Reinforcement Learning approach\nthat has been proposed to automate architecture design. NAS has been\nsuccessfully applied to generate Neural Networks that rival the best\nhuman-designed architectures. However, NAS requires sampling, constructing, and\ntraining hundreds to thousands of models to achieve well-performing\narchitectures. This procedure needs to be executed from scratch for each new\ntask. The application of NAS to a wide set of tasks currently lacks a way to\ntransfer generalizable knowledge across tasks. In this paper, we present the\nMultitask Neural Model Search (MNMS) controller. Our goal is to learn a\ngeneralizable framework that can condition model construction on successful\nmodel searches for previously seen tasks, thus significantly speeding up the\nsearch for new tasks. We demonstrate that MNMS can conduct an automated\narchitecture search for multiple tasks simultaneously while still learning\nwell-performing, specialized models for each task. We then show that\npre-trained MNMS controllers can transfer learning to new tasks. By leveraging\nknowledge from previous searches, we find that pre-trained MNMS models start\nfrom a better location in the search space and reduce search time on unseen\ntasks, while still discovering models that outperform published human-designed\nmodels.\n",
"title": "Transfer Learning to Learn with Multitask Neural Model Search"
} | null | null | null | null | true | null | 2640 | null | Default | null | null |
null | {
"abstract": " The actions of an autonomous vehicle on the road affect and are affected by\nthose of other drivers, whether overtaking, negotiating a merge, or avoiding an\naccident. This mutual dependence, best captured by dynamic game theory, creates\na strong coupling between the vehicle's planning and its predictions of other\ndrivers' behavior, and constitutes an open problem with direct implications on\nthe safety and viability of autonomous driving technology. Unfortunately,\ndynamic games are too computationally demanding to meet the real-time\nconstraints of autonomous driving in its continuous state and action space. In\nthis paper, we introduce a novel game-theoretic trajectory planning algorithm\nfor autonomous driving, that enables real-time performance by hierarchically\ndecomposing the underlying dynamic game into a long-horizon \"strategic\" game\nwith simplified dynamics and full information structure, and a short-horizon\n\"tactical\" game with full dynamics and a simplified information structure. The\nvalue of the strategic game is used to guide the tactical planning, implicitly\nextending the planning horizon, pushing the local trajectory optimization\ncloser to global solutions, and, most importantly, quantitatively accounting\nfor the autonomous vehicle and the human driver's ability and incentives to\ninfluence each other. In addition, our approach admits non-deterministic models\nof human decision-making, rather than relying on perfectly rational\npredictions. Our results showcase richer, safer, and more effective autonomous\nbehavior in comparison to existing techniques.\n",
"title": "Hierarchical Game-Theoretic Planning for Autonomous Vehicles"
} | null | null | [
"Computer Science"
]
| null | true | null | 2641 | null | Validated | null | null |
null | {
"abstract": " This paper introduces a method for efficiently inferring a high-dimensional\ndistributed quantity from a few observations. The quantity of interest (QoI) is\napproximated in a basis (dictionary) learned from a training set. The\ncoefficients associated with the approximation of the QoI in the basis are\ndetermined by minimizing the misfit with the observations. To obtain a\nprobabilistic estimate of the quantity of interest, a Bayesian approach is\nemployed. The QoI is treated as a random field endowed with a hierarchical\nprior distribution so that closed-form expressions can be obtained for the\nposterior distribution. The main contribution of the present work lies in the\nderivation of \\emph{a representation basis consistent with the observation\nchain} used to infer the associated coefficients. The resulting dictionary is\nthen tailored to be both observable by the sensors and accurate in\napproximating the posterior mean. An algorithm for deriving such an observable\ndictionary is presented. The method is illustrated with the estimation of the\nvelocity field of an open cavity flow from a handful of wall-mounted point\nsensors. Comparison with standard estimation approaches relying on Principal\nComponent Analysis and K-SVD dictionaries is provided and illustrates the\nsuperior performance of the present approach.\n",
"title": "Observable dictionary learning for high-dimensional statistical inference"
} | null | null | null | null | true | null | 2642 | null | Default | null | null |
null | {
"abstract": " Understanding the structure of ZnO surface reconstructions and their\nresultant properties is crucial to the rational design of ZnO-containing\ndevices ranging from optoelectronics to catalysts. Here, we are motivated by\nrecent experimental work which showed a new surface reconstruction containing\nZn vacancies ordered in a Zn(3x3) pattern in the subsurface of (0001)-O\nterminated ZnO. A reconstruction with Zn vacancies on (0001)-O is surprising\nand counterintuitive because Zn vacancies enhance the surface dipole rather\nthan reduce it. In this work, we show using Density Functional Theory (DFT)\nthat subsurface Zn vacancies can form on (0001)-O when coupled with adsorption\nof surface H and are in fact stable under a wide range of common conditions. We\nalso show these vacancies have a significant ordering tendency and that\nSb-doping created subsurface inversion domain boundaries (IDBs) enhances the\ndriving force of Zn vacancy alignment into large domains of the Zn(3x3)\nreconstruction.\n",
"title": "Counterintuitive Reconstruction of the Polar O-Terminated ZnO Surface With Zinc Vacancies and Hydrogen"
} | null | null | null | null | true | null | 2643 | null | Default | null | null |
null | {
"abstract": " In this paper, we consider the problem of determining when two tensor\nnetworks are equivalent under a heterogeneous change of basis. In particular,\nto a string diagram in a certain monoidal category (which we call tensor\ndiagrams), we formulate an associated abelian category of representations. Each\nrepresentation corresponds to a tensor network on that diagram. We then\nclassify which tensor diagrams give rise to categories that are finite, tame,\nor wild in the traditional sense of representation theory. For those tensor\ndiagrams of finite and tame type, we classify the indecomposable\nrepresentations. Our main result is that a tensor diagram is wild if and only\nif it contains a vertex of degree at least three. Otherwise, it is of tame or\nfinite type.\n",
"title": "A Finite-Tame-Wild Trichotomy Theorem for Tensor Diagrams"
} | null | null | null | null | true | null | 2644 | null | Default | null | null |
null | {
"abstract": " The quantile ratio index introduced by Prendergast and Staudte 2017 is a\nsimple and effective measure of relative inequality for income data that is\nresistant to outliers. It measures the average relative distance of a randomly\nchosen income from its symmetric quantile. Another useful property of this\nindex is investigated here: given a partition of the income distribution into a\nunion of sets of symmetric quantiles, one can find the conditional inequality\nfor each set as measured by the quantile ratio index and readily combine them\nin a weighted average to obtain the index for the entire population. When\napplied to data for various years, one can track how these contributions to\ninequality vary over time, as illustrated here for Australian Bureau of\nStatistics income and wealth data.\n",
"title": "Decomposing the Quantile Ratio Index with applications to Australian income and wealth data"
} | null | null | null | null | true | null | 2645 | null | Default | null | null |
null | {
"abstract": " This paper considers a practical scenario where a classical estimation method\nmight have already been implemented on a certain platform when one tries to\napply more advanced techniques such as moving horizon estimation (MHE). We are\ninterested to utilize MHE to upgrade, rather than completely discard, the\nexisting estimation technique. This immediately raises the question how one can\nimprove the estimation performance gradually based on the pre-estimator. To\nthis end, we propose a general methodology which incorporates the pre-estimator\nwith a tuning parameter {\\lambda} between 0 and 1 into the quadratic cost\nfunctions that are usually adopted in MHE. We examine the above idea in two\nstandard MHE frameworks that have been proposed in the existing literature. For\nboth frameworks, when {\\lambda} = 0, the proposed strategy exactly matches the\nexisting classical estimator; when the value of {\\lambda} is increased, the\nproposed strategy exhibits a more aggressive normalized forgetting effect\ntowards the old data, thereby increasing the estimation performance gradually.\n",
"title": "Metamorphic Moving Horizon Estimation"
