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": " Diffusion MRI measurements using hyperpolarized gases are generally acquired\nduring patient breath hold, which yields a compromise between achievable image\nresolution, lung coverage and number of b-values. In this work, we propose a\nnovel method that accelerates the acquisition of MR diffusion data by\nundersampling in both spatial and b-value dimensions, thanks to incorporating\nknowledge about the signal decay into the reconstruction (SIDER). SIDER is\ncompared to total variation (TV) reconstruction by assessing their effect on\nboth the recovery of ventilation images and estimated mean alveolar dimensions\n(MAD). Both methods are assessed by retrospectively undersampling diffusion\ndatasets of normal volunteers and COPD patients (n=8) for acceleration factors\nbetween x2 and x10. TV led to large errors and artefacts for acceleration\nfactors equal or larger than x5. SIDER improved TV, presenting lower errors and\nhistograms of MAD closer to those obtained from fully sampled data for\naccelerations factors up to x10. SIDER preserved image quality at all\nacceleration factors but images were slightly smoothed and some details were\nlost at x10. In conclusion, we have developed and validated a novel compressed\nsensing method for lung MRI imaging and achieved high acceleration factors,\nwhich can be used to increase the amount of data acquired during a breath-hold.\nThis methodology is expected to improve the accuracy of estimated lung\nmicrostructure dimensions and widen the possibilities of studying lung diseases\nwith MRI.\n",
"title": "Incorporation of prior knowledge of the signal behavior into the reconstruction to accelerate the acquisition of MR diffusion data"
}
| null | null | null | null | true | null |
16801
| null |
Default
| null | null |
null |
{
"abstract": " Understanding the nature of two-level tunneling defects is important for\nminimizing their disruptive effects in various nano-devices. By exploiting the\nresonant coupling of these defects to a superconducting qubit, one can probe\nand coherently manipulate them individually. In this work we utilize a phase\nqubit to induce Rabi oscillations of single tunneling defects and measure their\ndephasing rates as a function of the defect's asymmetry energy, which is tuned\nby an applied strain. The dephasing rates scale quadratically with the external\nstrain and are inversely proportional to the Rabi frequency. These results are\nanalyzed and explained within a model of interacting standard defects, in which\npure dephasing of coherent high-frequency (GHz) defects is caused by\ninteraction with incoherent low-frequency thermally excited defects.\n",
"title": "Rabi noise spectroscopy of individual two-level tunneling defects"
}
| null | null | null | null | true | null |
16802
| null |
Default
| null | null |
null |
{
"abstract": " We propose an algorithm for the adaptation of the learning rate for\nstochastic gradient descent (SGD) that avoids the need for validation set use.\nThe idea for the adaptiveness comes from the technique of extrapolation: to get\nan estimate for the error against the gradient flow which underlies SGD, we\ncompare the result obtained by one full step and two half-steps. The algorithm\nis applied in two separate frameworks: federated and differentially private\nlearning. Using examples of deep neural networks we empirically show that the\nadaptive algorithm is competitive with manually tuned commonly used\noptimisation methods for differentially privately training. We also show that\nit works robustly in the case of federated learning unlike commonly used\noptimisation methods.\n",
"title": "Learning rate adaptation for federated and differentially private learning"
}
| null | null | null | null | true | null |
16803
| null |
Default
| null | null |
null |
{
"abstract": " Generalizations of the Hermite polynomials to many variables and/or to the\ncomplex domain have been located in mathematical and physical literature for\nsome decades. Polynomials traditionally called complex Hermite ones are mostly\nunderstood as polynomials in $z$ and $\\bar{z}$ which in fact makes them\npolynomials in two real variables with complex coefficients. The present paper\nproposes to investigate for the first time holomorphic Hermite polynomials in\ntwo variables. Their algebraic and analytic properties are developed here.\nWhile the algebraic properties do not differ too much for those considered so\nfar, their analytic features are based on a kind of non-rotational\northogonality invented by van Eijndhoven and Meyers. Inspired by their\ninvention we merely follow the idea of Bargmann's seminal paper (1961) giving\nexplicit construction of reproducing kernel Hilbert spaces based on those\npolynomials. \"Homotopic\" behavior of our new formation culminates in comparing\nit to the very classical Bargmann space of two variables on one edge and the\naforementioned Hermite polynomials in $z$ and $\\bar{z}$ on the other. Unlike in\nthe case of Bargmann's basis our Hermite polynomials are not product ones but\nfactorize to it when bonded together with the first case of limit properties\nleading both to the Bargmann basis and suitable form of the reproducing kernel.\nAlso in the second limit we recover standard results obeyed by Hermite\npolynomials in $z$ and $\\bar{z}$.\n",
"title": "Holomorphic Hermite polynomials in two variables"
}
| null | null |
[
"Mathematics"
] | null | true | null |
16804
| null |
Validated
| null | null |
null |
{
"abstract": " Interactions between people are the basis on which the structure of our\nsociety arises as a complex system and, at the same time, are the starting\npoint of any physical description of it. In the last few years, much\ntheoretical research has addressed this issue by combining the physics of\ncomplex networks with a description of interactions in terms of evolutionary\ngame theory. We here take this research a step further by introducing a most\nsalient societal factor such as the individuals' preferences, a characteristic\nthat is key to understand much of the social phenomenology these days. We\nconsider a heterogeneous, agent-based model in which agents interact\nstrategically with their neighbors but their preferences and payoffs for the\npossible actions differ. We study how such a heterogeneous network behaves\nunder evolutionary dynamics and different strategic interactions, namely\ncoordination games and best shot games. With this model we study the emergence\nof the equilibria predicted analytically in random graphs under best response\ndynamics, and we extend this test to unexplored contexts like proportional\nimitation and scale free networks. We show that some theoretically predicted\nequilibria do not arise in simulations with incomplete Information, and we\ndemonstrate the importance of the graph topology and the payoff function\nparameters for some games. Finally, we discuss our results with available\nexperimental evidence on coordination games, showing that our model agrees\nbetter with the experiment that standard economic theories, and draw hints as\nto how to maximize social efficiency in situations of conflicting preferences.\n",
"title": "Equilibria, information and frustration in heterogeneous network games with conflicting preferences"
}
| null | null | null | null | true | null |
16805
| null |
Default
| null | null |
null |
{
"abstract": " Dynamic topic models (DTMs) model the evolution of prevalent themes in\nliterature, online media, and other forms of text over time. DTMs assume that\nword co-occurrence statistics change continuously and therefore impose\ncontinuous stochastic process priors on their model parameters. These dynamical\npriors make inference much harder than in regular topic models, and also limit\nscalability. In this paper, we present several new results around DTMs. First,\nwe extend the class of tractable priors from Wiener processes to the generic\nclass of Gaussian processes (GPs). This allows us to explore topics that\ndevelop smoothly over time, that have a long-term memory or are temporally\nconcentrated (for event detection). Second, we show how to perform scalable\napproximate inference in these models based on ideas around stochastic\nvariational inference and sparse Gaussian processes. This way we can train a\nrich family of DTMs to massive data. Our experiments on several large-scale\ndatasets show that our generalized model allows us to find interesting patterns\nthat were not accessible by previous approaches.\n",
"title": "Scalable Generalized Dynamic Topic Models"
}
| null | null | null | null | true | null |
16806
| null |
Default
| null | null |
null |
{
"abstract": " In the setting of the pi-calculus with binary sessions, we aim at relaxing\nthe notion of duality of session types by the concept of retractable compliance\ndeveloped in contract theory. This leads to extending session types with a new\ntype operator of \"speculative selection\" including choices not necessarily\noffered by a compliant partner. We address the problem of selecting successful\ncommunicating branches by means of an operational semantics based on\norchestrators, which has been shown to be equivalent to the retractable\nsemantics of contracts, but clearly more feasible. A type system, sound with\nrespect to such a semantics, is hence provided.\n",
"title": "Session Types for Orchestrated Interactions"
}
| null | null | null | null | true | null |
16807
| null |
Default
| null | null |
null |
{
"abstract": " Modularity in military vehicle designs enables on-base assembly, disassembly,\nand reconfiguration of vehicles, which can be beneficial in promoting fleet\nadaptability and life cycle cost savings. To properly manage the fleet\noperation and to control the resupply, demand prediction, and scheduling\nprocess, this paper illustrates an agent-based approach customized for highly\nmodularized military vehicle fleets and studies the feasibility and flexibility\nof modularity for various mission scenarios. Given deterministic field demands\nwith operation stochasticity, we compare the performance of a modular fleet to\na conventional fleet in equivalent operation strategies and also compare fleet\nperformance driven by heuristic rules and optimization. Several indicators are\nselected to quantify the fleet performance, including operation costs, total\nresupplied resources, and fleet readiness.\nWhen the model is implemented for military Joint Tactical Transport System\n(JTTS) mission, our results indicate that fleet modularity can reduce total\nresource supplies without significant losses in fleet readiness. The benefits\nof fleet modularity can also be amplified through a real-time optimized\noperation strategy. To highlight the feasibility of fleet modularity, a\nparametric study is performed to show the impacts from working capacity on\nmodular fleet performance. Finally, we provide practical suggestions of modular\nvehicle designs based on the analysis and other possible usage.\n",
"title": "An Agent-Based Approach for Optimizing Modular Vehicle Fleet Operation"
}
| null | null | null | null | true | null |
16808
| null |
Default
| null | null |
null |
{
"abstract": " Under certain general conditions, an explicit formula to compute the greatest\ndelta-epsilon function of a continuous function is given. From this formula, a\nnew way to analyze the uniform continuity of a continuous function is given.\nSeveral examples illustrating the theory are discussed.\n",
"title": "Delta-epsilon functions and uniform continuity on metric spaces"
}
| null | null | null | null | true | null |
16809
| null |
Default
| null | null |
null |
{
"abstract": " In this work, we study the problem of dispersion of mobile robots on dynamic\nrings. The problem of dispersion of $n$ robots on an $n$ node graph, introduced\nby Augustine and Moses Jr. [1], requires robots to coordinate with each other\nand reach a configuration where exactly one robot is present on each node. This\nproblem has real world applications and applies whenever we want to minimize\nthe total cost of $n$ agents sharing $n$ resources, located at various places,\nsubject to the constraint that the cost of an agent moving to a different\nresource is comparatively much smaller than the cost of multiple agents sharing\na resource (e.g. smart electric cars sharing recharge stations). The study of\nthis problem also provides indirect benefits to the study of scattering on\ngraphs, the study of exploration by mobile robots, and the study of load\nbalancing on graphs.\nWe solve the problem of dispersion in the presence of two types of dynamism\nin the underlying graph: (i) vertex permutation and (ii) 1-interval\nconnectivity. We introduce the notion of vertex permutation dynamism and have\nit mean that for a given set of nodes, in every round, the adversary ensures a\nring structure is maintained, but the connections between the nodes may change.\nWe use the idea of 1-interval connectivity from Di Luna et al. [10], where for\na given ring, in each round, the adversary chooses at most one edge to remove.\nWe assume robots have full visibility and present asymptotically time optimal\nalgorithms to achieve dispersion in the presence of both types of dynamism when\nrobots have chirality. When robots do not have chirality, we present\nasymptotically time optimal algorithms to achieve dispersion subject to certain\nconstraints. Finally, we provide impossibility results for dispersion when\nrobots have no visibility.\n",
"title": "Deterministic Dispersion of Mobile Robots in Dynamic Rings"
}
| null | null | null | null | true | null |
16810
| null |
Default
| null | null |
null |
{
"abstract": " Early prognosis of Alzheimer's dementia is hard. Mild cognitive impairment\n(MCI) typically precedes Alzheimer's dementia, yet only a fraction of MCI\nindividuals will progress to dementia, even when screened using biomarkers. We\npropose here to identify a subset of individuals who share a common brain\nsignature highly predictive of oncoming dementia. This signature was composed\nof brain atrophy and functional dysconnectivity and discovered using a machine\nlearning model in patients suffering from dementia. The model recognized the\nsame brain signature in MCI individuals, 90% of which progressed to dementia\nwithin three years. This result is a marked improvement on the state-of-the-art\nin prognostic precision, while the brain signature still identified 47% of all\nMCI progressors. We thus discovered a sizable MCI subpopulation which\nrepresents an excellent recruitment target for clinical trials at the prodromal\nstage of Alzheimer's disease.\n",
"title": "A brain signature highly predictive of future progression to Alzheimer's dementia"
}
| null | null | null | null | true | null |
16811
| null |
Default
| null | null |
null |
{
"abstract": " Automatic Music Transcription (AMT) is one of the oldest and most\nwell-studied problems in the field of music information retrieval. Within this\nchallenging research field, onset detection and instrument recognition take\nimportant places in transcription systems, as they respectively help to\ndetermine exact onset times of notes and to recognize the corresponding\ninstrument sources. The aim of this study is to explore the usefulness of\nmultiscale scattering operators for these two tasks on plucked string\ninstrument and piano music. After resuming the theoretical background and\nillustrating the key features of this sound representation method, we evaluate\nits performances comparatively to other classical sound representations. Using\nboth MIDI-driven datasets with real instrument samples and real musical pieces,\nscattering is proved to outperform other sound representations for these AMT\nsubtasks, putting forward its richer sound representation and invariance\nproperties.\n",
