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": " This paper describes a massively parallel code for a state-of-the art thermal\nlattice- Boltzmann method. Our code has been carefully optimized for\nperformance on one GPU and to have a good scaling behavior extending to a large\nnumber of GPUs. Versions of this code have been already used for large-scale\nstudies of convective turbulence. GPUs are becoming increasingly popular in HPC\napplications, as they are able to deliver higher performance than traditional\nprocessors. Writing efficient programs for large clusters is not an easy task\nas codes must adapt to increasingly parallel architectures, and the overheads\nof node-to-node communications must be properly handled. We describe the\nstructure of our code, discussing several key design choices that were guided\nby theoretical models of performance and experimental benchmarks. We present an\nextensive set of performance measurements and identify the corresponding main\nbot- tlenecks; finally we compare the results of our GPU code with those\nmeasured on other currently available high performance processors. Our results\nare a production-grade code able to deliver a sustained performance of several\ntens of Tflops as well as a design and op- timization methodology that can be\nused for the development of other high performance applications for\ncomputational physics.\n",
"title": "Massively parallel lattice-Boltzmann codes on large GPU clusters"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4401
| null |
Validated
| null | null |
null |
{
"abstract": " This article carries out a large dimensional analysis of standard regularized\ndiscriminant analysis classifiers designed on the assumption that data arise\nfrom a Gaussian mixture model with different means and covariances. The\nanalysis relies on fundamental results from random matrix theory (RMT) when\nboth the number of features and the cardinality of the training data within\neach class grow large at the same pace. Under mild assumptions, we show that\nthe asymptotic classification error approaches a deterministic quantity that\ndepends only on the means and covariances associated with each class as well as\nthe problem dimensions. Such a result permits a better understanding of the\nperformance of regularized discriminant analsysis, in practical large but\nfinite dimensions, and can be used to determine and pre-estimate the optimal\nregularization parameter that minimizes the misclassification error\nprobability. Despite being theoretically valid only for Gaussian data, our\nfindings are shown to yield a high accuracy in predicting the performances\nachieved with real data sets drawn from the popular USPS data base, thereby\nmaking an interesting connection between theory and practice.\n",
"title": "A Large Dimensional Study of Regularized Discriminant Analysis Classifiers"
}
| null | null | null | null | true | null |
4402
| null |
Default
| null | null |
null |
{
"abstract": " We study the differences and equivalences between the non-perturbative\ndescription of the evolution of cosmic structure furnished by the Szekeres dust\nmodels (a non-spherical exact solution of Einstein's equations) and the\ndynamics of Cosmological Perturbation Theory (CPT) for dust sources in a\n$\\Lambda$CDM background. We show how the dynamics of Szekeres models can be\ndescribed by evolution equations given in terms of \"exact fluctuations\" that\nidentically reduce (at all orders) to evolution equations of CPT in the\ncomoving isochronous gauge. We explicitly show how Szekeres linearised exact\nfluctuations are specific (deterministic) realisations of standard linear\nperturbations of CPT given as random fields but, as opposed to the latter\nperturbations, they can be evolved exactly into the full non-linear regime. We\nprove two important results: (i) the conservation of the curvature perturbation\n(at all scales) also holds for the appropriate approximation of the exact\nSzekeres fluctuations in a $\\Lambda$CDM background, and (ii) the different\ncollapse morphologies of Szekeres models yields, at nonlinear order, different\nfunctional forms for the growth factor that follows from the study of redshift\nspace distortions. The metric based potentials used in linear CPT are computed\nin terms of the parameters of the linearised Szekeres models, thus allowing us\nto relate our results to linear CPT results in other gauges. We believe that\nthese results provide a solid starting stage to examine the role of\nnon-perturbative General Relativity in current cosmological research.\n",
"title": "Non-Spherical Szekeres models in the language of Cosmological Perturbations"
}
| null | null | null | null | true | null |
4403
| null |
Default
| null | null |
null |
{
"abstract": " Vibrational energy harvesters capture mechanical energy from ambient\nvibrations and convert the mechanical energy into electrical energy to power\nwireless electronic systems. Challenges exist in the process of capturing\nmechanical energy from ambient vibrations. For example, resonant harvesters may\nbe used to improve power output near their resonance, but their narrow\nbandwidth makes them less suitable for applications with varying vibrational\nfrequencies. Higher operating frequencies can increase harvesters power output,\nbut many vibrational sources are characterized by lower frequencies, such as\nhuman motions. This paper provides a thorough review of state of the art energy\nharvesters based on various energy sources such as solar, thermal,\nelectromagnetic and mechanical energy, as well as smart materials including\npiezoelectric materials and carbon nanotubes. The paper will then focus on\nvibrational energy harvesters to review harvesters using typical transduction\nmechanisms and various techniques to address the challenges in capturing\nmechanical energy and delivering it to the transducers.\n",