} | null | null | null | null | true | null | 2646 | null | Default | null | null |
null | {
"abstract": " The persistence diagram of Cohen-Steiner, Edelsbrunner, and Harer was\nrecently generalized by Patel to the case of constructible persistence modules\nwith values in a symmetric monoidal category with images. Patel also introduced\na distance for persistence diagrams, the erosion distance. Motivated by this\nwork, we extend the erosion distance to a distance of rank invariants of\ngeneralized persistence modules by using the generalization of the interleaving\ndistance of Bubenik, de Silva, and Scott as a guideline. This extension of the\nerosion distance also gives, as a special case, a distance for multidimensional\npersistent homology groups with torsion introduced by Frosini. We show that the\nerosion distance is stable with respect to the interleaving distance, and that\nit gives a lower bound for the natural pseudo-distance in the case of sublevel\nset persistent homology of continuous functions.\n",
"title": "Erosion distance for generalized persistence modules"
} | null | null | [
"Mathematics"
]
| null | true | null | 2647 | null | Validated | null | null |
null | {
"abstract": " First-order optimization algorithms, often preferred for large problems,\nrequire the gradient of the differentiable terms in the objective function.\nThese gradients often involve linear operators and their adjoints, which must\nbe applied rapidly. We consider two example problems and derive methods for\nquickly evaluating the required adjoint operator. The first example is an image\ndeblurring problem, where we must compute efficiently the adjoint of\nmulti-stage wavelet reconstruction. Our formulation of the adjoint works for a\nvariety of boundary conditions, which allows the formulation to generalize to a\nlarger class of problems. The second example is a blind channel estimation\nproblem taken from the optimization literature where we must compute the\nadjoint of the convolution of two signals. In each example, we show how the\nadjoint operator can be applied efficiently while leveraging existing software.\n",
"title": "Efficient Adjoint Computation for Wavelet and Convolution Operators"
} | null | null | null | null | true | null | 2648 | null | Default | null | null |
null | {
"abstract": " Datacenter-based Cloud Computing services provide a flexible, scalable and\nyet economical infrastructure to host online services such as multimedia\nstreaming, email and bulk storage. Many such services perform geo-replication\nto provide necessary quality of service and reliability to users resulting in\nfrequent large inter- datacenter transfers. In order to meet tenant service\nlevel agreements (SLAs), these transfers have to be completed prior to a\ndeadline. In addition, WAN resources are quite scarce and costly, meaning they\nshould be fully utilized. Several recently proposed schemes, such as B4,\nTEMPUS, and SWAN have focused on improving the utilization of inter-datacenter\ntransfers through centralized scheduling, however, they fail to provide a\nmechanism to guarantee that admitted requests meet their deadlines. Also, in a\nrecent study, authors propose Amoeba, a system that allows tenants to define\ndeadlines and guarantees that the specified deadlines are met, however, to\nadmit new traffic, the proposed system has to modify the allocation of already\nadmitted transfers. In this paper, we propose Rapid Close to Deadline\nScheduling (RCD), a close to deadline traffic allocation technique that is fast\nand efficient. Through simulations, we show that RCD is up to 15 times faster\nthan Amoeba, provides high link utilization along with deadline guarantees, and\nis able to make quick decisions on whether a new request can be fully satisfied\nbefore its deadline.\n",
"title": "RCD: Rapid Close to Deadline Scheduling for Datacenter Networks"
} | null | null | null | null | true | null | 2649 | null | Default | null | null |
null | {
"abstract": " We prove that when $q$ is a power of $2$, every complex irreducible\nrepresentation of $\\mathrm{Sp}(2n, \\mathbb{F}_q)$ may be defined over the real\nnumbers, that is, all Frobenius-Schur indicators are 1. We also obtain a\ngenerating function for the sum of the degrees of the unipotent characters of\n$\\mathrm{Sp}(2n, \\mathbb{F}_q)$, or of $\\mathrm{SO}(2n+1, \\mathbb{F}_q)$, for\nany prime power $q$.\n",
"title": "Real representations of finite symplectic groups over fields of characteristic two"
} | null | null | null | null | true | null | 2650 | null | Default | null | null |
null | {
"abstract": " In this paper, we discuss the application of extreme value theory in the\ncontext of stationary $\\beta$-mixing sequences that belong to the Fréchet\ndomain of attraction. In particular, we propose a methodology to construct\nbias-corrected tail estimators. Our approach is based on the combination of two\nestimators for the extreme value index to cancel the bias. The resulting\nestimator is used to estimate an extreme quantile. In a simulation study, we\noutline the performance of our proposals that we compare to alternative\nestimators recently introduced in the literature. Also, we compute the\nasymptotic variance in specific examples when possible. Our methodology is\napplied to two datasets on finance and environment.\n",
"title": "Risk measure estimation for $β$-mixing time series and applications"
} | null | null | null | null | true | null | 2651 | null | Default | null | null |
null | {
"abstract": " With the help of transfer entropy, we analyze information flows between\ncommunities of complex networks. We show that the transfer entropy provides a\ncoherent description of interactions between communities, including non-linear\ninteractions. To put some flesh on the bare bones, we analyze transfer\nentropies between communities of five largest financial markets, represented as\nnetworks of interacting stocks. Additionally, we discuss information transfer\nof rare events, which is analyzed by Rényi transfer entropy.\n",
"title": "Transfer entropy between communities in complex networks"
} | null | null | null | null | true | null | 2652 | null | Default | null | null |
null | {
"abstract": " Many problems in machine learning and related application areas are\nfundamentally variants of conditional modeling and sampling across multi-aspect\ndata, either multi-view, multi-modal, or simply multi-group. For example,\nsampling from the distribution of English sentences conditioned on a given\nFrench sentence or sampling audio waveforms conditioned on a given piece of\ntext. Central to many of these problems is the issue of missing data: we can\nobserve many English, French, or German sentences individually but only\noccasionally do we have data for a sentence pair. Motivated by these\napplications and inspired by recent progress in variational autoencoders for\ngrouped data, we develop factVAE, a deep generative model capable of handling\nmulti-aspect data, robust to missing observations, and with a prior that\nencourages disentanglement between the groups and the latent dimensions. The\neffectiveness of factVAE is demonstrated on a variety of rich real-world\ndatasets, including motion capture poses and pictures of faces captured from\nvarying poses and perspectives.\n",
"title": "Disentangled VAE Representations for Multi-Aspect and Missing Data"
} | null | null | null | null | true | null | 2653 | null | Default | null | null |
null | {
"abstract": " In the previous article we derived a detailed asymptotic expansion of the\nheat trace for the Laplace-Beltrami operator on functions on manifolds with\nconic singularities. In this article we investigate how the terms in the\nexpansion reflect the geometry of the manifold. Since the general expansion\ncontains a logarithmic term, its vanishing is a necessary condition for\nsmoothness of the manifold. In the two-dimensional case this implies that the\nconstant term of the expansion contains a non-local term that determines the\nlength of the (circular) cross section and vanishes precisely if this length\nequals $2\\pi$, that is, in the smooth case. We proceed to the study of higher\ndimensions. In the four-dimensional case, the logarithmic term in the expansion\nvanishes precisely when the cross section is a spherical space form, and we\nexpect that the vanishing of a further singular term will imply again\nsmoothness, but this is not yet clear beyond the case of cyclic space forms. In\nhigher dimensions the situation is naturally more difficult. We illustrate this\nin the case of cross sections with constant curvature. Then the logarithmic\nterm becomes a polynomial in the curvature with roots that are different from\n1, which necessitates more vanishing of other terms, not isolated so far.\n",
"title": "On the spectral geometry of manifolds with conic singularities"
} | null | null | null | null | true | null | 2654 | null | Default | null | null |
null | {