"title": "Deep scattering transform applied to note onset detection and instrument recognition"
}
| null | null | null | null | true | null |
16812
| null |
Default
| null | null |
null |
{
"abstract": " We prove the Gaschütz Lemma holds for all metrisable compact groups.\n",
"title": "Gaschütz Lemma for Compact Groups"
}
| null | null | null | null | true | null |
16813
| null |
Default
| null | null |
null |
{
"abstract": " We present a simplified description for spin-dependent electronic transport\nin honeycomb-lattice structures with spin-orbit interactions, using\ngeneralizations of the stochastic non-equilibrium model known as the totally\nasymmetric simple exclusion process. Mean field theory and numerical\nsimulations are used to study currents, density profiles and current\npolarization in quasi- one dimensional systems with open boundaries, and\nexternally-imposed particle injection ($\\alpha$) and ejection ($\\beta$) rates.\nWe investigate the influence of allowing for double site occupancy, according\nto Pauli's exclusion principle, on the behavior of the quantities of interest.\nWe find that double occupancy shows strong signatures for specific combinations\nof rates, namely high $\\alpha$ and low $\\beta$, but otherwise its effects are\nquantitatively suppressed. Comments are made on the possible relevance of the\npresent results to experiments on suitably doped graphenelike structures.\n",
"title": "Driven flow with exclusion and spin-dependent transport in graphenelike structures"
}
| null | null | null | null | true | null |
16814
| null |
Default
| null | null |
null |
{
"abstract": " The interconnected nature of graphs often results in difficult to interpret\nclutter. Typically techniques focus on either decluttering by clustering nodes\nwith similar properties or grouping edges with similar relationship. We propose\nusing mapper, a powerful topological data analysis tool, to summarize the\nstructure of a graph in a way that both clusters data with similar properties\nand preserves relationships. Typically, mapper operates on a given data by\nutilizing a scalar function defined on every point in the data and a cover for\nscalar function codomain. The output of mapper is a graph that summarize the\nshape of the space. In this paper, we outline how to use this mapper\nconstruction on an input graphs, outline three filter functions that capture\nimportant structures of the input graph, and provide an interface for\ninteractively modifying the cover. To validate our approach, we conduct several\ncase studies on synthetic and real world data sets and demonstrate how our\nmethod can give meaningful summaries for graphs with various complexities\n",
"title": "MOG: Mapper on Graphs for Relationship Preserving Clustering"
}
| null | null | null | null | true | null |
16815
| null |
Default
| null | null |
null |
{
"abstract": " A compact version of the Variation Evolving Method (VEM) is developed for the\noptimal control computation. It follows the idea that originates from the\ncontinuous-time dynamics stability theory in the control field. The optimal\nsolution is analogized to the equilibrium point of a dynamic system and is\nanticipated to be obtained in an asymptotically evolving way. With the\nintroduction of a virtual dimension, the variation time, the Evolution Partial\nDifferential Equation (EPDE), which describes the variation motion towards the\noptimal solution, is deduced from the Optimal Control Problem (OCP), and the\nequivalent optimality conditions with no employment of costates are\nestablished. In particular, it is found that theoretically the analytic\nfeedback optimal control law does not exist for general OCPs because the\noptimal control is related to the future state. Since the derived EPDE is\nsuitable to be solved with the semi-discrete method in the field of PDE\nnumerical calculation, the resulting Initial-value Problems (IVPs) may be\nsolved with mature Ordinary Differential Equation (ODE) numerical integration\nmethods.\n",
"title": "Variation Evolving for Optimal Control Computation, A Compact Way"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16816
| null |
Validated
| null | null |
null |
{
"abstract": " Convolutional Neural Networks (CNNs) have become the state-of-the-art in\nvarious computer vision tasks, but they are still premature for most sensor\ndata, especially in pervasive and wearable computing. A major reason for this\nis the limited amount of annotated training data. In this paper, we propose the\nidea of leveraging the discriminative power of pre-trained deep CNNs on\n2-dimensional sensor data by transforming the sensor modality to the visual\ndomain. By three proposed strategies, 2D sensor output is converted into\npressure distribution imageries. Then we utilize a pre-trained CNN for transfer\nlearning on the converted imagery data. We evaluate our method on a gait\ndataset of floor surface pressure mapping. We obtain a classification accuracy\nof 87.66%, which outperforms the conventional machine learning methods by over\n10%.\n",
"title": "Transforming Sensor Data to the Image Domain for Deep Learning - an Application to Footstep Detection"
}
| null | null | null | null | true | null |
16817
| null |
Default
| null | null |
null |
{
"abstract": " We design differentially private algorithms for the problem of online linear\noptimization in the full information and bandit settings with optimal\n$\\tilde{O}(\\sqrt{T})$ regret bounds. In the full-information setting, our\nresults demonstrate that $\\epsilon$-differential privacy may be ensured for\nfree -- in particular, the regret bounds scale as\n$O(\\sqrt{T})+\\tilde{O}\\left(\\frac{1}{\\epsilon}\\right)$. For bandit linear\noptimization, and as a special case, for non-stochastic multi-armed bandits,\nthe proposed algorithm achieves a regret of\n$\\tilde{O}\\left(\\frac{1}{\\epsilon}\\sqrt{T}\\right)$, while the previously known\nbest regret bound was\n$\\tilde{O}\\left(\\frac{1}{\\epsilon}T^{\\frac{2}{3}}\\right)$.\n",
"title": "The Price of Differential Privacy For Online Learning"
}
| null | null | null | null | true | null |
16818
| null |
Default
| null | null |
null |
{
"abstract": " Acoustic neutrino detection is a promising approach to extend the energy\nrange of neutrino telescopes to energies beyond $10^{18}$\\,eV. Currently\noperational and planned water-Cherenkov neutrino telescopes, most notably\nKM3NeT, include acoustic sensors in addition to the optical ones. These\nacoustic sensors could be used as instruments for acoustic detection, while\ntheir main purpose is the position calibration of the detection units. In this\narticle, a Monte Carlo simulation chain for acoustic detectors will be\npresented, covering the initial interaction of the neutrino up to the signal\nclassification of recorded events. The ambient and transient background in the\nsimulation was implemented according to data recorded by the acoustic set-up\nAMADEUS inside the ANTARES detector. The effects of refraction on the neutrino\nsignature in the detector are studied, and a classification of the recorded\nevents is implemented. As bipolar waveforms similar to those of the expected\nneutrino signals are also emitted from other sound sources, additional features\nlike the geometrical shape of the propagation have to be considered for the\nsignal classification. This leads to a large improvement of the background\nsuppression by almost two orders of magnitude, since a flat cylindrical\n\"pancake\" propagation pattern is a distinctive feature of neutrino signals. An\noverview of the simulation chain and the signal classification will be\npresented and preliminary studies of the performance of the classification will\nbe discussed.\n",
"title": "Simulation chain and signal classification for acoustic neutrino detection in seawater"
}
| null | null |
[
"Physics"
] | null | true | null |
16819
| null |
Validated
| null | null |
null |
{
"abstract": " Deep reinforcement learning (RL) methods generally engage in exploratory\nbehavior through noise injection in the action space. An alternative is to add\nnoise directly to the agent's parameters, which can lead to more consistent\nexploration and a richer set of behaviors. Methods such as evolutionary\nstrategies use parameter perturbations, but discard all temporal structure in\nthe process and require significantly more samples. Combining parameter noise\nwith traditional RL methods allows to combine the best of both worlds. We\ndemonstrate that both off- and on-policy methods benefit from this approach\nthrough experimental comparison of DQN, DDPG, and TRPO on high-dimensional\ndiscrete action environments as well as continuous control tasks. Our results\nshow that RL with parameter noise learns more efficiently than traditional RL\nwith action space noise and evolutionary strategies individually.\n",
"title": "Parameter Space Noise for Exploration"
}
| null | null | null | null | true | null |
16820
| null |
Default
| null | null |
null |
{
"abstract": " We present Deep Illumination, a novel machine learning technique for\napproximating global illumination (GI) in real-time applications using a\nConditional Generative Adversarial Network. Our primary focus is on generating\nindirect illumination and soft shadows with offline rendering quality at\ninteractive rates. Inspired from recent advancement in image-to-image\ntranslation problems using deep generative convolutional networks, we introduce\na variant of this network that learns a mapping from Gbuffers (depth map,\nnormal map, and diffuse map) and direct illumination to any global illumination\nsolution. Our primary contribution is showing that a generative model can be\nused to learn a density estimation from screen space buffers to an advanced\nillumination model for a 3D environment. Once trained, our network can\napproximate global illumination for scene configurations it has never\nencountered before within the environment it was trained on. We evaluate Deep\nIllumination through a comparison with both a state of the art real-time GI\ntechnique (VXGI) and an offline rendering GI technique (path tracing). We show\nthat our method produces effective GI approximations and is also\ncomputationally cheaper than existing GI techniques. Our technique has the\npotential to replace existing precomputed and screen-space techniques for\nproducing global illumination effects in dynamic scenes with physically-based\nrendering quality.\n",
"title": "Deep Illumination: Approximating Dynamic Global Illumination with Generative Adversarial Network"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16821
| null |
Validated
| null | null |
null |
{
"abstract": " Recurrent neural networks (RNNs) are important class of architectures among\nneural networks useful for language modeling and sequential prediction.\nHowever, optimizing RNNs is known to be harder compared to feed-forward neural\nnetworks. A number of techniques have been proposed in literature to address\nthis problem. In this paper we propose a simple technique called fraternal\ndropout that takes advantage of dropout to achieve this goal. Specifically, we\npropose to train two identical copies of an RNN (that share parameters) with\ndifferent dropout masks while minimizing the difference between their\n(pre-softmax) predictions. In this way our regularization encourages the\nrepresentations of RNNs to be invariant to dropout mask, thus being robust. We\nshow that our regularization term is upper bounded by the expectation-linear\ndropout objective which has been shown to address the gap due to the difference\nbetween the train and inference phases of dropout. We evaluate our model and\nachieve state-of-the-art results in sequence modeling tasks on two benchmark\ndatasets - Penn Treebank and Wikitext-2. We also show that our approach leads\nto performance improvement by a significant margin in image captioning\n(Microsoft COCO) and semi-supervised (CIFAR-10) tasks.\n",
"title": "Fraternal Dropout"
}
| null | null | null | null | true | null |
16822
| null |
Default
| null | null |
null |
{
"abstract": " The Behrens-Fisher problem is a well-known hypothesis testing problem in\nstatistics concerning two-sample mean comparison. In this article, we confirm\none conjecture in Eaton and Olshen (1972), which provides stochastic bounds for\nthe multivariate Behrens-Fisher test statistic under the null hypothesis. We\nalso extend their results on the stochastic ordering of random quotients to the\narbitrary finite dimensional case. This work can also be seen as a\ngeneralization of Hsu (1938) that provided the bounds for the univariate\nBehrens-Fisher problem. The results obtained in this article can be used to\nderive a testing procedure for the multivariate Behrens-Fisher problem that\nstrongly controls the Type I error.\n",
"title": "Finite-sample bounds for the multivariate Behrens-Fisher distribution with proportional covariances"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
16823
| null |
Validated
| null | null |
null |
{
"abstract": " Understanding the mechanisms underlying the formation of cultural traits,\nsuch as preferences, opinions and beliefs is an open challenge. Trait formation\nis intimately connected to cultural dynamics, which has been the focus of a\nvariety of quantitative models. Recently, some studies have emphasized the\nimportance of connecting those models to snapshots of cultural dynamics that\nare empirically accessible. By analyzing data obtained from different sources,\nit has been suggested that culture has properties that are universally present,\nand that empirical cultural states differ systematically from randomized\ncounterparts. Hence, a question about the mechanism responsible for the\nobserved patterns naturally arises. This study proposes a stochastic structural\nmodel for generating cultural states that retain those robust, empirical\nproperties. One ingredient of the model, already used in previous work, assumes\nthat every individual's set of traits is partly dictated by one of several,\nuniversal \"rationalities\", informally postulated by several social science\ntheories. The second, new ingredient taken from the same theories assumes that,\napart from a dominant rationality, each individual also has a certain exposure\nto the other rationalities. It is shown that both ingredients are required for\nreproducing the empirical regularities. This key result suggests that the\neffects of cultural dynamics in the real world can be described as an interplay\nof multiple, mixing rationalities, and thus provides indirect evidence for the\nclass of social science theories postulating such mixing. The model should be\nseen as a static, effective description of culture, while a dynamical, more\nfundamental description is left for future research.\n",
"title": "Evidence for mixed rationalities in preference formation"
}
| null | null | null | null | true | null |
16824
| null |
Default
| null | null |
null |
{