"title": "Smart materials and structures for energy harvesters"
}
| null | null |
[
"Physics"
] | null | true | null |
4404
| null |
Validated
| null | null |
null |
{
"abstract": " The paper presents a novel, principled approach to train recurrent neural\nnetworks from the Reservoir Computing family that are robust to missing part of\nthe input features at prediction time. By building on the ensembling properties\nof Dropout regularization, we propose a methodology, named DropIn, which\nefficiently trains a neural model as a committee machine of subnetworks, each\ncapable of predicting with a subset of the original input features. We discuss\nthe application of the DropIn methodology in the context of Reservoir Computing\nmodels and targeting applications characterized by input sources that are\nunreliable or prone to be disconnected, such as in pervasive wireless sensor\nnetworks and ambient intelligence. We provide an experimental assessment using\nreal-world data from such application domains, showing how the Dropin\nmethodology allows to maintain predictive performances comparable to those of a\nmodel without missing features, even when 20\\%-50\\% of the inputs are not\navailable.\n",
"title": "DropIn: Making Reservoir Computing Neural Networks Robust to Missing Inputs by Dropout"
}
| null | null | null | null | true | null |
4405
| null |
Default
| null | null |
null |
{
"abstract": " An area efficient row-parallel architecture is proposed for the real-time\nimplementation of bivariate algebraic integer (AI) encoded 2-D discrete cosine\ntransform (DCT) for image and video processing. The proposed architecture\ncomputes 8$\\times$8 2-D DCT transform based on the Arai DCT algorithm. An\nimproved fast algorithm for AI based 1-D DCT computation is proposed along with\na single channel 2-D DCT architecture. The design improves on the 4-channel AI\nDCT architecture that was published recently by reducing the number of integer\nchannels to one and the number of 8-point 1-D DCT cores from 5 down to 2. The\narchitecture offers exact computation of 8$\\times$8 blocks of the 2-D DCT\ncoefficients up to the FRS, which converts the coefficients from the AI\nrepresentation to fixed-point format using the method of expansion factors.\nPrototype circuits corresponding to FRS blocks based on two expansion factors\nare realized, tested, and verified on FPGA-chip, using a Xilinx Virtex-6\nXC6VLX240T device. Post place-and-route results show a 20% reduction in terms\nof area compared to the 2-D DCT architecture requiring five 1-D AI cores. The\narea-time and area-time${}^2$ complexity metrics are also reduced by 23% and\n22% respectively for designs with 8-bit input word length. The digital\nrealizations are simulated up to place and route for ASICs using 45 nm CMOS\nstandard cells. The maximum estimated clock rate is 951 MHz for the CMOS\nrealizations indicating 7.608$\\cdot$10$^9$ pixels/seconds and a 8$\\times$8\nblock rate of 118.875 MHz.\n",
"title": "A Single-Channel Architecture for Algebraic Integer Based 8$\\times$8 2-D DCT Computation"
}
| null | null | null | null | true | null |
4406
| null |
Default
| null | null |
null |
{
"abstract": " Following a paper in which the fundamental aspects of probabilistic inference\nwere introduced by means of a toy experiment, details of the analysis of\nsimulated long sequences of extractions are shown here. In fact, the striking\nperformance of probability-based inference and forecasting, compared to those\nobtained by simple `rules', might impress those practitioners who are usually\nunderwhelmed by the philosophical foundation of the different methods. The\nanalysis of the sequences also shows how the smallness of the probability of\nwhat has been actually observed, given the hypotheses of interest, is\nirrelevant for the purpose of inference.\n",
"title": "More lessons from the six box toy experiment"
}
| null | null | null | null | true | null |
4407
| null |
Default
| null | null |
null |
{
"abstract": " Osteonecrosis occurs due to the loss of blood supply to the bone, leading to\nspontaneous death of the trabecular bone. Delayed treatment of the involved\npatients results in collapse of the femoral head, which leads to a need for\ntotal hip arthroplasty surgery. Core decompression, as the most popular\ntechnique for treatment of the osteonecrosis, includes removal of the lesion\narea by drilling a straight tunnel to the lesion, debriding the dead bone and\nreplacing it with bone substitutes. However, there are two drawbacks for this\ntreatment method. First, due to the rigidity of the instruments currently used\nduring core decompression, lesions cannot be completely removed and/or\nexcessive healthy bone may also be removed with the lesion. Second, the use of\nbone substitutes, despite its biocompatibility and osteoconductivity, may not\nprovide sufficient mechanical strength and support for the bone. To address\nthese shortcomings, a novel robot-assisted curved core decompression (CCD)\ntechnique is introduced to provide surgeons with direct access to the lesions\ncausing minimal damage to the healthy bone. In this study, with the aid of\nfinite element (FE) simulations, we investigate biomechanical performance of\ncore decompression using the curved drilling technique in the presence of\nnormal gait loading. In this regard, we compare the result of the CCD using\nbone substitutes and flexible implants with other conventional core\ndecompression techniques. The study finding shows that the maximum principal\nstress occurring at the superior domain of the neck is smaller in the CCD\ntechniques (i.e. 52.847 MPa) compared to the other core decompression methods.\n",