"abstract": " In this work, we propose a novel robot learning framework called Neural Task\nProgramming (NTP), which bridges the idea of few-shot learning from\ndemonstration and neural program induction. NTP takes as input a task\nspecification (e.g., video demonstration of a task) and recursively decomposes\nit into finer sub-task specifications. These specifications are fed to a\nhierarchical neural program, where bottom-level programs are callable\nsubroutines that interact with the environment. We validate our method in three\nrobot manipulation tasks. NTP achieves strong generalization across sequential\ntasks that exhibit hierarchal and compositional structures. The experimental\nresults show that NTP learns to generalize well to- wards unseen tasks with\nincreasing lengths, variable topologies, and changing objectives.\n",
"title": "Neural Task Programming: Learning to Generalize Across Hierarchical Tasks"
} | null | null | null | null | true | null | 2655 | null | Default | null | null |
null | {
"abstract": " Scientific publishing conveys the outputs of an academic or research\nactivity, in this sense; it also reflects the efforts and issues in which\npeople engage. To identify potential collaborative networks one of the simplest\napproaches is to leverage the co-authorship relations. In this approach,\nsemantic and hierarchic relationships defined by a Knowledge Organization\nSystem are used in order to improve the system's ability to recommend potential\nnetworks beyond the lexical or syntactic analysis of the topics or concepts\nthat are of interest to academics.\n",
"title": "Discovery of potential collaboration networks from open knowledge sources"
} | null | null | null | null | true | null | 2656 | null | Default | null | null |
null | {
"abstract": " We present a multi-query recovery policy for a hybrid system with goal limit\ncycle. The sample trajectories and the hybrid limit cycle of the dynamical\nsystem are stabilized using locally valid Time Varying LQR controller policies\nwhich probabilistically cover a bounded region of state space. The original LQR\nTree algorithm builds such trees for non-linear static and non-hybrid systems\nlike a pendulum or a cart-pole. We leverage the idea of LQR trees to plan with\na continuous control set, unlike methods that rely on discretization like\ndynamic programming to plan for hybrid dynamical systems where it is hard to\ncapture the exact event of discrete transition. We test the algorithm on a\ncompass gait model by stabilizing a dynamic walking hybrid limit cycle with\npoint foot contact from random initial conditions. We show results from the\nsimulation where the system comes back to a stable behavior with initial\nposition or velocity perturbation and noise.\n",
"title": "Towards Planning and Control of Hybrid Systems with Limit Cycle using LQR Trees"
} | null | null | null | null | true | null | 2657 | null | Default | null | null |
null | {
"abstract": " t-distributed Stochastic Neighborhood Embedding (t-SNE), a clustering and\nvisualization method proposed by van der Maaten & Hinton in 2008, has rapidly\nbecome a standard tool in a number of natural sciences. Despite its\noverwhelming success, there is a distinct lack of mathematical foundations and\nthe inner workings of the algorithm are not well understood. The purpose of\nthis paper is to prove that t-SNE is able to recover well-separated clusters;\nmore precisely, we prove that t-SNE in the `early exaggeration' phase, an\noptimization technique proposed by van der Maaten & Hinton (2008) and van der\nMaaten (2014), can be rigorously analyzed. As a byproduct, the proof suggests\nnovel ways for setting the exaggeration parameter $\\alpha$ and step size $h$.\nNumerical examples illustrate the effectiveness of these rules: in particular,\nthe quality of embedding of topological structures (e.g. the swiss roll)\nimproves. We also discuss a connection to spectral clustering methods.\n",
"title": "Clustering with t-SNE, provably"
} | null | null | null | null | true | null | 2658 | null | Default | null | null |
null | {
"abstract": " Winds arising from galaxies, star clusters, and active galactic nuclei are\ncrucial players in star and galaxy formation, but it has proven remarkably\ndifficult to use observations of them to determine physical properties of\ninterest, particularly mass fluxes. Much of the difficulty stems from a lack of\na theory that links a physically-realistic model for winds' density, velocity,\nand covering factors to calculations of light emission and absorption. In this\npaper we provide such a model. We consider a wind launched from a turbulent\nregion with a range of column densities, derive the differential acceleration\nof gas as a function of column density, and use this result to compute winds'\nabsorption profiles, emission profiles, and emission intensity maps in both\noptically thin and optically thick species. The model is sufficiently simple\nthat all required computations can be done analytically up to straightforward\nnumerical integrals, rendering it suitable for the problem of deriving physical\nparameters by fitting models to observed data. We show that our model produces\nrealistic absorption and emission profiles for some example cases, and argue\nthat the most promising methods of deducing mass fluxes are based on\ncombinations of absorption lines of different optical depths, or on combining\nabsorption with measurements of molecular line emission. In the second paper in\nthis series, we expand on these ideas by introducing a set of observational\ndiagnostics that are significantly more robust that those commonly in use, and\nthat can be used to obtain improved estimates of wind properties.\n",
"title": "The Observable Properties of Cool Winds from Galaxies, AGN, and Star Clusters. I. Theoretical Framework"
} | null | null | null | null | true | null | 2659 | null | Default | null | null |
null | {
"abstract": " Neural models enjoy widespread use across a variety of tasks and have grown\nto become crucial components of many industrial systems. Despite their\neffectiveness and extensive popularity, they are not without their exploitable\nflaws. Initially applied to computer vision systems, the generation of\nadversarial examples is a process in which seemingly imperceptible\nperturbations are made to an image, with the purpose of inducing a deep\nlearning based classifier to misclassify the image. Due to recent trends in\nspeech processing, this has become a noticeable issue in speech recognition\nmodels. In late 2017, an attack was shown to be quite effective against the\nSpeech Commands classification model. Limited-vocabulary speech classifiers,\nsuch as the Speech Commands model, are used quite frequently in a variety of\napplications, particularly in managing automated attendants in telephony\ncontexts. As such, adversarial examples produced by this attack could have\nreal-world consequences. While previous work in defending against these\nadversarial examples has investigated using audio preprocessing to reduce or\ndistort adversarial noise, this work explores the idea of flooding particular\nfrequency bands of an audio signal with random noise in order to detect\nadversarial examples. This technique of flooding, which does not require\nretraining or modifying the model, is inspired by work done in computer vision\nand builds on the idea that speech classifiers are relatively robust to natural\nnoise. A combined defense incorporating 5 different frequency bands for\nflooding the signal with noise outperformed other existing defenses in the\naudio space, detecting adversarial examples with 91.8% precision and 93.5%\nrecall.\n",
"title": "Noise Flooding for Detecting Audio Adversarial Examples Against Automatic Speech Recognition"
} | null | null | null | null | true | null | 2660 | null | Default | null | null |
null | {
"abstract": " In this paper, we propose a novel ranking framework for collaborative\nfiltering with the overall aim of learning user preferences over items by\nminimizing a pairwise ranking loss. We show the minimization problem involves\ndependent random variables and provide a theoretical analysis by proving the\nconsistency of the empirical risk minimization in the worst case where all\nusers choose a minimal number of positive and negative items. We further derive\na Neural-Network model that jointly learns a new representation of users and\nitems in an embedded space as well as the preference relation of users over the\npairs of items. The learning objective is based on three scenarios of ranking\nlosses that control the ability of the model to maintain the ordering over the\nitems induced from the users' preferences, as well as, the capacity of the\ndot-product defined in the learned embedded space to produce the ordering. The\nproposed model is by nature suitable for implicit feedback and involves the\nestimation of only very few parameters. Through extensive experiments on\nseveral real-world benchmarks on implicit data, we show the interest of\nlearning the preference and the embedding simultaneously when compared to\nlearning those separately. We also demonstrate that our approach is very\ncompetitive with the best state-of-the-art collaborative filtering techniques\nproposed for implicit feedback.\n",
"title": "Representation Learning and Pairwise Ranking for Implicit Feedback in Recommendation Systems"