"abstract": " Active learning aims to train a classifier as fast as possible with as few\nlabels as possible. The core element in virtually any active learning strategy\nis the criterion that measures the usefulness of the unlabeled data based on\nwhich new points to be labeled are picked. We propose a novel approach which we\nrefer to as maximizing variance for active learning or MVAL for short. MVAL\nmeasures the value of unlabeled instances by evaluating the rate of change of\noutput variables caused by changes in the next sample to be queried and its\npotential labelling. In a sense, this criterion measures how unstable the\nclassifier's output is for the unlabeled data points under perturbations of the\ntraining data. MVAL maintains, what we refer to as, retraining information\nmatrices to keep track of these output scores and exploits two kinds of\nvariance to measure the informativeness and representativeness, respectively.\nBy fusing these variances, MVAL is able to select the instances which are both\ninformative and representative. We employ our technique both in combination\nwith logistic regression and support vector machines and demonstrate that MVAL\nachieves state-of-the-art performance in experiments on a large number of\nstandard benchmark datasets.\n",
"title": "A Variance Maximization Criterion for Active Learning"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
16825
| null |
Validated
| null | null |
null |
{
"abstract": " We compute the polarization function in a doped three-dimensional\nanisotropic-Weyl semimetal, in which the fermion energy dispersion is linear in\ntwo components of the momenta and quadratic in the third. Through detailed\ncalculations, we find that the long wavelength plasmon mode depends on the\nfermion density $n_e$ in the form $\\Omega_{p}^{\\bot}\\propto n_{e}^{3/10}$\nwithin the basal plane and behaves as $\\Omega_{p}^{z}\\propto n_{e}^{1/2}$ along\nthe third direction. This unique characteristic of the plasmon mode can be\nprobed by various experimental techniques, such as electron energy-loss\nspectroscopy. The Debye screening at finite chemical potential and finite\ntemperature is also analyzed based on the polarization function.\n",
"title": "Polarization, plasmon, and Debye screening in doped 3D ani-Weyl semimetal"
}
| null | null | null | null | true | null |
16826
| null |
Default
| null | null |
null |
{
"abstract": " Unsupervised machine learning via a restricted Boltzmann machine is an useful\ntool in distinguishing an ordered phase from a disordered phase. Here we study\nits application on the two-dimensional Ashkin-Teller model, which features a\npartially ordered product phase. We train the neural network with spin\nconfiguration data generated by Monte Carlo simulations and show that distinct\nfeatures of the product phase can be learned from non-ergodic samples resulting\nfrom symmetry breaking. Careful analysis of the weight matrices inspires us to\ndefine a nontrivial machine-learning motivated quantity of the product form,\nwhich resembles the conventional product order parameter.\n",
"title": "Identifying Product Order with Restricted Boltzmann Machines"
}
| null | null | null | null | true | null |
16827
| null |
Default
| null | null |
null |
{
"abstract": " We study the thermodynamics of ideal Bose gas as well as the transport\nproperties of non interacting bosons and fermions in a one dimensional\nquasi-periodic potential, namely Aubry-André (AA) model at finite\ntemperature. For bosons in finite size systems, the effect of quasi-periodic\npotential on the crossover phenomena corresponding to Bose-Einstein\ncondensation (BEC), superfluidity and localization phenomena at finite\ntemperatures are investigated. From the ground state number fluctuation we\ncalculate the crossover temperature of BEC which exhibits a non monotonic\nbehavior with the strength of AA potential and vanishes at the self-dual\ncritical point following power law. Appropriate rescaling of the crossover\ntemperatures reveals universal behavior which is studied for different\nquasi-periodicity of the AA model. Finally, we study the temperature and flux\ndependence of the persistent current of fermions in presence of a\nquasi-periodic potential to identify the localization at the Fermi energy from\nthe decay of the current.\n",
"title": "A finite temperature study of ideal quantum gases in the presence of one dimensional quasi-periodic potential"
}
| null | null | null | null | true | null |
16828
| null |
Default
| null | null |
null |
{
"abstract": " Using a high-frequency expansion in periodically driven extended Hubbard\nmodels, where the strengths and ranges of density-density interactions are\narbitrary, we obtain the effective interactions and bandwidth, which depend\nsensitively on the polarization of the driving field. Then, we numerically\ncalculate modulations of correlation functions in a quarter-filled extended\nHubbard model with nearest-neighbor interactions on a triangular lattice with\ntrimers after monocycle pulse excitation. We discuss how the resultant\nmodulations are compatible with the effective interactions and bandwidth\nderived above on the basis of their dependence on the polarization of\nphotoexcitation, which is easily accessible by experiments. Some correlation\nfunctions after monocycle pulse excitation are consistent with the effective\ninteractions, which are weaker or stronger than the original ones. However, the\nphotoinduced enhancement of anisotropic charge correlations previously\ndiscussed for the three-quarter-filled organic conductor\n$\\alpha$-(bis[ethylenedithio]-tetrathiafulvalene)$_2$I$_3$\n[$\\alpha$-(BEDT-TTF)$_2$I$_3$] in the metallic phase is not fully explained by\nthe effective interactions or bandwidth, which are derived independently of the\nfilling.\n",
"title": "High-Frequency Analysis of Effective Interactions and Bandwidth for Transient States after Monocycle Pulse Excitation of Extended Hubbard Model"
}
| null | null | null | null | true | null |
16829
| null |
Default
| null | null |
null |
{
"abstract": " This paper deals with two related problems, namely distance-preserving binary\nembeddings and quantization for compressed sensing . First, we propose fast\nmethods to replace points from a subset $\\mathcal{X} \\subset \\mathbb{R}^n$,\nassociated with the Euclidean metric, with points in the cube $\\{\\pm 1\\}^m$ and\nwe associate the cube with a pseudo-metric that approximates Euclidean distance\namong points in $\\mathcal{X}$. Our methods rely on quantizing fast\nJohnson-Lindenstrauss embeddings based on bounded orthonormal systems and\npartial circulant ensembles, both of which admit fast transforms. Our\nquantization methods utilize noise-shaping, and include Sigma-Delta schemes and\ndistributed noise-shaping schemes. The resulting approximation errors decay\npolynomially and exponentially fast in $m$, depending on the embedding method.\nThis dramatically outperforms the current decay rates associated with binary\nembeddings and Hamming distances. Additionally, it is the first such binary\nembedding result that applies to fast Johnson-Lindenstrauss maps while\npreserving $\\ell_2$ norms.\nSecond, we again consider noise-shaping schemes, albeit this time to quantize\ncompressed sensing measurements arising from bounded orthonormal ensembles and\npartial circulant matrices. We show that these methods yield a reconstruction\nerror that again decays with the number of measurements (and bits), when using\nconvex optimization for reconstruction. Specifically, for Sigma-Delta schemes,\nthe error decays polynomially in the number of measurements, and it decays\nexponentially for distributed noise-shaping schemes based on beta encoding.\nThese results are near optimal and the first of their kind dealing with bounded\northonormal systems.\n",
"title": "Fast binary embeddings, and quantized compressed sensing with structured matrices"
}
| null | null | null | null | true | null |
16830
| null |
Default
| null | null |
null |
{
"abstract": " Most past work on social network link fraud detection tries to separate\ngenuine users from fraudsters, implicitly assuming that there is only one type\nof fraudulent behavior. But is this assumption true? And, in either case, what\nare the characteristics of such fraudulent behaviors? In this work, we set up\nhoneypots (\"dummy\" social network accounts), and buy fake followers (after\ncareful IRB approval). We report the signs of such behaviors including oddities\nin local network connectivity, account attributes, and similarities and\ndifferences across fraud providers. Most valuably, we discover and characterize\nseveral types of fraud behaviors. We discuss how to leverage our insights in\npractice by engineering strongly performing entropy-based features and\ndemonstrating high classification accuracy. Our contributions are (a)\ninstrumentation: we detail our experimental setup and carefully engineered data\ncollection process to scrape Twitter data while respecting API rate-limits, (b)\nobservations on fraud multimodality: we analyze our honeypot fraudster\necosystem and give surprising insights into the multifaceted behaviors of these\nfraudster types, and (c) features: we propose novel features that give strong\n(>0.95 precision/recall) discriminative power on ground-truth Twitter data.\n",
"title": "The Many Faces of Link Fraud"
}
| null | null | null | null | true | null |
16831
| null |
Default
| null | null |
null |
{
"abstract": " We show that whenever $\\delta>0$, $\\eta$ is real and constants $\\lambda_i$\nsatisfy some necessary conditions, there are infinitely many prime triples\n$p_1,\\, p_2,\\, p_3$ satisfying the inequality $|\\lambda_1p_1 + \\lambda_2p_2 +\n\\lambda_3p_3+\\eta|<(\\max p_j)^{-1/12+\\delta}$ and such that, for each\n$i\\in\\{1,2,3\\}$, $p_i+2$ has at most $28$ prime factors.\n",
"title": "Diophantine approximation by special primes"
}
| null | null | null | null | true | null |
16832
| null |
Default
| null | null |
null |
{
"abstract": " Compositional Game Theory is a new, recently introduced model of economic\ngames based upon the computer science idea of compositionality. In it, complex\nand irregular games can be built up from smaller and simpler games, and the\nequilibria of these complex games can be defined recursively from the\nequilibria of their simpler subgames. This paper extends the model by providing\na final coalgebra semantics for infinite games. In the course of this, we\nintroduce a new operator on games to model the economic concept of subgame\nperfection.\n",
"title": "A Compositional Treatment of Iterated Open Games"
}
| null | null | null | null | true | null |
16833
| null |
Default
| null | null |
null |
{
"abstract": " Let $X_1, \\ldots, X_n$ be i.i.d. sample in $\\mathbb{R}^p$ with zero mean and\nthe covariance matrix $\\mathbf{\\Sigma^*}$. The classic principal component\nanalysis estimates the projector $\\mathbf{P^*_{\\mathcal{J}}}$ onto the direct\nsum of some eigenspaces of $\\mathbf{\\Sigma^*}$ by its empirical counterpart\n$\\mathbf{\\widehat{P}_{\\mathcal{J}}}$. Recent papers [Koltchinskii, Lounici\n(2017)], [Naumov et al. (2017)] investigate the asymptotic distribution of the\nFrobenius distance between the projectors $\\|\n\\mathbf{\\widehat{P}_{\\mathcal{J}}} - \\mathbf{P^*_{\\mathcal{J}}} \\|_2$. The\nproblem arises when one tries to build a confidence set for the true projector\neffectively. We consider the problem from Bayesian perspective and derive an\napproximation for the posterior distribution of the Frobenius distance between\nprojectors. The derived theorems hold true for non-Gaussian data: the only\nassumption that we impose is the concentration of the sample covariance\n$\\mathbf{\\widehat{\\Sigma}}$ in a vicinity of $\\mathbf{\\Sigma^*}$. The obtained\nresults are applied to construction of sharp confidence sets for the true\nprojector. Numerical simulations illustrate good performance of the proposed\nprocedure even on non-Gaussian data in quite challenging regime.\n",
"title": "Bayesian inference for spectral projectors of covariance matrix"
}
| null | null | null | null | true | null |
16834
| null |
Default
| null | null |
null |
{
"abstract": " Variational autoencoders (VAEs), as well as other generative models, have\nbeen shown to be efficient and accurate to capture the latent structure of vast\namounts of complex high-dimensional data. However, existing VAEs can still not\ndirectly handle data that are heterogenous (mixed continuous and discrete) or\nincomplete (with missing data at random), which is indeed common in real-world\napplications.\nIn this paper, we propose a general framework to design VAEs, suitable for\nfitting incomplete heterogenous data. The proposed HI-VAE includes likelihood\nmodels for real-valued, positive real valued, interval, categorical, ordinal\nand count data, and allows to estimate (and potentially impute) missing data\naccurately. Furthermore, HI-VAE presents competitive predictive performance in\nsupervised tasks, outperforming super- vised models when trained on incomplete\ndata\n",
"title": "Handling Incomplete Heterogeneous Data using VAEs"
}
| null | null | null | null | true | null |
16835
| null |
Default
| null | null |
null |
{
"abstract": " The space of all probability measures having positive density function on a\nconnected compact smooth manifold $M$, denoted by $\\mathcal{P}(M)$, carries the\nFisher information metric $G$. We define the geometric mean of probability\nmeasures by the aid of which we investigate information geometry of\n$\\mathcal{P}(M)$, equipped with $G$. We show that a geodesic segment joining\narbitrary probability measures $\\mu_1$ and $\\mu_2$ is expressed by using the\nnormalized geometric mean of its endpoints. As an application, we show that any\ntwo points of $\\mathcal{P}(M)$ can be joined by a geodesic. Moreover, we prove\nthat the function $\\ell$ defined by $\\ell(\\mu_1, \\mu_2):=2\\arccos\\int_M\n\\sqrt{p_1\\,p_2}\\,d\\lambda$, $\\mu_i=p_i\\,\\lambda$, $i=1,2$ gives the distance\nfunction on $\\mathcal{P}(M)$. It is shown that geodesics are all minimal.\n",
"title": "Geometric mean of probability measures and geodesics of Fisher information metric"
}
| null | null | null | null | true | null |
16836
| null |
Default
| null | null |
null |
{
"abstract": " In this article we study automorphisms of Toeplitz subshifts. Such groups are\nabelian and any finitely generated torsion subgroup is finite and cyclic. When\nthe complexity is non superlinear, we prove that the automorphism group is,\nmodulo a finite cyclic group, generated by a unique root of the shift. In the\nsubquadratic complexity case, we show that the automorphism group modulo the\ntorsion is generated by the roots of the shift map and that the result of the\nnon superlinear case is optimal. Namely, for any $\\varepsilon > 0$ we construct\nexamples of minimal Toeplitz subshifts with complexity bounded by $C\nn^{1+\\epsilon}$ whose automorphism groups are not finitely generated. Finally,\nwe observe the coalescence and the automorphism group give no restriction on\nthe complexity since we provide a family of coalescent Toeplitz subshifts with\npositive entropy such that their automorphism groups are arbitrary finitely\ngenerated infinite abelian groups with cyclic torsion subgroup (eventually\nrestricted to powers of the shift).\n",
"title": "On automorphism groups of Toeplitz subshifts"
}
| null | null | null | null | true | null |
16837
| null |
Default
| null | null |
null |
{
"abstract": " Recent results by Alagic and Russell have given some evidence that the\nEven-Mansour cipher may be secure against quantum adversaries with quantum\nqueries, if considered over other groups than $(\\mathbb{Z}/2)^n$. This prompts\nthe question as to whether or not other classical schemes may be generalized to\narbitrary groups and whether classical results still apply to those generalized\nschemes. In this thesis, we generalize the Even-Mansour cipher and the Feistel\ncipher. We show that Even and Mansour's original notions of secrecy are\nobtained on a one-key, group variant of the Even-Mansour cipher. We generalize\nthe result by Kilian and Rogaway, that the Even-Mansour cipher is pseudorandom,\nto super pseudorandomness, also in the one-key, group case. Using a Slide\nAttack we match the bound found above. After generalizing the Feistel cipher to\narbitrary groups we resolve an open problem of Patel, Ramzan, and Sundaram by\nshowing that the 3-round Feistel cipher over an arbitrary group is not super\npseudorandom. We generalize a result by Gentry and Ramzan showing that the\nEven-Mansour cipher can be implemented using the Feistel cipher as the public\npermutation. In this result, we also consider the one-key case over a group and\ngeneralize their bound. Finally, we consider Zhandry's result on quantum\npseudorandom permutations, showing that his result may be generalized to hold\nfor arbitrary groups. In this regard, we consider whether certain card shuffles\nmay be generalized as well.\n",
"title": "How to Generate Pseudorandom Permutations Over Other Groups"
}
| null | null | null | null | true | null |
16838
| null |
Default
| null | null |
null |
{
"abstract": " In the present work, we use information theory to understand the empirical\nconvergence rate of tractography, a widely-used approach to reconstruct\nanatomical fiber pathways in the living brain. Based on diffusion MRI data,\ntractography is the starting point for many methods to study brain\nconnectivity. Of the available methods to perform tractography, most\nreconstruct a finite set of streamlines, or 3D curves, representing probable\nconnections between anatomical regions, yet relatively little is known about\nhow the sampling of this set of streamlines affects downstream results, and how\nexhaustive the sampling should be. Here we provide a method to measure the\ninformation theoretic surprise (self-cross entropy) for tract sampling schema.\nWe then empirically assess four streamline methods. We demonstrate that the\nrelative information gain is very low after a moderate number of streamlines\nhave been generated for each tested method. The results give rise to several\nguidelines for optimal sampling in brain connectivity analyses.\n",
"title": "Measures of Tractography Convergence"
}
| null | null | null | null | true | null |
16839
| null |
Default
| null | null |
null |
{
"abstract": " Collaborative filtering is a broad and powerful framework for building\nrecommendation systems that has seen widespread adoption. Over the past decade,\nthe propensity of such systems for favoring popular products and thus creating\necho chambers have been observed. This has given rise to an active area of\nresearch that seeks to diversify recommendations generated by such algorithms.\nWe address the problem of increasing diversity in recommendation systems that\nare based on collaborative filtering that use past ratings to predicting a\nrating quality for potential recommendations. Following our earlier work, we\nformulate recommendation system design as a subgraph selection problem from a\ncandidate super-graph of potential recommendations where both diversity and\nrating quality are explicitly optimized: (1) On the modeling side, we define a\nnew flexible notion of diversity that allows a system designer to prescribe the\nnumber of recommendations each item should receive, and smoothly penalizes\ndeviations from this distribution. (2) On the algorithmic side, we show that\nminimum-cost network flow methods yield fast algorithms in theory and practice\nfor designing recommendation subgraphs that optimize this notion of diversity.\n(3) On the empirical side, we show the effectiveness of our new model and\nmethod to increase diversity while maintaining high rating quality in standard\nrating data sets from Netflix and MovieLens.\n",
"title": "Network Flow Based Post Processing for Sales Diversity"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16840
| null |
Validated
| null | null |
null |
{
"abstract": " We define a lattice model for rock, absorbers, and gas that makes it possible\nto examine the flow of gas to a complicated absorbing boundary over long\nperiods of time. The motivation is to deduce the geometry of the boundary from\nthe time history of gas absorption. We find a solution to this model using\nGreen's function techniques, and apply the solution to three absorbing networks\nof increasing complexity.\n",
"title": "Lattice Model for Production of Gas"
}
| null | null | null | null | true | null |
16841
| null |
Default
| null | null |
null |
{
"abstract": " We consider an extension of the contextual bandit setting, motivated by\nseveral practical applications, where an unlabeled history of contexts can\nbecome available for pre-training before the online decision-making begins. We\npropose an approach for improving the performance of contextual bandit in such\nsetting, via adaptive, dynamic representation learning, which combines offline\npre-training on unlabeled history of contexts with online selection and\nmodification of embedding functions. Our experiments on a variety of datasets\nand in different nonstationary environments demonstrate clear advantages of our\napproach over the standard contextual bandit.\n",
"title": "Adaptive Representation Selection in Contextual Bandit"
}
| null | null | null | null | true | null |
16842
| null |
Default
| null | null |
null |
{
"abstract": " Using the theory of cohomology support locus, we give a necessary condition\nfor the Albanese map of a smooth projective surface being a submersion. More\nprecisely, assuming the cohomology support locus of any finite abelian cover of\na smooth projective surface consists of finitely many points, we prove that the\nsurface has trivial first Betti number, or is a ruled surface of genus one, or\nis an abelian surface.\n",
"title": "Algebraic surfaces with zero-dimensional cohomology support locus"
}
| null | null | null | null | true | null |
16843
| null |
Default
| null | null |
null |
{
"abstract": " Analyzing temperature dependent photoemission (PE) data of the ferromagnetic\nKondo-lattice (KL) system YbNiSn in the light of the Periodic Anderson model\n(PAM) we show that the KL behavior is not limited to temperatures below a\ntemperature T_K, defined empirically from resistivity and specificic heat\nmeasurements. As characteristic for weakly hybridized Ce and Yb systems, the PE\nspectra reveal a 4f-derived Fermi level peak, which reflects contributions from\nthe Kondo resonance and its crystal electric field (CEF) satellites. In YbNiSn\nthis peak has an unusual temperature dependence: With decreasing temperature a\nsteady linear increase of intensity is observed which extends over a large\ninterval ranging from 100 K down to 1 K without showing any peculiarities in\nthe region of T_K ~ TC= 5.6 K. In the light of the single-impurity Anderson\nmodel (SIAM) this intensity variation reflects a linear increase of 4f\noccupancy with decreasing temperature, indicating an onset of Kondo screening\nat temperatures above 100 K. Within the PAM this phenomenon could be described\nby a non-Fermi liquid like T- linear damping of the self-energy which accounts\nphenomenologically for the feedback from the closely spaced CEF-states.\n",
"title": "Insight into the temperature dependent properties of the ferromagnetic Kondo lattice YbNiSn"
}
| null | null | null | null | true | null |
16844
| null |
Default
| null | null |
null |
{
"abstract": " Although proportional hazard rate model is a very popular model to analyze\nfailure time data, sometimes it becomes important to study the additive hazard\nrate model. Again, sometimes the concept of the hazard rate function is\nabstract, in comparison to the concept of mean residual life function. A new\nmodel called `dynamic additive mean residual life model' where the covariates\nare time-dependent has been defined in the literature. Here we study the\nclosure properties of the model for different positive and negative ageing\nclasses under certain condition(s). Quite a few examples are presented to\nillustrate different properties of the model.\n",
"title": "Some Ageing Properties of Dynamic Additive Mean Residual Life Model"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
16845
| null |
Validated
| null | null |
null |
{
"abstract": " We propose a Las Vegas transformation of Markov Chain Monte Carlo (MCMC)\nestimators of Restricted Boltzmann Machines (RBMs). We denote our approach\nMarkov Chain Las Vegas (MCLV). MCLV gives statistical guarantees in exchange\nfor random running times. MCLV uses a stopping set built from the training data\nand has maximum number of Markov chain steps K (referred as MCLV-K). We present\na MCLV-K gradient estimator (LVS-K) for RBMs and explore the correspondence and\ndifferences between LVS-K and Contrastive Divergence (CD-K), with LVS-K\nsignificantly outperforming CD-K training RBMs over the MNIST dataset,\nindicating MCLV to be a promising direction in learning generative models.\n",
"title": "From Monte Carlo to Las Vegas: Improving Restricted Boltzmann Machine Training Through Stopping Sets"
}
| null | null | null | null | true | null |
16846
| null |
Default
| null | null |
null |
{
"abstract": " We show that if a noncollapsed $CD(K,n)$ space $X$ with $n\\ge 2$ has\ncurvature bounded above by $\\kappa$ in the sense of Alexandrov then $K\\le\n(n-1)\\kappa$ and $X$ is an Alexandrov space of curvature bounded below by\n$K-\\kappa (n-2)$. We also show that if a $CD(K,n)$ space $Y$ with finite $n$\nhas curvature bounded above then it is infinitesimally Hilbertian.\n",
"title": "CD meets CAT"
}
| null | null | null | null | true | null |
16847
| null |
Default
| null | null |
null |
{
"abstract": " Data warehouse performance is usually achieved through physical data\nstructures such as indexes or materialized views. In this context, cost models\ncan help select a relevant set ofsuch performance optimization structures.\nNevertheless, selection becomes more complex in the cloud. The criterion to\noptimize is indeed at least two-dimensional, with monetary cost balancing\noverall query response time. This paper introduces new cost models that fit\ninto the pay-as-you-go paradigm of cloud computing. Based on these cost models,\nan optimization problem is defined to discover, among candidate views, those to\nbe materialized to minimize both the overall cost of using and maintaining the\ndatabase in a public cloud and the total response time ofa given query\nworkload. We experimentally show that maintaining materialized views is always\nadvantageous, both in terms of performance and cost.\n",
"title": "Cost Models for Selecting Materialized Views in Public Clouds"
}
| null | null | null | null | true | null |
16848
| null |
Default
| null | null |
null |
{
"abstract": " Bayesian optimization (BO) methods are useful for optimizing functions that\nare expensive to evaluate, lack an analytical expression and whose evaluations\ncan be contaminated by noise. These methods rely on a probabilistic model of\nthe objective function, typically a Gaussian process (GP), upon which an\nacquisition function is built. This function guides the optimization process\nand measures the expected utility of performing an evaluation of the objective\nat a new point. GPs assume continous input variables. When this is not the\ncase, such as when some of the input variables take integer values, one has to\nintroduce extra approximations. A common approach is to round the suggested\nvariable value to the closest integer before doing the evaluation of the\nobjective. We show that this can lead to problems in the optimization process\nand describe a more principled approach to account for input variables that are\ninteger-valued. We illustrate in both synthetic and a real experiments the\nutility of our approach, which significantly improves the results of standard\nBO methods on problems involving integer-valued variables.\n",
"title": "Dealing with Integer-valued Variables in Bayesian Optimization with Gaussian Processes"
}
| null | null | null | null | true | null |
16849
| null |
Default
| null | null |
null |
{
"abstract": " The aim of this work is to study, from an intrinsic and geometric point of\nview, second-order constrained variational problems on Lie algebroids, that is,\noptimization problems defined by a cost functional which depends on\nhigher-order derivatives of admissible curves on a Lie algebroid. Extending the\nclassical Skinner and Rusk formalism for the mechanics in the context of Lie\nalgebroids, for second-order constrained mechanical systems, we derive the\ncorresponding dynamical equations. We find a symplectic Lie subalgebroid where,\nunder some mild regularity conditions, the second-order constrained variational\nproblem, seen as a presymplectic Hamiltonian system, has a unique solution. We\nstudy the relationship of this formalism with the second-order constrained\nEuler-Poincaré and Lagrange-Poincaré equations, among others. Our study is\napplied to the optimal control of mechanical systems.\n",
"title": "Second-order constrained variational problems on Lie algebroids: applications to optimal control"
}
| null | null | null | null | true | null |
16850
| null |
Default
| null | null |
null |
{
"abstract": " The cosmic ray electrons measured by Voyager 1 between 3-70 MeV beyond the\nheliopause have intensities several hundred times those measured at the Earth\nby PAMELA at nearly the same energies. This paper compares this new V1 data\nwith data from the earth-orbiting PAMELA experiment up to energies greater than\n10 GeV where solar modulation effects are negligible. In this energy regime we\nassume the main parameters governing electron propagation are diffusion and\nenergy loss and we use a Monte Carlo program to describe this propagation in\nthe galaxy. To reproduce the new Voyager electron spectrum, which is E-1.3,\ntogether with that measured by PAMELA which is E-3.20 above 10 GeV, we require\na diffusion coefficient which is P 0.45 at energies above 0.5 GeV changing to a\nP-1.00 dependence at lower rigidities. The entire electron spectrum observed at\nboth V1 and PAMELA from 3 MeV to 30 GeV can then be described by a simple\nsource spectrum, dj/dP P-2.25, with a spectral exponent that is independent of\nrigidity. The change in exponent of the measured electron spectrum from -1.3 at\nlow energies to 3.2 at the highest energies can be explained by galactic\npropagation effects related to the changing dependence of the diffusion\ncoefficient below 0.5 GeV, and the increasing importance above 0.5 GV of energy\nloss from synchrotron and inverse Compton radiation, which are both E2, and\nwhich are responsible for most of the changing spectral exponent above 1.0 GV.\nAs a result of the P-1.00 dependence of the diffusion coefficient below 0.5\nGV that is required to fit the V1 electron spectrum, there is a rapid flow of\nthese low energy electrons out of the galaxy. These electrons in local IG space\nare unobservable to us at any wave length and therefore form a dark energy\ncomponent which is 100 times the electrons rest energy.\n",