} | null | null | null | null | true | null | 2661 | null | Default | null | null |
null | {
"abstract": " This paper introduces consensus-based primal-dual methods for distributed\nonline optimization where the time-varying system objective function\n$f_t(\\mathbf{x})$ is given as the sum of local agents' objective functions,\ni.e., $f_t(\\mathbf{x}) = \\sum_i f_{i,t}(\\mathbf{x}_i)$, and the system\nconstraint function $\\mathbf{g}(\\mathbf{x})$ is given as the sum of local\nagents' constraint functions, i.e., $\\mathbf{g}(\\mathbf{x}) = \\sum_i\n\\mathbf{g}_i (\\mathbf{x}_i) \\preceq \\mathbf{0}$. At each stage, each agent\ncommits to an adaptive decision pertaining only to the past and locally\navailable information, and incurs a new cost function reflecting the change in\nthe environment. Our algorithm uses weighted averaging of the iterates for each\nagent to keep local estimates of the global constraints and dual variables. We\nshow that the algorithm achieves a regret of order $O(\\sqrt{T})$ with the time\nhorizon $T$, in scenarios when the underlying communication topology is\ntime-varying and jointly-connected. The regret is measured in regard to the\ncost function value as well as the constraint violation. Numerical results for\nonline routing in wireless multi-hop networks with uncertain channel rates are\nprovided to illustrate the performance of the proposed algorithm.\n",
"title": "On the Sublinear Regret of Distributed Primal-Dual Algorithms for Online Constrained Optimization"
} | null | null | null | null | true | null | 2662 | null | Default | null | null |
null | {
"abstract": " The proportional odds model gives a method of generating new family of\ndistributions by adding a parameter, called tilt parameter, to expand an\nexisting family of distributions. The new family of distributions so obtained\nis known as Marshall-Olkin family of distributions or Marshall-Olkin extended\ndistributions. In this paper, we consider Marshall-Olkin family of\ndistributions in discrete case with fixed tilt parameter. We study different\nageing properties, as well as different stochastic orderings of this family of\ndistributions. All the results of this paper are supported by several examples.\n",
"title": "Reliability study of proportional odds family of discrete distributions"
} | null | null | null | null | true | null | 2663 | null | Default | null | null |
null | {
"abstract": " We describe global embeddings of fractional D3 branes at orientifolded\nsingularities in type IIB flux compactifications. We present an explicit\nCalabi-Yau example where the chiral visible sector lives on a local\norientifolded quiver while non-perturbative effects, $\\alpha'$ corrections and\na T-brane hidden sector lead to full closed string moduli stabilisation in a de\nSitter vacuum. The same model can also successfully give rise to inflation\ndriven by a del Pezzo divisor. Our model represents the first explicit\nCalabi-Yau example featuring both an inflationary and a chiral visible sector.\n",
"title": "Global Orientifolded Quivers with Inflation"
} | null | null | null | null | true | null | 2664 | null | Default | null | null |
null | {
"abstract": " Consider a nilpotent element e in a simple complex Lie algebra. The Springer\nfibre corresponding to e admits a discretization (discrete analogue) introduced\nby the author in 1999. In this paper we propose a conjectural description of\nthat discretization which is more amenable to computation.\n",
"title": "Discretization of Springer fibers"
} | null | null | null | null | true | null | 2665 | null | Default | null | null |
null | {
"abstract": " We consider the problem of proving that each point in a given set of states\n(\"target set\") can indeed be reached by a given nondeterministic\ncontinuous-time dynamical system from some initial state. We consider this\nproblem for abstract continuous-time models that can be concretized as various\nkinds of continuous and hybrid dynamical systems.\nThe approach to this problem proposed in this paper is based on finding a\nsuitable superset S of the target set which has the property that each partial\ntrajectory of the system which lies entirely in S either is defined as the\ninitial time moment, or can be locally extended backward in time, or can be\nlocally modified in such a way that the resulting trajectory can be locally\nextended back in time.\nThis reformulation of the problem has a relatively simple logical expression\nand is convenient for applying various local existence theorems and local\ndynamics analysis methods to proving reachability which makes it suitable for\nreasoning about the behavior of continuous and hybrid dynamical systems in\nproof assistants such as Mizar, Isabelle, etc.\n",
"title": "On the Underapproximation of Reach Sets of Abstract Continuous-Time Systems"
} | null | null | null | null | true | null | 2666 | null | Default | null | null |
null | {
"abstract": " The estimation of a log-concave density on $\\mathbb{R}$ is a canonical\nproblem in the area of shape-constrained nonparametric inference. We present a\nBayesian nonparametric approach to this problem based on an exponentiated\nDirichlet process mixture prior and show that the posterior distribution\nconverges to the log-concave truth at the (near-) minimax rate in Hellinger\ndistance. Our proof proceeds by establishing a general contraction result based\non the log-concave maximum likelihood estimator that prevents the need for\nfurther metric entropy calculations. We also present two computationally more\nfeasible approximations and a more practical empirical Bayes approach, which\nare illustrated numerically via simulations.\n",
"title": "A Bayesian nonparametric approach to log-concave density estimation"
} | null | null | null | null | true | null | 2667 | null | Default | null | null |
null | {
"abstract": " Metric graphs are special types of metric spaces used to model and represent\nsimple, ubiquitous, geometric relations in data such as biological networks,\nsocial networks, and road networks. We are interested in giving a qualitative\ndescription of metric graphs using topological summaries. In particular, we\nprovide a complete characterization of the 1-dimensional intrinsic Cech\npersistence diagrams for metric graphs using persistent homology. Together with\ncomplementary results by Adamaszek et. al, which imply results on intrinsic\nCech persistence diagrams in all dimensions for a single cycle, our results\nconstitute important steps toward characterizing intrinsic Cech persistence\ndiagrams for arbitrary metric graphs across all dimensions.\n",
"title": "A Complete Characterization of the 1-Dimensional Intrinsic Cech Persistence Diagrams for Metric Graphs"
} | null | null | null | null | true | null | 2668 | null | Default | null | null |
null | {
"abstract": " The critcal exponent $\\omega$ is evaluated at $O(1/N)$ in $d$-dimensions in\nthe Gross-Neveu model using the large $N$ critical point formalism. It is shown\nto be in agreement with the recently determined three loop $\\beta$-functions of\nthe Gross-Neveu-Yukawa model in four dimensions. The same exponent is computed\nfor the chiral Gross-Neveu and non-abelian Nambu-Jona-Lasinio universality\nclasses.\n",
"title": "Critical exponent $ω$ in the Gross-Neveu-Yukawa model at $O(1/N)$"
} | null | null | null | null | true | null | 2669 | null | Default | null | null |
null | {
"abstract": " This article presents a framework and develops a formulation to solve a path\nplanning problem for multiple heterogeneous Unmanned Vehicles (UVs) with\nuncertain service times for each vehicle--target pair. The vehicles incur a\npenalty proportional to the duration of their total service time in excess of a\npreset constant. The vehicles differ in their motion constraints and are\nlocated at distinct depots at the start of the mission. The vehicles may also\nbe equipped with disparate sensors. The objective is to find a tour for each\nvehicle that starts and ends at its respective depot such that every target is\nvisited and serviced by some vehicle while minimizing the sum of the total\ntravel distance and the expected penalty incurred by all the vehicles. We\nformulate the problem as a two-stage stochastic program with recourse, present\nthe theoretical properties of the formulation and advantages of using such a\nformulation, as opposed to a deterministic expected value formulation, to solve\nthe problem. Extensive numerical simulations also corroborate the effectiveness\nof the proposed approach.\n",
"title": "Path Planning for Multiple Heterogeneous Unmanned Vehicles with Uncertain Service Times"
} | null | null | null | null | true | null | 2670 | null | Default | null | null |
null | {