"title": "The Galactic Cosmic Ray Electron Spectrum from 3 to 70 MeV Measured by Voyager 1 Beyond the Heliopause, What This Tells Us About the Propagation of Electrons and Nuclei In and Out of the Galaxy at Low Energies"
}
| null | null |
[
"Physics"
] | null | true | null |
16851
| null |
Validated
| null | null |
null |
{
"abstract": " In the following, we present example illustrative and experimental results\ncomparing fair schedulers allocating resources from multiple servers to\ndistributed application frameworks. Resources are allocated so that at least\none resource is exhausted in every server. Schedulers considered include DRF\n(DRFH) and Best-Fit DRF (BF-DRF), TSF, and PS-DSF. We also consider server\nselection under Randomized Round Robin (RRR) and based on their residual\n(unreserved) resources. In the following, we consider cases with frameworks of\nequal priority and without server-preference constraints. We first give typical\nresults of a illustrative numerical study and then give typical results of a\nstudy involving Spark workloads on Mesos which we have modified and\nopen-sourced to prototype different schedulers.\n",
"title": "Online Scheduling of Spark Workloads with Mesos using Different Fair Allocation Algorithms"
}
| null | null | null | null | true | null |
16852
| null |
Default
| null | null |
null |
{
"abstract": " For monomial special multiserial algebras, which in general are of wild\nrepresentation type, we construct radical embeddings into algebras of finite\nrepresentation type. As a consequence, we show that the representation\ndimension of monomial and self-injective special multiserial algebras is less\nor equal to three. This implies that the finitistic dimension conjecture holds\nfor all special multiserial algebras.\n",
"title": "On the representation dimension and finitistic dimension of special multiserial algebras"
}
| null | null | null | null | true | null |
16853
| null |
Default
| null | null |
null |
{
"abstract": " Context. Considering the importance of software testing to the development of\nhigh quality and reliable software systems, this paper aims to investigate how\ncan work-related factors influence the motivation of software testers. Method.\nWe applied a questionnaire that was developed using a previous theory of\nmotivation and satisfaction of software engineers to conduct a survey-based\nstudy to explore and understand how professional software testers perceive and\nvalue work-related factors that could influence their motivation at work.\nResults. With a sample of 80 software testers we observed that software testers\nare strongly motivated by variety of work, creative tasks, recognition for\ntheir work, and activities that allow them to acquire new knowledge, but in\ngeneral the social impact of this activity has low influence on their\nmotivation. Conclusion. This study discusses the difference of opinions among\nsoftware testers, regarding work-related factors that could impact their\nmotivation, which can be relevant for managers and leaders in software\nengineering practice.\n",
"title": "Would You Like to Motivate Software Testers? Ask Them How"
}
| null | null | null | null | true | null |
16854
| null |
Default
| null | null |
null |
{
"abstract": " This article provides a short review of some structural results in controlled\nsensing when the problem is formulated as a partially observed Markov decision\nprocess. In particular, monotone value functions, Blackwell dominance and\nquickest detection are described.\n",
"title": "POMDP Structural Results for Controlled Sensing"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16855
| null |
Validated
| null | null |
null |
{
"abstract": " We study a data model in which the data matrix D can be expressed as D = L +\nS + C, where L is a low rank matrix, S an element-wise sparse matrix and C a\nmatrix whose non-zero columns are outlying data points. To date, robust PCA\nalgorithms have solely considered models with either S or C, but not both. As\nsuch, existing algorithms cannot account for simultaneous element-wise and\ncolumn-wise corruptions. In this paper, a new robust PCA algorithm that is\nrobust to simultaneous types of corruption is proposed. Our approach hinges on\nthe sparse approximation of a sparsely corrupted column so that the sparse\nexpansion of a column with respect to the other data points is used to\ndistinguish a sparsely corrupted inlier column from an outlying data point. We\nalso develop a randomized design which provides a scalable implementation of\nthe proposed approach. The core idea of sparse approximation is analyzed\nanalytically where we show that the underlying ell_1-norm minimization can\nobtain the representation of an inlier in presence of sparse corruptions.\n",
"title": "Low Rank Matrix Recovery with Simultaneous Presence of Outliers and Sparse Corruption"
}
| null | null | null | null | true | null |
16856
| null |
Default
| null | null |
null |
{
"abstract": " The power sum $1^n + 2^n + \\cdots + x^n$ has been of interest to\nmathematicians since classical times. Johann Faulhaber, Jacob Bernoulli, and\nothers who followed expressed power sums as polynomials in $x$ of degree $n+1$\nwith rational coefficients. Here we consider the denominators of these\npolynomials, and prove some of their properties. A remarkable one is that such\na denominator equals $n+1$ times the squarefree product of certain primes $p$\nobeying the condition that the sum of the base-$p$ digits of $n+1$ is at least\n$p$. As an application, we derive a squarefree product formula for the\ndenominators of the Bernoulli polynomials.\n",
"title": "Power-Sum Denominators"
}
| null | null | null | null | true | null |
16857
| null |
Default
| null | null |
null |
{
"abstract": " Indexing massive data sets is extremely expensive for large scale problems.\nIn many fields, huge amounts of data are currently generated, however\nextracting meaningful information from voluminous data sets, such as computing\nsimilarity between elements, is far from being trivial. It remains nonetheless\na fundamental need. This work proposes a probabilistic data structure based on\na minimal perfect hash function for indexing large sets of keys. Our structure\nout-compete the hash table for construction, query times and for memory usage,\nin the case of the indexation of a static set. To illustrate the impact of\nalgorithms performances, we provide two applications based on similarity\ncomputation between collections of sequences, and for which this calculation is\nan expensive but required operation. In particular, we show a practical case in\nwhich other bioinformatics tools fail to scale up the tested data set or\nprovide lower recall quality results.\n",
"title": "A resource-frugal probabilistic dictionary and applications in bioinformatics"
}
| null | null | null | null | true | null |
16858
| null |
Default
| null | null |
null |
{
"abstract": " We develop a new theoretical framework to analyze the generalization error of\ndeep learning, and derive a new fast learning rate for two representative\nalgorithms: empirical risk minimization and Bayesian deep learning. The series\nof theoretical analyses of deep learning has revealed its high expressive power\nand universal approximation capability. Although these analyses are highly\nnonparametric, existing generalization error analyses have been developed\nmainly in a fixed dimensional parametric model. To compensate this gap, we\ndevelop an infinite dimensional model that is based on an integral form as\nperformed in the analysis of the universal approximation capability. This\nallows us to define a reproducing kernel Hilbert space corresponding to each\nlayer. Our point of view is to deal with the ordinary finite dimensional deep\nneural network as a finite approximation of the infinite dimensional one. The\napproximation error is evaluated by the degree of freedom of the reproducing\nkernel Hilbert space in each layer. To estimate a good finite dimensional\nmodel, we consider both of empirical risk minimization and Bayesian deep\nlearning. We derive its generalization error bound and it is shown that there\nappears bias-variance trade-off in terms of the number of parameters of the\nfinite dimensional approximation. We show that the optimal width of the\ninternal layers can be determined through the degree of freedom and the\nconvergence rate can be faster than $O(1/\\sqrt{n})$ rate which has been shown\nin the existing studies.\n",
"title": "Fast learning rate of deep learning via a kernel perspective"
}
| null | null | null | null | true | null |
16859
| null |
Default
| null | null |
null |
{
"abstract": " Closed-loop field development (CLFD) optimization is a comprehensive\nframework for optimal development of subsurface resources. CLFD involves three\nmajor steps: 1) optimization of full development plan based on current set of\nmodels, 2) drilling new wells and collecting new spatial and temporal\n(production) data, 3) model calibration based on all data. This process is\nrepeated until the optimal number of wells is drilled. This work introduces an\nefficient CLFD implementation for complex systems described by multipoint\ngeostatistics (MPS). Model calibration is accomplished in two steps:\nconditioning to spatial data by a geostatistical simulation method, and\nconditioning to production data by optimization-based PCA. A statistical\nprocedure is presented to assess the performance of CLFD. Methodology is\napplied to an oil reservoir example for 25 different true-model cases.\nApplication of a single-step of CLFD, improved the true NPV in 64%--80% of\ncases. The full CLFD procedure (with three steps) improved the true NPV in 96%\nof cases, with an average improvement of 37%.\n",
"title": "Closed-loop field development optimization with multipoint geostatistics and statistical assessment"
}
| null | null | null | null | true | null |
16860
| null |
Default
| null | null |
null |
{
"abstract": " A regular $t$-balanced Cayley map (RBCM$_t$ for short) on a group $\\Gamma$ is\nan embedding of a Cayley graph on $\\Gamma$ into a surface with some special\nsymmetric properties. We propose a reduction method to study RBCM$_t$'s, and as\na first practice, we completely classify RBCM$_t$'s for a class of split\nmetacyclic 2-groups.\n",
"title": "Reduction and regular $t$-balanced Cayley maps on split metacyclic 2-groups"
}
| null | null | null | null | true | null |
16861
| null |
Default
| null | null |
null |
{
"abstract": " Transition metal oxides are promising candidates for thermoelectric\napplications, because they are stable at high temperature and because strong\nelectronic correlations can generate large Seebeck coefficients, but their\nthermoelectric power factors are limited by the low electrical conductivity. We\nreport transport measurements on Ca3Co4O9 films on various perovskite\nsubstrates and show that reversible incorporation of oxygen into SrTiO3 and\nLaAlO3 substrates activates a parallel conduction channel for p-type carriers,\ngreatly enhancing the thermoelectric performance of the film-substrate system\nat temperatures above 450 °C. Thin-film structures that take advantage of\nboth electronic correlations and the high oxygen mobility of transition metal\noxides thus open up new perspectives for thermopower generation at high\ntemperature.\n",
"title": "Perovskite Substrates Boost the Thermopower of Cobaltate Thin Films at High Temperatures"
}
| null | null |
[
"Physics"
] | null | true | null |
16862
| null |
Validated
| null | null |
null |
{
"abstract": " We analyze the motion of a rod floating in a weightless environment in space\nwhen a force is applied at some point on the rod in a direction perpendicular\nto its length. If the force applied is at the centre of mass, then the rod gets\na linear motion perpendicular to its length. However, if the same force is\napplied at a point other than the centre of mass, say, near one end of the rod,\nthereby giving rise to a torque, then there will also be a rotation of the rod\nabout its centre of mass, in addition to the motion of the centre of mass\nitself. If the force applied is for a very short duration, but imparting\nnevertheless a finite impulse, like in a sudden (quick) hit at one end of the\nrod, then the centre of mass will move with a constant linear speed and\nsuperimposed on it will be a rotation of the rod with constant angular speed\nabout the centre of mass. However, if force is applied continuously, say by\nstrapping a tiny rocket at one end of the rod, then the rod will spin faster\nand faster about the centre of mass, with angular speed increasing linearly\nwith time. As the direction of the applied force, as seen by an external\n(inertial) observer, will be changing continuously with the rotation of the\nrod, the acceleration of the centre of mass would also be not in one fixed\ndirection. However, it turns out that the locus of the velocity vector of the\ncentre of mass will describe a Cornu spiral, with the velocity vector reaching\na final constant value with time. The mean motion of the centre of mass will be\nin a straight line, with superposed initial oscillations that soon die down.\n",
"title": "Motion of a rod pushed at one point in a weightless environment in space"
}
| null | null | null | null | true | null |
16863
| null |
Default
| null | null |
null |
{
"abstract": " We prove some basic results on the dimension theory of algebraic stacks, and\non the multiplicities of their irreducible components, for which we do not know\na reference.\n",
"title": "Dimension theory and components of algebraic stacks"
}
| null | null |
[
"Mathematics"
] | null | true | null |
16864
| null |
Validated
| null | null |
null |
{
"abstract": " Can an ideal I in a polynomial ring k[x] over a field be moved by a change of\ncoordinates into a position where it is generated by binomials $x^a - cx^b$\nwith c in k, or by unital binomials (i.e., with c = 0 or 1)? Can a variety be\nmoved into a position where it is toric? By fibering the G-translates of I over\nan algebraic group G acting on affine space, these problems are special cases\nof questions about a family F of ideals over an arbitrary base B. The main\nresults in this general setting are algorithms to find the locus of points in B\nover which the fiber of F\n- is contained in the fiber of a second family F' of ideals over B;\n- defines a variety of dimension at least d;\n- is generated by binomials; or\n- is generated by unital binomials.\nA faster containment algorithm is also presented when the fibers of F are\nprime. The big-fiber algorithm is probabilistic but likely faster than known\ndeterministic ones. Applications include the setting where a second group T\nacts on affine space, in addition to G, in which case algorithms compute the\nset of G-translates of I\n- whose stabilizer subgroups in T have maximal dimension; or\n- that admit a faithful multigrading by $Z^r$ of maximal rank r.\nEven with no ambient group action given, the final application is an\nalgorithm to\n- decide whether a normal projective variety is abstractly toric.\nAll of these loci in B and subsets of G are constructible; in some cases they\nare closed.\n",
"title": "When is a polynomial ideal binomial after an ambient automorphism?"