"abstract": " Active learning is relevant and challenging for high-dimensional regression\nmodels when the annotation of the samples is expensive. Yet most of the\nexisting sampling methods cannot be applied to large-scale problems, consuming\ntoo much time for data processing. In this paper, we propose a fast active\nlearning algorithm for regression, tailored for neural network models. It is\nbased on uncertainty estimation from stochastic dropout output of the network.\nExperiments on both synthetic and real-world datasets show comparable or better\nperformance (depending on the accuracy metric) as compared to the baselines.\nThis approach can be generalized to other deep learning architectures. It can\nbe used to systematically improve a machine-learning model as it offers a\ncomputationally efficient way of sampling additional data.\n",
"title": "Dropout-based Active Learning for Regression"
} | null | null | null | null | true | null | 2671 | null | Default | null | null |
null | {
"abstract": " Over the last few decades, psychologists have developed sophisticated formal\nmodels of human categorization using simple artificial stimuli. In this paper,\nwe use modern machine learning methods to extend this work into the realm of\nnaturalistic stimuli, enabling human categorization to be studied over the\ncomplex visual domain in which it evolved and developed. We show that\nrepresentations derived from a convolutional neural network can be used to\nmodel behavior over a database of >300,000 human natural image classifications,\nand find that a group of models based on these representations perform well,\nnear the reliability of human judgments. Interestingly, this group includes\nboth exemplar and prototype models, contrasting with the dominance of exemplar\nmodels in previous work. We are able to improve the performance of the\nremaining models by preprocessing neural network representations to more\nclosely capture human similarity judgments.\n",
"title": "Modeling Human Categorization of Natural Images Using Deep Feature Representations"
} | null | null | null | null | true | null | 2672 | null | Default | null | null |
null | {
"abstract": " Comprehensive Two dimensional gas chromatography (GCxGC) plays a central role\ninto the elucidation of complex samples. The automation of the identification\nof peak areas is of prime interest to obtain a fast and repeatable analysis of\nchromatograms. To determine the concentration of compounds or pseudo-compounds,\ntemplates of blobs are defined and superimposed on a reference chromatogram.\nThe templates then need to be modified when different chromatograms are\nrecorded. In this study, we present a chromatogram and template alignment\nmethod based on peak registration called BARCHAN. Peaks are identified using a\nrobust mathematical morphology tool. The alignment is performed by a\nprobabilistic estimation of a rigid transformation along the first dimension,\nand a non-rigid transformation in the second dimension, taking into account\nnoise, outliers and missing peaks in a fully automated way. Resulting aligned\nchromatograms and masks are presented on two datasets. The proposed algorithm\nproves to be fast and reliable. It significantly reduces the time to results\nfor GCxGC analysis.\n",
"title": "BARCHAN: Blob Alignment for Robust CHromatographic ANalysis"
} | null | null | null | null | true | null | 2673 | null | Default | null | null |
null | {
"abstract": " Panel data analysis is an important topic in statistics and econometrics.\nTraditionally, in panel data analysis, all individuals are assumed to share the\nsame unknown parameters, e.g. the same coefficients of covariates when the\nlinear models are used, and the differences between the individuals are\naccounted for by cluster effects. This kind of modelling only makes sense if\nour main interest is on the global trend, this is because it would not be able\nto tell us anything about the individual attributes which are sometimes very\nimportant. In this paper, we proposed a modelling based on the single index\nmodels embedded with homogeneity for panel data analysis, which builds the\nindividual attributes in the model and is parsimonious at the same time. We\ndevelop a data driven approach to identify the structure of homogeneity, and\nestimate the unknown parameters and functions based on the identified\nstructure. Asymptotic properties of the resulting estimators are established.\nIntensive simulation studies conducted in this paper also show the resulting\nestimators work very well when sample size is finite. Finally, the proposed\nmodelling is applied to a public financial dataset and a UK climate dataset,\nthe results reveal some interesting findings.\n",
"title": "Homogeneity Pursuit in Single Index Models based Panel Data Analysis"
} | null | null | null | null | true | null | 2674 | null | Default | null | null |
null | {
"abstract": " In order to understand the origin of observed molecular cloud properties, it\nis critical to understand how clouds interact with their environments during\ntheir formation, growth, and collapse. It has been suggested that\naccretion-driven turbulence can maintain clouds in a highly turbulent state,\npreventing runaway collapse, and explaining the observed non-thermal velocity\ndispersions. We present 3D, AMR, MHD simulations of a kiloparsec-scale,\nstratified, supernova-driven, self-gravitating, interstellar medium, including\ndiffuse heating and radiative cooling. These simulations model the formation\nand evolution of a molecular cloud population in the turbulent interstellar\nmedium. We use zoom-in techniques to focus on the dynamics of the mass\naccretion and its history for individual molecular clouds. We find that mass\naccretion onto molecular clouds proceeds as a combination of turbulent and near\nfree-fall accretion of a gravitationally bound envelope. Nearby supernova\nexplosions have a dual role, compressing the envelope, boosting accreted mass,\nbut also disrupting parts of the envelope and eroding mass from the cloud's\nsurface. It appears that the inflow rate of kinetic energy onto clouds from\nsupernova explosions is insufficient to explain the net rate of charge of the\ncloud kinetic energy. In the absence of self-consistent star formation,\nconversion of gravitational potential into kinetic energy during contraction\nseems to be the main driver of non-thermal motions within clouds. We conclude\nthat although clouds interact strongly with their environments, bound clouds\nare always in a state of gravitational contraction, close to runaway, and their\nproperties are a natural result of this collapse.\n",
"title": "Feeding vs. Falling: The growth and collapse of molecular clouds in a turbulent interstellar medium"
} | null | null | null | null | true | null | 2675 | null | Default | null | null |
null | {
"abstract": " We theoretically investigate the dispersion and polarization properties of\nthe electromagnetic waves in a multi-layered structure composed of a\nmagneto-optic waveguide on dielectric substrate covered by one-dimensional\ndielectric photonic crystal. The numerical analysis of such a complex structure\nshows polarization filtration of TE- and TM-modes depending on geometrical\nparameters of the waveguide and photonic crystal. We consider different regimes\nof the modes propagation inside such a structure: when guiding modes propagate\ninside the magnetic film and decay in the photonic crystal; when they propagate\nin both magnetic film and photonic crystal.\n",
"title": "Complex waveguide based on a magneto-optic layer and a dielectric photonic crystal"
} | null | null | [
"Physics"
]
| null | true | null | 2676 | null | Validated | null | null |
null | {
"abstract": " In this paper, we develop a new approach to the discrimi-nant of a complete\nintersection curve in the 3-dimensional projective space. By relying on the\nresultant theory, we first prove a new formula that allows us to define this\ndiscrimi-nant without ambiguity and over any commutative ring, in particular in\nany characteristic. This formula also provides a new method for evaluating and\ncomputing this discrimi-nant efficiently, without the need to introduce new\nvariables as with the well-known Cayley trick. Then, we obtain new properties\nand computational rules such as the covariance and the invariance formulas.\nFinally, we show that our definition of the discriminant satisfies to the\nexpected geometric property and hence yields an effective smoothness criterion\nfor complete intersection space curves. Actually, we show that in the generic\nsetting, it is the defining equation of the discriminant scheme if the ground\nring is assumed to be a unique factorization domain.\n",
"title": "Discriminants of complete intersection space curves"
} | null | null | [
"Computer Science",
"Mathematics"
]
| null | true | null | 2677 | null | Validated | null | null |
null | {
"abstract": " There is a digraph corresponding to every square matrix over $\\mathbb{C}$. We\ngenerate a recurrence relation using the Laplace expansion to calculate the\ncharacteristic, and permanent polynomials of a square matrix. Solving this\nrecurrence relation, we found that the characteristic, and permanent\npolynomials can be calculated in terms of characteristic, and permanent\npolynomials of some specific induced subdigraphs of blocks in the digraph,\nrespectively. Interestingly, these induced subdigraphs are vertex-disjoint and\nthey partition the digraph. Similar to the characteristic, and permanent\npolynomials; the determinant, and permanent can also be calculated. Therefore,\nthis article provides a combinatorial meaning of these useful quantities of the\nmatrix theory. We conclude this article with a number of open problems which\nmay be attempted for further research in this direction.\n",