}
| null | null | null | null | true | null |
16865
| null |
Default
| null | null |
null |
{
"abstract": " It has been shown by McCoy that a right ideal of a polynomial ring with\nseveral indeterminates has a non-trivial homogeneous right annihilator of\ndegree 0 provided its right annihilator is non-trivial to begin with. In this\nnote, it is documented that any $\\mathbb{N}$-graded ring $R$ has a slightly\nweaker property: the right annihilator of a right ideal contains a homogeneous\nnon-zero element, if it is non-trivial to begin with. If $R$ is a subring of a\n$\\mathbb{Z}^k$ -graded ring $S$ satisfying a certain non-annihilation property\n(which is the case if $S$ is strongly graded, for example), then it is possible\nto find annihilators of degree 0.\n",
"title": "Annihilators in $\\mathbb{N}^k$-graded and $\\mathbb{Z}^k$-graded rings"
}
| null | null | null | null | true | null |
16866
| null |
Default
| null | null |
null |
{
"abstract": " We propose a new generic type of stochastic neurons, called $q$-neurons, that\nconsiders activation functions based on Jackson's $q$-derivatives with\nstochastic parameters $q$. Our generalization of neural network architectures\nwith $q$-neurons is shown to be both scalable and very easy to implement. We\ndemonstrate experimentally consistently improved performances over\nstate-of-the-art standard activation functions, both on training and testing\nloss functions.\n",
"title": "q-Neurons: Neuron Activations based on Stochastic Jackson's Derivative Operators"
}
| null | null |
[
"Statistics"
] | null | true | null |
16867
| null |
Validated
| null | null |
null |
{
"abstract": " Randomly generated programs are popular for testing compilers and program\nanalysis tools, with hundreds of bugs in real-world C compilers found by random\ntesting. However, existing random program generators may generate large amounts\nof dead code (computations whose result is never used). This leaves relatively\nlittle code to exercise a target compiler's more complex optimizations.\nTo address this shortcoming, we introduce liveness-driven random program\ngeneration. In this approach the random program is constructed bottom-up,\nguided by a simultaneous structural data-flow analysis to ensure that the\ngenerator never generates dead code.\nThe algorithm is implemented as a plugin for the Frama-C framework. We\nevaluate it in comparison to Csmith, the standard random C program generator.\nOur tool generates programs that compile to more machine code with a more\ncomplex instruction mix.\n",
"title": "Liveness-Driven Random Program Generation"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16868
| null |
Validated
| null | null |
null |
{
"abstract": " Pre-exposure prophylaxis (PrEP) consists in the use of an antiretroviral\nmedication to prevent the acquisition of HIV infection by uninfected\nindividuals and has recently demonstrated to be highly efficacious for HIV\nprevention. We propose a new epidemiological model for HIV/AIDS transmission\nincluding PrEP. Existence, uniqueness and global stability of the disease free\nand endemic equilibriums are proved. The model with no PrEP is calibrated with\nthe cumulative cases of infection by HIV and AIDS reported in Cape Verde from\n1987 to 2014, showing that it predicts well such reality. An optimal control\nproblem with a mixed state control constraint is then proposed and analyzed,\nwhere the control function represents the PrEP strategy and the mixed\nconstraint models the fact that, due to PrEP costs, epidemic context and\nprogram coverage, the number of individuals under PrEP is limited at each\ninstant of time. The objective is to determine the PrEP strategy that satisfies\nthe mixed state control constraint and minimizes the number of individuals with\npre-AIDS HIV-infection as well as the costs associated with PrEP. The optimal\ncontrol problem is studied analytically. Through numerical simulations, we\ndemonstrate that PrEP reduces HIV transmission significantly.\n",
"title": "Modeling and optimal control of HIV/AIDS prevention through PrEP"
}
| null | null | null | null | true | null |
16869
| null |
Default
| null | null |
null |
{
"abstract": " Based on the observation that the correlation between observed traffic at two\nmeasurement points or traffic stations may be time-varying, attributable to the\ntime-varying speed which subsequently causes variations in the time required to\ntravel between the two points, in this paper, we develop a modified Space-Time\nAutoregressive Integrated Moving Average (STARIMA) model with time-varying lags\nfor short-term traffic flow prediction. Particularly, the temporal lags in the\nmodified STARIMA change with the time-varying speed at different time of the\nday or equivalently change with the (time-varying) time required to travel\nbetween two measurement points. Firstly, a technique is developed to evaluate\nthe temporal lag in the STARIMA model, where the temporal lag is formulated as\na function of the spatial lag (spatial distance) and the average speed.\nSecondly, an unsupervised classification algorithm based on ISODATA algorithm\nis designed to classify different time periods of the day according to the\nvariation of the speed. The classification helps to determine the appropriate\ntime lag to use in the STARIMA model. Finally, a STARIMA-based model with\ntime-varying lags is developed for short-term traffic prediction. Experimental\nresults using real traffic data show that the developed STARIMA-based model\nwith time-varying lags has superior accuracy compared with its counterpart\ndeveloped using the traditional cross-correlation function and without\nemploying time-varying lags.\n",
"title": "STARIMA-based Traffic Prediction with Time-varying Lags"
}
| null | null | null | null | true | null |
16870
| null |
Default
| null | null |
null |
{
"abstract": " For accommodating more electric vehicles (EVs) to battle against fossil fuel\nemission, the problem of charging station placement is inevitable and could be\ncostly if done improperly. Some researches consider a general setup, using\nconditions such as driving ranges for planning. However, most of the EV growths\nin the next decades will happen in the urban area, where driving ranges is not\nthe biggest concern. For such a need, we consider several practical aspects of\nurban systems, such as voltage regulation cost and protection device upgrade\nresulting from the large integration of EVs. Notably, our diversified objective\ncan reveal the trade-off between different factors in different cities\nworldwide. To understand the global optimum of large-scale analysis, we add\nconstraint one-by-one to see how to preserve the problem convexity. Our\nsensitivity analysis before and after convexification shows that our approach\nis not only universally applicable but also has a small approximation error for\nprioritizing the most urgent constraint in a specific setup. Finally, numerical\nresults demonstrate the trade-off, the relationship between different factors\nand the global objective, and the small approximation error. A unique\nobservation in this study shows the importance of incorporating the protection\ndevice upgrade in urban system planning on charging stations.\n",
"title": "Electric Vehicle Charging Station Placement Method for Urban Areas"
}
| null | null | null | null | true | null |
16871
| null |
Default
| null | null |
null |
{
"abstract": " Let G be a reductive algebraic group over a field of positive characteristic\nand denote by C(G) the category of rational G-modules. In this note we\ninvestigate the subcategory of C(G) consisting of those modules whose\ncomposition factors all have highest weights linked to the Steinberg weight.\nThis subcategory is denoted ST and called the Steinberg component. We give an\nexplicit equivalence between ST and C(G) and we derive some consequences. In\nparticular, our result allows us to relate the Frobenius contracting functor to\nthe projection functor from C(G) onto ST .\n",
"title": "The Steinberg linkage class for a reductive algebraic group"
}
| null | null |
[
"Mathematics"
] | null | true | null |
16872
| null |
Validated
| null | null |
null |
{
"abstract": " Despite the increasing use of social media platforms for information and news\ngathering, its unmoderated nature often leads to the emergence and spread of\nrumours, i.e. pieces of information that are unverified at the time of posting.\nAt the same time, the openness of social media platforms provides opportunities\nto study how users share and discuss rumours, and to explore how natural\nlanguage processing and data mining techniques may be used to find ways of\ndetermining their veracity. In this survey we introduce and discuss two types\nof rumours that circulate on social media; long-standing rumours that circulate\nfor long periods of time, and newly-emerging rumours spawned during fast-paced\nevents such as breaking news, where reports are released piecemeal and often\nwith an unverified status in their early stages. We provide an overview of\nresearch into social media rumours with the ultimate goal of developing a\nrumour classification system that consists of four components: rumour\ndetection, rumour tracking, rumour stance classification and rumour veracity\nclassification. We delve into the approaches presented in the scientific\nliterature for the development of each of these four components. We summarise\nthe efforts and achievements so far towards the development of rumour\nclassification systems and conclude with suggestions for avenues for future\nresearch in social media mining for detection and resolution of rumours.\n",
"title": "Detection and Resolution of Rumours in Social Media: A Survey"
}
| null | null | null | null | true | null |
16873
| null |
Default
| null | null |
null |
{
"abstract": " This paper contains a non-trivial generalization of the Harish-Chandra\ntransforms on a connected semisimple Lie group $G,$ with finite center, into\nwhat we term spherical convolutions. Among other results we show that its\nintegral over the collection of bounded spherical functions at the identity\nelement $e \\in G$ is a weighted Fourier transforms of the Abel transform at\n$0.$ Being a function on $G,$ the restriction of this integral of its spherical\nFourier transforms to the positive-definite spherical functions is then shown\nto be (the non-zero constant multiple of) a positive-definite distribution on\n$G,$ which is tempered and invariant on $G=SL(2,\\mathbb{R}).$ These results\nsuggest the consideration of a calculus on the Schwartz algebras of spherical\nfunctions. The Plancherel measure of the spherical convolutions is also\nexplicitly computed.\n",
"title": "On harmonic analysis of spherical convolutions on semisimple Lie groups"
}
| null | null | null | null | true | null |
16874
| null |
Default
| null | null |
null |
{
"abstract": " Magnetic Particle Imaging (MPI) has been shown to provide remarkable contrast\nfor imaging applications such as angiography, stem cell tracking, and cancer\nimaging. Recently, there is growing interest in the functional imaging\ncapabilities of MPI, where color MPI techniques have explored separating\ndifferent nanoparticles, which could potentially be used to distinguish\nnanoparticles in different states or environments. Viscosity mapping is a\npromising functional imaging application for MPI, as increased viscosity levels\nin vivo have been associated with numerous diseases such as hypertension,\natherosclerosis, and cancer. In this work, we propose a viscosity mapping\ntechnique for MPI through the estimation of the relaxation time constant of the\nnanoparticles. Importantly, the proposed time constant estimation scheme does\nnot require any prior information regarding the nanoparticles. We validate this\nmethod with extensive experiments in an in-house magnetic particle spectroscopy\n(MPS) setup at four different frequencies (between 250 Hz and 10.8 kHz) and at\nthree different field strengths (between 5 mT and 15 mT) for viscosities\nranging between 0.89 mPa.s to 15.33 mPa.s. Our results demonstrate the\nviscosity mapping ability of MPI in the biologically relevant viscosity range.\n",
"title": "Relaxation-based viscosity mapping for magnetic particle imaging"
}
| null | null | null | null | true | null |
16875
| null |
Default
| null | null |
null |
{
"abstract": " Community detection is a key data analysis problem across different fields.\nDuring the past decades, numerous algorithms have been proposed to address this\nissue. However, most work on community detection does not address the issue of\nstatistical significance. Although some research efforts have been made towards\nmining statistically significant communities, deriving an analytical solution\nof p-value for one community under the configuration model is still a\nchallenging mission that remains unsolved. To partially fulfill this void, we\npresent a tight upper bound on the p-value of a single community under the\nconfiguration model, which can be used for quantifying the statistical\nsignificance of each community analytically. Meanwhile, we present a local\nsearch method to detect statistically significant communities in an iterative\nmanner. Experimental results demonstrate that our method is comparable with the\ncompeting methods on detecting statistically significant communities.\n",
"title": "Detecting Statistically Significant Communities"
}
| null | null | null | null | true | null |
16876
| null |
Default
| null | null |
null |
{
"abstract": " Let k be an infinite perfect field. We provide a general criterion for a\nspectrum in the stable homotopy category over k to be effective, i.e. to be in\nthe localizing subcategory generated by the suspension spectra of smooth\nschemes. As a consequence, we show that two recent versions of generalized\nmotivic cohomology theories coincide.\n",
"title": "On the effectivity of spectra representing motivic cohomology theories"
}
| null | null | null | null | true | null |
16877
| null |
Default
| null | null |
null |
{
"abstract": " Momentum is a simple and widely used trick which allows gradient-based\noptimizers to pick up speed along low curvature directions. Its performance\ndepends crucially on a damping coefficient $\\beta$. Large $\\beta$ values can\npotentially deliver much larger speedups, but are prone to oscillations and\ninstability; hence one typically resorts to small values such as 0.5 or 0.9. We\npropose Aggregated Momentum (AggMo), a variant of momentum which combines\nmultiple velocity vectors with different $\\beta$ parameters. AggMo is trivial\nto implement, but significantly dampens oscillations, enabling it to remain\nstable even for aggressive $\\beta$ values such as 0.999. We reinterpret\nNesterov's accelerated gradient descent as a special case of AggMo and analyze\nrates of convergence for quadratic objectives. Empirically, we find that AggMo\nis a suitable drop-in replacement for other momentum methods, and frequently\ndelivers faster convergence.\n",