"title": "On the Characteristic and Permanent Polynomials of a Matrix"
} | null | null | [
"Computer Science"
]
| null | true | null | 2678 | null | Validated | null | null |
null | {
"abstract": " We study the Loschmidt echo for quenches in open one-dimensional lattice\nmodels with symmetry protected topological phases. For quenches where dynamical\nquantum phase transitions do occur we find that cusps in the bulk return rate\nat critical times tc are associated with sudden changes in the boundary\ncontribution. For our main example, the Su-Schrieffer-Heeger model, we show\nthat these sudden changes are related to the periodical appearance of two\neigenvalues close to zero in the dynamical Loschmidt matrix. We demonstrate,\nfurthermore, that the structure of the Loschmidt spectrum is linked to the\nperiodic creation of long-range entanglement between the edges of the system.\n",
"title": "A bulk-boundary correspondence for dynamical phase transitions in one-dimensional topological insulators and superconductors"
} | null | null | null | null | true | null | 2679 | null | Default | null | null |
null | {
"abstract": " A new prior is proposed for representation learning, which can be combined\nwith other priors in order to help disentangling abstract factors from each\nother. It is inspired by the phenomenon of consciousness seen as the formation\nof a low-dimensional combination of a few concepts constituting a conscious\nthought, i.e., consciousness as awareness at a particular time instant. This\nprovides a powerful constraint on the representation in that such\nlow-dimensional thought vectors can correspond to statements about reality\nwhich are true, highly probable, or very useful for taking decisions. The fact\nthat a few elements of the current state can be combined into such a predictive\nor useful statement is a strong constraint and deviates considerably from the\nmaximum likelihood approaches to modelling data and how states unfold in the\nfuture based on an agent's actions. Instead of making predictions in the\nsensory (e.g. pixel) space, the consciousness prior allows the agent to make\npredictions in the abstract space, with only a few dimensions of that space\nbeing involved in each of these predictions. The consciousness prior also makes\nit natural to map conscious states to natural language utterances or to express\nclassical AI knowledge in the form of facts and rules, although the conscious\nstates may be richer than what can be expressed easily in the form of a\nsentence, a fact or a rule.\n",
"title": "The Consciousness Prior"
} | null | null | null | null | true | null | 2680 | null | Default | null | null |
null | {
"abstract": " We propose a multi-scale edge-detection algorithm to search for the\nGott-Kaiser-Stebbins imprints of a cosmic string (CS) network on the Cosmic\nMicrowave Background (CMB) anisotropies. Curvelet decomposition and extended\nCanny algorithm are used to enhance the string detectability. Various\nstatistical tools are then applied to quantify the deviation of CMB maps having\na cosmic string contribution with respect to pure Gaussian anisotropies of\ninflationary origin. These statistical measures include the one-point\nprobability density function, the weighted two-point correlation function\n(TPCF) of the anisotropies, the unweighted TPCF of the peaks and of the\nup-crossing map, as well as their cross-correlation. We use this algorithm on a\nhundred of simulated Nambu-Goto CMB flat sky maps, covering approximately\n$10\\%$ of the sky, and for different string tensions $G\\mu$. On noiseless sky\nmaps with an angular resolution of $0.9'$, we show that our pipeline detects\nCSs with $G\\mu$ as low as $G\\mu\\gtrsim 4.3\\times 10^{-10}$. At the same\nresolution, but with a noise level typical to a CMB-S4 phase II experiment, the\ndetection threshold would be to $G\\mu\\gtrsim 1.2 \\times 10^{-7}$.\n",
"title": "Multi-Scale Pipeline for the Search of String-Induced CMB Anisotropies"
} | null | null | null | null | true | null | 2681 | null | Default | null | null |
null | {
"abstract": " Inspired by recent work of I. Baoulina, we give a simultaneous generalization\nof the theorems of Chevalley-Warning and Morlaye.\n",
"title": "A simultaneous generalization of the theorems of Chevalley-Warning and Morlaye"
} | null | null | null | null | true | null | 2682 | null | Default | null | null |
null | {
"abstract": " In this paper, we study the fractional Poisson process (FPP) time-changed by\nan independent Lévy subordinator and the inverse of the Lévy subordinator,\nwhich we call TCFPP-I and TCFPP-II, respectively. Various distributional\nproperties of these processes are established. We show that, under certain\nconditions, the TCFPP-I has the long-range dependence property and also its law\nof iterated logarithm is proved. It is shown that the TCFPP-II is a renewal\nprocess and its waiting time distribution is identified. Its bivariate\ndistributions and also the governing difference-differential equation are\nderived. Some specific examples for both the processes are discussed. Finally,\nwe present the simulations of the sample paths of these processes.\n",
"title": "Some Time-changed fractional Poisson processes"
} | null | null | null | null | true | null | 2683 | null | Default | null | null |
null | {
"abstract": " Adaptive Fourier decomposition (AFD, precisely 1-D AFD or Core-AFD) was\noriginated for the goal of positive frequency representations of signals. It\nachieved the goal and at the same time offered fast decompositions of signals.\nThere then arose several types of AFDs. AFD merged with the greedy algorithm\nidea, and in particular, motivated the so-called pre-orthogonal greedy\nalgorithm (Pre-OGA) that was proven to be the most efficient greedy algorithm.\nThe cost of the advantages of the AFD type decompositions is, however, the high\ncomputational complexity due to the involvement of maximal selections of the\ndictionary parameters. The present paper offers one formulation of the 1-D AFD\nalgorithm by building the FFT algorithm into it. Accordingly, the algorithm\ncomplexity is reduced, from the original $\\mathcal{O}(M N^2)$ to $\\mathcal{O}(M\nN\\log_2 N)$, where $N$ denotes the number of the discretization points on the\nunit circle and $M$ denotes the number of points in $[0,1)$. This greatly\nenhances the applicability of AFD. Experiments are carried out to show the high\nefficiency of the proposed algorithm.\n",
"title": "Fast algorithm of adaptive Fourier series"
} | null | null | [
"Mathematics"
]
| null | true | null | 2684 | null | Validated | null | null |
null | {
"abstract": " Due to the growth of geo-tagged images, recent web and mobile applications\nprovide search capabilities for images that are similar to a given query image\nand simultaneously within a given geographical area. In this paper, we focus on\ndesigning index structures to expedite these spatial-visual searches. We start\nby baseline indexes that are straightforward extensions of the current popular\nspatial (R*-tree) and visual (LSH) index structures. Subsequently, we propose\nhybrid index structures that evaluate both spatial and visual features in\ntandem. The unique challenge of this type of query is that there are\ninaccuracies in both spatial and visual features. Therefore, different\ntraversals of the index structures may produce different images as output, some\nof which more relevant to the query than the others. We compare our hybrid\nstructures with a set of baseline indexes in both performance and result\naccuracy using three real world datasets from Flickr, Google Street View, and\nGeoUGV.\n",
"title": "Hybrid Indexes to Expedite Spatial-Visual Search"
} | null | null | null | null | true | null | 2685 | null | Default | null | null |
null | {
"abstract": " Electron Cryo-Tomography (ECT) enables 3D visualization of macromolecule\nstructure inside single cells. Macromolecule classification approaches based on\nconvolutional neural networks (CNN) were developed to separate millions of\nmacromolecules captured from ECT systematically. However, given the fast\naccumulation of ECT data, it will soon become necessary to use CNN models to\nefficiently and accurately separate substantially more macromolecules at the\nprediction stage, which requires additional computational costs. To speed up\nthe prediction, we compress classification models into compact neural networks\nwith little in accuracy for deployment. Specifically, we propose to perform\nmodel compression through knowledge distillation. Firstly, a complex teacher\nnetwork is trained to generate soft labels with better classification\nfeasibility followed by training of customized student networks with simple\narchitectures using the soft label to compress model complexity. Our tests\ndemonstrate that our compressed models significantly reduce the number of\nparameters and time cost while maintaining similar classification accuracy.\n",
"title": "Model compression for faster structural separation of macromolecules captured by Cellular Electron Cryo-Tomography"
} | null | null | null | null | true | null | 2686 | null | Default | null | null |
null | {
"abstract": " We use the Kotliar-Ruckenstein slave-boson formalism to study the temperature\ndependence of paramagnetic phases of the one-band Hubbard model for a variety\nof band structures. We calculate the Fermi liquid quasiparticle spectral weight\n$Z$ and identify the temperature at which it decreases significantly to a\ncrossover to a bad metal region. Near the Mott metal-insulator transition, this\ncoherence temperature $T_\\textrm{coh}$ is much lower than the Fermi temperature\nof the uncorrelated Fermi gas, as is observed in a broad range of strongly\ncorrelated electron materials. After a proper rescaling of temperature and\ninteraction, we find a universal behavior that is independent of the band\nstructure of the system. We obtain the temperature-interaction phase diagram as\na function of doping, and we compare the temperature dependence of the double\noccupancy, entropy, and charge compressibility with previous results obtained\nwith Dynamical Mean-Field Theory. We analyse the stability of the method by\ncalculating the charge compressibility.\n",
"title": "Low quasiparticle coherence temperature in the one band-Hubbard model: A slave-boson approach"
} | null | null | [
"Physics"
]
| null | true | null | 2687 | null | Validated | null | null |
null | {
"abstract": " Schmerl and Beklemishev's work on iterated reflection achieves two aims: It\nintroduces the important notion of $\\Pi^0_1$-ordinal, characterizing the\n$\\Pi^0_1$-theorems of a theory in terms of transfinite iterations of\nconsistency; and it provides an innovative calculus to compute the\n$\\Pi^0_1$-ordinals for a range of theories. The present note demonstrates that\nthese achievements are independent: We read off $\\Pi^0_1$-ordinals from a\nSchütte-style ordinal analysis via infinite proofs, in a direct and\ntransparent way.\n",
"title": "A Note on Iterated Consistency and Infinite Proofs"
} | null | null | null | null | true | null | 2688 | null | Default | null | null |
null | {
"abstract": " Internet of Things (IoT) is the next big evolutionary step in the world of\ninternet. The main intention behind the IoT is to enable safer living and risk\nmitigation on different levels of life. With the advent of IoT botnets, the\nview towards IoT devices has changed from enabler of enhanced living into\nInternet of vulnerabilities for cyber criminals. IoT botnets has exposed two\ndifferent glaring issues, 1) A large number of IoT devices are accessible over\npublic Internet. 2) Security (if considered at all) is often an afterthought in\nthe architecture of many wide spread IoT devices. In this article, we briefly\noutline the anatomy of the IoT botnets and their basic mode of operations. Some\nof the major DDoS incidents using IoT botnets in recent times along with the\ncorresponding exploited vulnerabilities will be discussed. We also provide\nremedies and recommendations to mitigate IoT related cyber risks and briefly\nillustrate the importance of cyber insurance in the modern connected world.\n",
"title": "Turning Internet of Things(IoT) into Internet of Vulnerabilities (IoV) : IoT Botnets"
} | null | null | null | null | true | null | 2689 | null | Default | null | null |
null | {
"abstract": " In this paper, we propose a new approach to Cwikel estimates both for the\nEuclidean space and for the noncommutative Euclidean space.\n",
"title": "Cwikel estimates revisited"
} | null | null | null | null | true | null | 2690 | null | Default | null | null |
null | {
"abstract": " Uncertainty analysis in the form of probabilistic forecasting can\nsignificantly improve decision making processes in the smart power grid for\nbetter integrating renewable energy sources such as wind. Whereas point\nforecasting provides a single expected value, probabilistic forecasts provide\nmore information in the form of quantiles, prediction intervals, or full\npredictive densities. This paper analyzes the effectiveness of a novel approach\nfor nonparametric probabilistic forecasting of wind power that combines a\nsmooth approximation of the pinball loss function with a neural network\narchitecture and a weighting initialization scheme to prevent the quantile\ncross over problem. A numerical case study is conducted using publicly\navailable wind data from the Global Energy Forecasting Competition 2014.\nMultiple quantiles are estimated to form 10%, to 90% prediction intervals which\nare evaluated using a quantile score and reliability measures. Benchmark models\nsuch as the persistence and climatology distributions, multiple quantile\nregression, and support vector quantile regression are used for comparison\nwhere results demonstrate the proposed approach leads to improved performance\nwhile preventing the problem of overlapping quantile estimates.\n",
"title": "Smooth Pinball Neural Network for Probabilistic Forecasting of Wind Power"
} | null | null | null | null | true | null | 2691 | null | Default | null | null |
null | {
"abstract": " Asynchronous-parallel algorithms have the potential to vastly speed up\nalgorithms by eliminating costly synchronization. However, our understanding to\nthese algorithms is limited because the current convergence of asynchronous\n(block) coordinate descent algorithms are based on somewhat unrealistic\nassumptions. In particular, the age of the shared optimization variables being\nused to update a block is assumed to be independent of the block being updated.\nAlso, it is assumed that the updates are applied to randomly chosen blocks. In\nthis paper, we argue that these assumptions either fail to hold or will imply\nless efficient implementations. We then prove the convergence of\nasynchronous-parallel block coordinate descent under more realistic\nassumptions, in particular, always without the independence assumption. The\nanalysis permits both the deterministic (essentially) cyclic and random rules\nfor block choices. Because a bound on the asynchronous delays may or may not be\navailable, we establish convergence for both bounded delays and unbounded\ndelays. The analysis also covers nonconvex, weakly convex, and strongly convex\nfunctions. We construct Lyapunov functions that directly model both objective\nprogress and delays, so delays are not treated errors or noise. A\ncontinuous-time ODE is provided to explain the construction at a high level.\n",
"title": "Asynchronous Coordinate Descent under More Realistic Assumptions"
} | null | null | null | null | true | null | 2692 | null | Default | null | null |
null | {
"abstract": " The ground-state magnetic response of fullerene molecules with up to 36\nvertices is calculated, when spins classical or with magnitude $s=\\frac{1}{2}$\nare located on their vertices and interact according to the nearest-neighbor\nantiferromagnetic Heisenberg model. The frustrated topology, which originates\nin the pentagons of the fullerenes and is enhanced by their close proximity,\nleads to a significant number of classical magnetization and susceptibility\ndiscontinuities, something not expected for a model lacking magnetic\nanisotropy. This establishes the classical discontinuities as a generic feature\nof fullerene molecules irrespective of their symmetry. The largest number of\ndiscontinuities have the molecule with 26 sites, four of the magnetization and\ntwo of the susceptibility, and an isomer with 34 sites, which has three each.\nIn addition, for several of the fullerenes the classical zero-field lowest\nenergy configuration has finite magnetization, which is unexpected for\nantiferromagnetic interactions between an even number of spins and with each\nspin having the same number of nearest-neighbors. The molecules come in\ndifferent symmetries and topologies and there are only a few patterns of\nmagnetic behavior that can be detected from such a small sample of relatively\nsmall fullerenes. Contrary to the classical case, in the full quantum limit\n$s=\\frac{1}{2}$ there are no discontinuities for a subset of the molecules that\nwas considered. This leaves the icosahedral symmetry fullerenes as the only\nones known supporting ground-state magnetization discontinuities for\n$s=\\frac{1}{2}$. It is also found that a molecule with 34 sites has a\ndoubly-degenerate ground state when $s=\\frac{1}{2}$.\n",
"title": "Zero-temperature magnetic response of small fullerene molecules at the classical and full quantum limit"
} | null | null | null | null | true | null | 2693 | null | Default | null | null |
null | {