"title": "Aggregated Momentum: Stability Through Passive Damping"
}
| null | null | null | null | true | null |
16878
| null |
Default
| null | null |
null |
{
"abstract": " We investigate the equilibrium behavior for the decentralized quadratic cheap\ntalk problem in which an encoder and a decoder, viewed as two decision makers,\nhave misaligned objective functions. In prior work, we have shown that the\nnumber of bins under any equilibrium has to be at most countable, generalizing\na classical result due to Crawford and Sobel who considered sources with\ndensity supported on $[0,1]$. In this paper, we refine this result in the\ncontext of exponential and Gaussian sources. For exponential sources, a\nrelation between the upper bound on the number of bins and the misalignment in\nthe objective functions is derived, the equilibrium costs are compared, and it\nis shown that there also exist equilibria with infinitely many bins under\ncertain parametric assumptions. For Gaussian sources, it is shown that there\nexist equilibria with infinitely many bins.\n",
"title": "On the Number of Bins in Equilibria for Signaling Games"
}
| null | null | null | null | true | null |
16879
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, a linear model of diffusion processes with unknown drift and\ndiagonal diffusion matrices is discussed. We will consider the estimation\nproblems for unknown parameters based on the discrete time observation in\nhigh-dimensional and sparse settings. To estimate drift matrices, the Dantzig\nselector which was proposed by Candés and Tao in 2007 will be applied. Then,\nwe will prove two types of consistency of the estimator of drift matrix; one is\nthe consistency in the sense of $l_q$ norm for every $q \\in [1,\\infty]$ and the\nother is the variable selection consistency. Moreover, we will construct an\nasymptotically normal estimator of the drift matrix by using the variable\nselection consistency of the Dantzig selector.\n",
"title": "The Dantzig selector for a linear model of diffusion processes"
}
| null | null | null | null | true | null |
16880
| null |
Default
| null | null |
null |
{
"abstract": " Among the proposals for joint disease mapping, the shared component model has\nbecome more popular. Another recent advance to strengthen inference of disease\ndata has been the extension of purely spatial models to include time and\nspace-time interaction. Such analyses have additional benefits over purely\nspatial models. However, only a few proposed spatio-temporal models could\naddress analysing multiple diseases jointly.\nIn the proposed model, each component is shared by different subsets of\ndiseases, spatial and temporal trends are considered for each component, and\nthe relative weight of these trends for each component for each relevant\ndisease can be estimated. We present an application of the proposed method on\nincidence rates of seven prevalent cancers in Iran. The effect of the shared\ncomponents on the individual cancer types can be identified. Regional and\ntemporal variation in relative risks is shown. We present a model which\ncombines the benefits of shared-components with spatio-temporal techniques for\nmultivariate data. We show, how the model allows to analyse geographical and\ntemporal variation among diseases beyond previous approaches.\n",
"title": "A Spatio-Temporal Multivariate Shared Component Model with an Application in Iran Cancer Data"
}
| null | null |
[
"Statistics"
] | null | true | null |
16881
| null |
Validated
| null | null |
null |
{
"abstract": " We show that in decaying hydromagnetic turbulence with initial kinetic\nhelicity, a weak magnetic field eventually becomes fully helical. The sign of\nmagnetic helicity is opposite to that of the kinetic helicity - regardless of\nwhether or not the initial magnetic field was helical. The magnetic field\nundergoes inverse cascading with the magnetic energy decaying approximately\nlike t^{-1/2}. This is even slower than in the fully helical case, where it\ndecays like t^{-2/3}. In this parameter range, the product of magnetic energy\nand correlation length raised to a certain power slightly larger than unity, is\napproximately constant. This scaling of magnetic energy persists over long time\nscales. At very late times and for domain sizes large enough to accommodate the\ngrowing spatial scales, we expect a cross-over to the t^{-2/3} decay law that\nis commonly observed for fully helical magnetic fields. Regardless of the\npresence or absence of initial kinetic helicity, the magnetic field experiences\nexponential growth during the first few turnover times, which is suggestive of\nsmall-scale dynamo action. Our results have applications to a wide range of\nexperimental dynamos and astrophysical time-dependent plasmas, including\nprimordial turbulence in the early universe.\n",
"title": "The dynamo effect in decaying helical turbulence"
}
| null | null |
[
"Physics"
] | null | true | null |
16882
| null |
Validated
| null | null |
null |
{
"abstract": " We prove that, for any two finite volume hyperbolic $3$-manifolds, the\namalgamation of their fundamental groups along any nontrivial geometrically\nfinite subgroup is not LERF. This generalizes the author's previous work on\nnonLERFness of amalgamations of hyperbolic $3$-manifold groups along abelian\nsubgroups. A consequence of this result is that closed arithmetic hyperbolic\n$4$-manifolds have nonLERF fundamental groups. Along with the author's previous\nwork, we get that, for any arithmetic hyperbolic manifold with dimension at\nleast $4$, with possible exceptions in $7$-dimensional manifolds defined by the\noctonion, its fundamental group is not LERF.\n",
"title": "Geometrically finite amalgamations of hyperbolic 3-manifold groups are not LERF"
}
| null | null | null | null | true | null |
16883
| null |
Default
| null | null |
null |
{
"abstract": " We provide the first quantum (exact) protocol for the Dining Philosophers\nproblem (DP), a central problem in distributed algorithms. It is well known\nthat the problem cannot be solved exactly in the classical setting. We then use\nour DP protocol to provide a new quantum protocol for the tightly related\nproblem of exact leader election (LE) on a ring, improving significantly in\nboth time and memory complexity over the known LE protocol by Tani et. al. To\ndo this, we show that in some sense the exact DP and exact LE problems are\nequivalent; interestingly, in the classical non-exact setting they are not.\nHopefully, the results will lead to exact quantum protocols for other important\ndistributed algorithmic questions; in particular, we discuss interesting\nconnections to the ring size problem, as well as to a physically motivated\nquestion of breaking symmetry in 1D translationally invariant systems.\n",
"title": "Dining Philosophers, Leader Election and Ring Size problems, in the quantum setting"
}
| null | null | null | null | true | null |
16884
| null |
Default
| null | null |
null |
{
"abstract": " Fog computing is seen as a promising approach to perform distributed,\nlow-latency computation for supporting Internet of Things applications.\nHowever, due to the unpredictable arrival of available neighboring fog nodes,\nthe dynamic formation of a fog network can be challenging. In essence, a given\nfog node must smartly select the set of neighboring fog nodes that can provide\nlow-latency computations. In this paper, this problem of fog network formation\nand task distribution is studied considering a hybrid cloud-fog architecture.\nThe goal of the proposed framework is to minimize the maximum computational\nlatency by enabling a given fog node to form a suitable fog network, under\nuncertainty on the arrival process of neighboring fog nodes. To solve this\nproblem, a novel approach based on the online secretary framework is proposed.\nTo find the desired set of neighboring fog nodes, an online algorithm is\ndeveloped to enable a task initiating fog node to decide on which other nodes\ncan be used as part of its fog network, to offload computational tasks, without\nknowing any prior information on the future arrivals of those other nodes.\nSimulation results show that the proposed online algorithm can successfully\nselect an optimal set of neighboring fog nodes while achieving a latency that\nis as small as the one resulting from an ideal, offline scheme that has\ncomplete knowledge of the system. The results also show how, using the proposed\napproach, the computational tasks can be properly distributed between the fog\nnetwork and a remote cloud server.\n",
"title": "An Online Secretary Framework for Fog Network Formation with Minimal Latency"
}
| null | null | null | null | true | null |
16885
| null |
Default
| null | null |
null |
{
"abstract": " We present a computer-assisted proof of heteroclinic connections in the\none-dimensional Ohta-Kawasaki model of diblock copolymers. The model is a\nfourth-order parabolic partial differential equation subject to homogeneous\nNeumann boundary conditions, which contains as a special case the celebrated\nCahn-Hilliard equation. While the attractor structure of the latter model is\ncompletely understood for one-dimensional domains, the diblock copolymer\nextension exhibits considerably richer long-term dynamical behavior, which\nincludes a high level of multistability. In this paper, we establish the\nexistence of certain heteroclinic connections between the homogeneous\nequilibrium state, which represents a perfect copolymer mixture, and all local\nand global energy minimizers. In this way, we show that not every solution\noriginating near the homogeneous state will converge to the global energy\nminimizer, but rather is trapped by a stable state with higher energy. This\nphenomenon can not be observed in the one-dimensional Cahn-Hillard equation,\nwhere generic solutions are attracted by a global minimizer.\n",
"title": "Computer-assisted proof of heteroclinic connections in the one-dimensional Ohta-Kawasaki model"
}
| null | null | null | null | true | null |
16886
| null |
Default
| null | null |
null |
{
"abstract": " We present millimetre dust emission measurements of two Lyman Break Galaxies\nat z~3 and construct for the first time fully sampled infrared spectral energy\ndistributions (SEDs), from mid-IR to the Rayleigh-Jeans tail, of individually\ndetected, unlensed, UV-selected, main sequence (MS) galaxies at $z=3$. The SED\nmodelling of the two sources confirms previous findings, based on stacked\nensembles, of an increasing mean radiation field <U> with redshift, consistent\nwith a rapidly decreasing gas metallicity in z > 2 galaxies. Complementing our\nstudy with CO[3-2] emission line observations, we measure the molecular gas\nmass (M_H2) reservoir of the systems using three independent approaches: 1) CO\nline observations, 2) the dust to gas mass ratio vs metallicity relation and 3)\na single band, dust emission flux on the Rayleigh-Jeans side of the SED. All\ntechniques return consistent M_H2 estimates within a factor of ~2 or less,\nyielding gas depletion time-scales (tau_dep ~ 0.35 Gyrs) and gas-to-stellar\nmass ratios (M_H2/M* ~ 0.5-1) for our z~3 massive MS galaxies. The overall\nproperties of our galaxies are consistent with trends and relations established\nat lower redshifts, extending the apparent uniformity of star-forming galaxies\nover the last 11.5 billion years.\n",
"title": "Dust and Gas in Star Forming Galaxies at z~3 - Extending Galaxy Uniformity to 11.5 Billion Years"
}
| null | null | null | null | true | null |
16887
| null |
Default
| null | null |
null |
{
"abstract": " Experiments are reported on the performance of a pitching and heaving\ntwo-dimensional foil in a water channel in either continuous or intermittent\nmotion. We find that the thrust and power are independent of the mean\nfreestream velocity for two-fold changes in the mean velocity (four-fold in the\ndynamic pressure), and for oscillations in the velocity up to 38\\% of the mean,\nwhere the oscillations are intended to mimic those of freely swimming motions\nwhere the thrust varies during the flapping cycle. We demonstrate that the\ncorrect velocity scale is not the flow velocity but the mean velocity of the\ntrailing edge. We also find little or no impact of streamwise velocity change\non the wake characteristics such as vortex organization, vortex strength, and\ntime-averaged velocity profile development---the wake is both qualitatively and\nquantitatively unchanged. Our results suggest that constant velocity studies\ncan be used to make robust conclusions about swimming performance without a\nneed to explore the free-swimming condition.\n",
"title": "Flow speed has little impact on propulsive characteristics of oscillating foils"
}
| null | null |
[
"Physics"
] | null | true | null |
16888
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper, we consider the problem of attack-resilient state estimation,\nthat is to reliably estimate the true system states despite two classes of\nattacks: (i) attacks on the switching mechanisms and (ii) false data injection\nattacks on actuator and sensor signals, in the presence of unbounded stochastic\nprocess and measurement noise signals. We model the systems under attack as\nhidden mode stochastic switched linear systems with unknown inputs and propose\nthe use of a multiple-model inference algorithm to tackle these security\nissues. Moreover, we characterize fundamental limitations to resilient\nestimation (e.g., upper bound on the number of tolerable signal attacks) and\ndiscuss the topics of attack detection, identification and mitigation under\nthis framework. Simulation examples of switching and false data injection\nattacks on a benchmark system and an IEEE 68-bus test system show the efficacy\nof our approach to recover resilient (i.e., asymptotically unbiased) state\nestimates as well as to identify and mitigate the attacks.\n",
"title": "Switching and Data Injection Attacks on Stochastic Cyber-Physical Systems: Modeling, Resilient Estimation and Attack Mitigation"
}
| null | null | null | null | true | null |
16889
| null |
Default
| null | null |
null |
{
"abstract": " There has been a recent explosion in applications for dialogue interaction\nranging from direction-giving and tourist information to interactive story\nsystems. Yet the natural language generation (NLG) component for many of these\nsystems remains largely handcrafted. This limitation greatly restricts the\nrange of applications; it also means that it is impossible to take advantage of\nrecent work in expressive and statistical language generation that can\ndynamically and automatically produce a large number of variations of given\ncontent. We propose that a solution to this problem lies in new methods for\ndeveloping language generation resources. We describe the ES-Translator, a\ncomputational language generator that has previously been applied only to\nfables, and quantitatively evaluate the domain independence of the EST by\napplying it to personal narratives from weblogs. We then take advantage of\nrecent work on language generation to create a parameterized sentence planner\nfor story generation that provides aggregation operations, variations in\ndiscourse and in point of view. Finally, we present a user evaluation of\ndifferent personal narrative retellings.\n",