"abstract": " We show that discrete distributions on the $d$-dimensional non-negative\ninteger lattice can be approximated arbitrarily well via the marginals of\nstationary distributions for various classes of stochastic chemical reaction\nnetworks. We begin by providing a class of detailed balanced networks and prove\nthat they can approximate any discrete distribution to any desired accuracy.\nHowever, these detailed balanced constructions rely on the ability to\ninitialize a system precisely, and are therefore susceptible to perturbations\nin the initial conditions. We therefore provide another construction based on\nthe ability to approximate point mass distributions and prove that this\nconstruction is capable of approximating arbitrary discrete distributions for\nany choice of initial condition. In particular, the developed models are\nergodic, so their limit distributions are robust to a finite number of\nperturbations over time in the counts of molecules.\n",
"title": "Stochastic Chemical Reaction Networks for Robustly Approximating Arbitrary Probability Distributions"
} | null | null | null | null | true | null | 2694 | null | Default | null | null |
null | {
"abstract": " The incorporation of macro-actions (temporally extended actions) into\nmulti-agent decision problems has the potential to address the curse of\ndimensionality associated with such decision problems. Since macro-actions last\nfor stochastic durations, multiple agents executing decentralized policies in\ncooperative environments must act asynchronously. We present an algorithm that\nmodifies Generalized Advantage Estimation for temporally extended actions,\nallowing a state-of-the-art policy optimization algorithm to optimize policies\nin Dec-POMDPs in which agents act asynchronously. We show that our algorithm is\ncapable of learning optimal policies in two cooperative domains, one involving\nreal-time bus holding control and one involving wildfire fighting with unmanned\naircraft. Our algorithm works by framing problems as \"event-driven decision\nprocesses,\" which are scenarios where the sequence and timing of actions and\nevents are random and governed by an underlying stochastic process. In addition\nto optimizing policies with continuous state and action spaces, our algorithm\nalso facilitates the use of event-driven simulators, which do not require time\nto be discretized into time-steps. We demonstrate the benefit of using\nevent-driven simulation in the context of multiple agents taking asynchronous\nactions. We show that fixed time-step simulation risks obfuscating the sequence\nin which closely-separated events occur, adversely affecting the policies\nlearned. Additionally, we show that arbitrarily shrinking the time-step scales\npoorly with the number of agents.\n",
"title": "Deep Reinforcement Learning for Event-Driven Multi-Agent Decision Processes"
} | null | null | null | null | true | null | 2695 | null | Default | null | null |
null | {
"abstract": " The current dominant visual processing paradigm in both human and machine\nresearch is the feedforward, layered hierarchy of neural-like processing\nelements. Within this paradigm, visual saliency is seen by many to have a\nspecific role, namely that of early selection. Early selection is thought to\nenable very fast visual performance by limiting processing to only the most\nrelevant candidate portions of an image. Though this strategy has indeed led to\nimproved processing time efficiency in machine algorithms, at least one set of\ncritical tests of this idea has never been performed with respect to the role\nof early selection in human vision. How would the best of the current saliency\nmodels perform on the stimuli used by experimentalists who first provided\nevidence for this visual processing paradigm? Would the algorithms really\nprovide correct candidate sub-images to enable fast categorization on those\nsame images? Here, we report on a new series of tests of these questions whose\nresults suggest that it is quite unlikely that such an early selection process\nhas any role in human rapid visual categorization.\n",
"title": "Early Salient Region Selection Does Not Drive Rapid Visual Categorization"
} | null | null | null | null | true | null | 2696 | null | Default | null | null |
null | {
"abstract": " We describe algorithms for symbolic reasoning about executable models of type\nsystems, supporting three queries intended for designers of type systems.\nFirst, we check for type soundness bugs and synthesize a counterexample program\nif such a bug is found. Second, we compare two versions of a type system,\nsynthesizing a program accepted by one but rejected by the other. Third, we\nminimize the size of synthesized counterexample programs.\nThese algorithms symbolically evaluate typecheckers and interpreters,\nproducing formulas that characterize the set of programs that fail or succeed\nin the typechecker and the interpreter. However, symbolically evaluating\ninterpreters poses efficiency challenges, which are caused by having to merge\nexecution paths of the various possible input programs. Our main contribution\nis the Bonsai tree, a novel symbolic representation of programs and program\nstates which addresses these challenges. Bonsai trees encode complex syntactic\ninformation in terms of logical constraints, enabling more efficient merging.\nWe implement these algorithms in the Bonsai tool, an assistant for type\nsystem designers. We perform case studies on how Bonsai helps test and explore\na variety of type systems. Bonsai efficiently synthesizes counterexamples for\nsoundness bugs that have been inaccessible to automatic tools, and is the first\nautomated tool to find a counterexample for the recently discovered Scala\nsoundness bug SI-9633.\n",
"title": "Bonsai: Synthesis-Based Reasoning for Type Systems"
} | null | null | null | null | true | null | 2697 | null | Default | null | null |
null | {
"abstract": " For Time-Domain Global Similarity (TDGS) method, which transforms the data\ncleaning problem into a binary classification problem about the physical\nsimilarity between channels, directly adopting common performance measures\ncould only guarantee the performance for physical similarity. Nevertheless,\npractical data cleaning tasks have preferences for the correctness of original\ndata sequences. To obtain the general expressions of performance measures based\non the preferences of tasks, the mapping relations between performance of TDGS\nmethod about physical similarity and correctness of data sequences are\ninvestigated by probability theory in this paper. Performance measures for TDGS\nmethod in several common data cleaning tasks are set. Cases when these\npreference-based performance measures could be simplified are introduced.\n",
"title": "Preference-based performance measures for Time-Domain Global Similarity method"
} | null | null | [
"Computer Science"
]
| null | true | null | 2698 | null | Validated | null | null |
null | {
"abstract": " If dark matter interactions with Standard Model particles are $CP$-violating,\nthen dark matter annihilation/decay can produce photons with a net circular\npolarization. We consider the prospects for experimentally detecting evidence\nfor such a circular polarization. We identify optimal models for dark matter\ninteractions with the Standard Model, from the point of view of detectability\nof the net polarization, for the case of either symmetric or asymmetric dark\nmatter. We find that, for symmetric dark matter, evidence for net polarization\ncould be found by a search of the Galactic Center by an instrument sensitive to\ncircular polarization with an efficiency-weighted exposure of at least\n$50000~\\text{cm}^2~\\text{yr}$, provided the systematic detector uncertainties\nare constrained at the $1\\%$ level. Better sensitivity can be obtained in the\ncase of asymmetric dark matter. We discuss the prospects for achieving the\nneeded level of performance using possible detector technologies.\n",
"title": "On the Prospects for Detecting a Net Photon Circular Polarization Produced by Decaying Dark Matter"
} | null | null | [
"Physics"
]
| null | true | null | 2699 | null | Validated | null | null |
null | {
"abstract": " We develop a strong diagnostic for bubbles and crashes in bitcoin, by\nanalyzing the coincidence (and its absence) of fundamental and technical\nindicators. Using a generalized Metcalfe's law based on network properties, a\nfundamental value is quantified and shown to be heavily exceeded, on at least\nfour occasions, by bubbles that grow and burst. In these bubbles, we detect a\nuniversal super-exponential unsustainable growth. We model this universal\npattern with the Log-Periodic Power Law Singularity (LPPLS) model, which\nparsimoniously captures diverse positive feedback phenomena, such as herding\nand imitation. The LPPLS model is shown to provide an ex-ante warning of market\ninstabilities, quantifying a high crash hazard and probabilistic bracket of the\ncrash time consistent with the actual corrections; although, as always, the\nprecise time and trigger (which straw breaks the camel's back) being exogenous\nand unpredictable. Looking forward, our analysis identifies a substantial but\nnot unprecedented overvaluation in the price of bitcoin, suggesting many months\nof volatile sideways bitcoin prices ahead (from the time of writing, March\n2018).\n",
"title": "Are Bitcoin Bubbles Predictable? Combining a Generalized Metcalfe's Law and the LPPLS Model"
} | null | null | null | null | true | null | 2700 | null | Default | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.