"title": "Generating Sentence Planning Variations for Story Telling"
}
| null | null | null | null | true | null |
16890
| null |
Default
| null | null |
null |
{
"abstract": " Aims. The purpose of this paper is to detect and investigate the nature of\nlong-term radial velocity (RV) variations of K-type giants and to confirm\nplanetary companions around the stars.\nMethods. We have conducted two planet search programs by precise RV\nmeasurement using the 1.8 m telescope at Bohyunsan Optical Astronomy\nObservatory (BOAO) and the 1.88 m telescope at Okayama Astrophysical\nObservatory (OAO). The BOAO program searches for planets around 55 early K\ngiants. The OAO program is looking for 190 G-K type giants.\nResults. In this paper, we report the detection of long-period RV variations\nof three K giant stars, HD 40956, HD 111591, and HD 113996. We investigated the\ncause of the observed RV variations and conclude the substellar companions are\nmost likely the cause of the RV variations. The orbital analyses yield P =\n578.6 $\\pm$ 3.3 d, $m$ sin $i$ = 2.7 $\\pm$ 0.6 $M_{\\rm{J}}$, $a$ = 1.4 $\\pm$\n0.1 AU for HD 40956; P = 1056.4 $\\pm$ 14.3 d, $m$ sin $i$ = 4.4 $\\pm$ 0.4\n$M_{\\rm{J}}$, $a$ = 2.5 $\\pm$ 0.1 AU for HD 111591; P = 610.2 $\\pm$ 3.8 d, $m$\nsin $i$ = 6.3 $\\pm$ 1.0 $M_{\\rm{J}}$, $a$ = 1.6 $\\pm$ 0.1 AU for HD 113996.\n",
"title": "Detection of planet candidates around K giants, HD 40956, HD 111591, and HD 113996"
}
| null | null | null | null | true | null |
16891
| null |
Default
| null | null |
null |
{
"abstract": " We introduce perfect half space games, in which the goal of Player 2 is to\nmake the sums of encountered multi-dimensional weights diverge in a direction\nwhich is consistent with a chosen sequence of perfect half spaces (chosen\ndynamically by Player 2). We establish that the bounding games of Jurdziński\net al. (ICALP 2015) can be reduced to perfect half space games, which in turn\ncan be translated to the lexicographic energy games of Colcombet and\nNiwiński, and are positionally determined in a strong sense (Player 2 can\nplay without knowing the current perfect half space). We finally show how\nperfect half space games and bounding games can be employed to solve\nmulti-dimensional energy parity games in pseudo-polynomial time when both the\nnumbers of energy dimensions and of priorities are fixed, regardless of whether\nthe initial credit is given as part of the input or existentially quantified.\nThis also yields an optimal 2-EXPTIME complexity with given initial credit,\nwhere the best known upper bound was non-elementary.\n",
"title": "Perfect Half Space Games"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16892
| null |
Validated
| null | null |
null |
{
"abstract": " The research challenge of current Wireless Sensor Networks~(WSNs) is to\ndesign energy-efficient, low-cost, high-accuracy, self-healing, and scalable\nsystems for applications such as environmental monitoring. Traditional WSNs\nconsist of low density, power-hungry digital motes that are expensive and\ncannot remain functional for long periods on a single charge. In order to\naddress these challenges, a \\textit{dumb-sensing and smart-processing}\narchitecture that splits sensing and computation capabilities among tiers is\nproposed. Tier-1 consists of dumb sensors that only sense and transmit, while\nthe nodes in Tier-2 do all the smart processing on Tier-1 sensor data. A\nlow-power and low-cost solution for Tier-1 sensors has been proposed using\nAnalog Joint Source Channel Coding~(AJSCC). An analog circuit that realizes the\nrectangular type of AJSCC has been proposed and realized on a Printed Circuit\nBoard for feasibility analysis. A prototype consisting of three Tier-1 sensors\n(sensing temperature and humidity) communicating to a Tier-2 Cluster Head has\nbeen demonstrated to verify the proposed approach. Results show that our\nframework is indeed feasible to support large scale high density and persistent\nWSN deployment.\n",
"title": "Energy-efficient Analog Sensing for Large-scale, High-density Persistent Wireless Monitoring"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16893
| null |
Validated
| null | null |
null |
{
"abstract": " We address the computation of ground-state properties of chemical systems and\nrealistic materials within the auxiliary-field quantum Monte Carlo method. The\nphase constraint to control the fermion phase problem requires the random walks\nin Slater determinant space to be open-ended with branching. This in turn makes\nit necessary to use back-propagation (BP) to compute averages and correlation\nfunctions of operators that do not commute with the Hamiltonian. Several BP\nschemes are investigated and their optimization with respect to the phaseless\nconstraint is considered. We propose a modified BP method for the computation\nof observables in electronic systems, discuss its numerical stability and\ncomputational complexity, and assess its performance by computing ground-state\nproperties for several substances, including constituents of the primordial\nterrestrial atmosphere and small organic molecules.\n",
"title": "Computation of ground-state properties in molecular systems: back-propagation with auxiliary-field quantum Monte Carlo"
}
| null | null | null | null | true | null |
16894
| null |
Default
| null | null |
null |
{
"abstract": " Inverse Uncertainty Quantification (UQ), or Bayesian calibration, is the\nprocess to quantify the uncertainties of random input parameters based on\nexperimental data. The introduction of model discrepancy term is significant\nbecause \"over-fitting\" can theoretically be avoided. But it also poses\nchallenges in the practical applications. One of the mostly concerned and\nunresolved problem is the \"lack of identifiability\" issue. With the presence of\nmodel discrepancy, inverse UQ becomes \"non-identifiable\" in the sense that it\nis difficult to precisely distinguish between the parameter uncertainties and\nmodel discrepancy when estimating the calibration parameters. Previous research\nto alleviate the non-identifiability issue focused on using informative priors\nfor the calibration parameters and the model discrepancy, which is usually not\na viable solution because one rarely has such accurate and informative prior\nknowledge. In this work, we show that identifiability is largely related to the\nsensitivity of the calibration parameters with regards to the chosen responses.\nWe adopted an improved modular Bayesian approach for inverse UQ that does not\nrequire priors for the model discrepancy term. The relationship between\nsensitivity and identifiability was demonstrated with a practical example in\nnuclear engineering. It was shown that, in order for a certain calibration\nparameter to be statistically identifiable, it should be significant to at\nleast one of the responses whose data are used for inverse UQ. Good\nidentifiability cannot be achieved for a certain calibration parameter if it is\nnot significant to any of the responses. It is also demonstrated that \"fake\nidentifiability\" is possible if model responses are not appropriately chosen,\nor inaccurate but informative priors are specified.\n",
"title": "Demonstration of the Relationship between Sensitivity and Identifiability for Inverse Uncertainty Quantification"
}
| null | null | null | null | true | null |
16895
| null |
Default
| null | null |
null |
{
"abstract": " Mendelian randomization (MR) is a popular instrumental variable (IV)\napproach. A key IV identification condition known as the exclusion restriction\nrequires no direct effect of an IV on the outcome not through the exposure\nwhich is unrealistic in most MR analyses. As a result, possible violation of\nthe exclusion restriction can seldom be ruled out in such studies. To address\nthis concern, we introduce a new class of IV estimators which are robust to\nviolation of the exclusion restriction under a large collection of data\ngenerating mechanisms consistent with parametric models commonly assumed in the\nMR literature. Our approach named \"MR G-Estimation under No Interaction with\nUnmeasured Selection\" (MR GENIUS) may be viewed as a modification to Robins'\nG-estimation approach that is robust to both additive unmeasured confounding\nand violation of the exclusion restriction assumption. We also establish that\nestimation with MR GENIUS may also be viewed as a robust generalization of the\nwell-known Lewbel estimator for a triangular system of structural equations\nwith endogeneity. Specifically, we show that unlike Lewbel estimation, MR\nGENIUS is under fairly weak conditions also robust to unmeasured confounding of\nthe effects of the genetic IVs, another possible violation of a key IV\nIdentification condition. Furthermore, while Lewbel estimation involves\nspecification of linear models both for the outcome and the exposure, MR GENIUS\ngenerally does not require specification of a structural model for the direct\neffect of invalid IVs on the outcome, therefore allowing the latter model to be\nunrestricted. Finally, unlike Lewbel estimation, MR GENIUS is shown to equally\napply for binary, discrete or continuous exposure and outcome variables and can\nbe used under prospective sampling, or retrospective sampling such as in a\ncase-control study.\n",
"title": "The GENIUS Approach to Robust Mendelian Randomization Inference"
}
| null | null | null | null | true | null |
16896
| null |
Default
| null | null |
null |
{
"abstract": " Trace alignment algorithms have been used in process mining for discovering\nthe consensus treatment procedures and process deviations. Different alignment\nalgorithms, however, may produce very different results. No widely-adopted\nmethod exists for evaluating the results of trace alignment. Existing\nreference-free evaluation methods cannot adequately and comprehensively assess\nthe alignment quality. We analyzed and compared the existing evaluation\nmethods, identifying their limitations, and introduced improvements in two\nreference-free evaluation methods. Our approach assesses the alignment result\nglobally instead of locally, and therefore helps the algorithm to optimize\noverall alignment quality. We also introduced a novel metric to measure the\nalignment complexity, which can be used as a constraint on alignment algorithm\noptimization. We tested our evaluation methods on a trauma resuscitation\ndataset and provided the medical explanation of the activities and patterns\nidentified as deviations using our proposed evaluation methods.\n",
"title": "Evaluation of Trace Alignment Quality and its Application in Medical Process Mining"
}
| null | null |
[
"Computer Science"
] | null | true | null |
16897
| null |
Validated
| null | null |
null |
{
"abstract": " Topological insulator surfaces in proximity to superconductors have been\nproposed as a way to produce Majorana fermions in condensed matter physics. One\nof the simplest proposed experiments with such a system is Majorana\ninterferometry. Here, we consider two possibly conflicting constraints on the\nsize of such an interferometer. Coupling of a Majorana mode from the edge (the\narms) of the interferometer to vortices in the centre of the device sets a\nlower bound on the size of the device. On the other hand, scattering to the\nusually imperfectly insulating bulk sets an upper bound. From estimates of\nexperimental parameters, we find that typical samples may have no size window\nin which the Majorana interferometer can operate, implying that a new\ngeneration of more highly insulating samples must be explored.\n",
"title": "Size Constraints on Majorana Beamsplitter Interferometer: Majorana Coupling and Surface-Bulk Scattering"
}
| null | null | null | null | true | null |
16898
| null |
Default
| null | null |
null |
{
"abstract": " Let $G$ be a finite, simple, connected graph. An arithmetical structure on\n$G$ is a pair of positive integer vectors $\\mathbf{d},\\mathbf{r}$ such that\n$(\\mathrm{diag}(\\mathbf{d})-A)\\mathbf{r}=0$, where $A$ is the adjacency matrix\nof $G$. We investigate the combinatorics of arithmetical structures on path and\ncycle graphs, as well as the associated critical groups (the cokernels of the\nmatrices $(\\mathrm{diag}(\\mathbf{d})-A)$). For paths, we prove that\narithmetical structures are enumerated by the Catalan numbers, and we obtain\nrefined enumeration results related to ballot sequences. For cycles, we prove\nthat arithmetical structures are enumerated by the binomial coefficients\n$\\binom{2n-1}{n-1}$, and we obtain refined enumeration results related to\nmultisets. In addition, we determine the critical groups for all arithmetical\nstructures on paths and cycles.\n",
"title": "Counting Arithmetical Structures on Paths and Cycles"
}
| null | null | null | null | true | null |
16899
| null |
Default
| null | null |
null |
{
"abstract": " The study of neuronal interactions is currently at the center of several\nneuroscience big collaborative projects (including the Human Connectome, the\nBlue Brain, the Brainome, etc.) which attempt to obtain a detailed map of the\nentire brain matrix. Under certain constraints, mathematical theory can advance\npredictions of the expected neural dynamics based solely on the statistical\nproperties of such synaptic interaction matrix. This work explores the\napplication of free random variables (FRV) to the study of large synaptic\ninteraction matrices. Besides recovering in a straightforward way known results\non eigenspectra of neural networks, we extend them to heavy-tailed\ndistributions of interactions. More importantly, we derive analytically the\nbehavior of eigenvector overlaps, which determine stability of the spectra. We\nobserve that upon imposing the neuronal excitation/inhibition balance, although\nthe eigenvalues remain unchanged, their stability dramatically decreases due to\nstrong non-orthogonality of associated eigenvectors. It leads us to the\nconclusion that the understanding of the temporal evolution of asymmetric\nneural networks requires considering the entangled dynamics of both\neigenvectors and eigenvalues, which might bear consequences for learning and\nmemory processes in these models. Considering the success of FRV analysis in a\nwide variety of branches disciplines, we hope that the results presented here\nfoster additional application of these ideas in the area of brain sciences.\n",
"title": "From synaptic interactions to collective dynamics in random neuronal networks models: critical role of eigenvectors and transient behavior"
}
| null | null | null | null | true | null |
16900
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.