"title": "A Biomechanical Study on the Use of Curved Drilling Technique for Treatment of Osteonecrosis of Femoral Head"
}
| null | null | null | null | true | null |
4408
| null |
Default
| null | null |
null |
{
"abstract": " We study the output feedback exponential stabilization of a one-dimensional\nunstable wave equation, where the boundary input, given by the Neumann trace at\none end of the domain, is the sum of the control input and the total\ndisturbance. The latter is composed of a nonlinear uncertain feedback term and\nan external bounded disturbance. Using the two boundary displacements as output\nsignals, we design a disturbance estimator that does not use high gain. It is\nshown that the disturbance estimator can estimate the total disturbance in the\nsense that the estimation error signal is in $L^2[0,\\infty)$. Using the\nestimated total disturbance, we design an observer whose state is exponentially\nconvergent to the state of original system. Finally, we design an\nobserver-based output feedback stabilizing controller. The total disturbance is\napproximately canceled in the feedback loop by its estimate. The closed-loop\nsystem is shown to be exponentially stable while guaranteeing that all the\ninternal signals are uniformly bounded.\n",
"title": "Output feedback exponential stabilization for 1-D unstable wave equations with boundary control matched disturbance"
}
| null | null | null | null | true | null |
4409
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we introduce a design principle to develop novel soft modular\nrobots based on tensegrity structures and inspired by the cytoskeleton of\nliving cells. We describe a novel strategy to realize tensegrity structures\nusing planar manufacturing techniques, such as 3D printing. We use this\nstrategy to develop icosahedron tensegrity structures with programmable\nvariable stiffness that can deform in a three-dimensional space. We also\ndescribe a tendon-driven contraction mechanism to actively control the\ndeformation of the tensegrity mod-ules. Finally, we validate the approach in a\nmodular locomotory worm as a proof of concept.\n",
"title": "Bio-inspired Tensegrity Soft Modular Robots"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4410
| null |
Validated
| null | null |
null |
{
"abstract": " The well-known Bayes theorem assumes that a posterior distribution is a\nprobability distribution. However, the posterior distribution may no longer be\na probability distribution if an improper prior distribution (non-probability\nmeasure) such as an unbounded uniform prior is used. Improper priors are often\nused in the astronomical literature to reflect a lack of prior knowledge, but\nchecking whether the resulting posterior is a probability distribution is\nsometimes neglected. It turns out that 23 articles out of 75 articles (30.7%)\npublished online in two renowned astronomy journals (ApJ and MNRAS) between Jan\n1, 2017 and Oct 15, 2017 make use of Bayesian analyses without rigorously\nestablishing posterior propriety. A disturbing aspect is that a Gibbs-type\nMarkov chain Monte Carlo (MCMC) method can produce a seemingly reasonable\nposterior sample even when the posterior is not a probability distribution\n(Hobert and Casella, 1996). In such cases, researchers may erroneously make\nprobabilistic inferences without noticing that the MCMC sample is from a\nnon-existing probability distribution. We review why checking posterior\npropriety is fundamental in Bayesian analyses, and discuss how to set up\nscientifically motivated proper priors.\n",
"title": "How proper are Bayesian models in the astronomical literature?"
}
| null | null |
[
"Physics"
] | null | true | null |
4411
| null |
Validated
| null | null |
null |
{
"abstract": " I consider a Jovian planet on a highly eccentric orbit around its host star,\na situation produced by secular interactions with its planetary or stellar\ncompanions. The tidal interactions at every periastron passage exchange energy\nbetween the orbit and the planet's degree-2 fundamental-mode. Starting from\nzero energy, the f-mode can diffusively grow to large amplitudes if its\none-kick energy gain > 10^-5 of the orbital energy. This requires a pericentre\ndistance of < 4 tidal radii (or 1.6 Roche radii). If the f-mode has a\nnon-negligible initial energy, diffusive evolution can occur at a lower\nthreshold. The first effect can stall the secular migration as the f-mode can\nabsorb orbital energy and decouple the planet from its secular perturbers,\nparking all migrating jupiters safely outside the zone of tidal disruption. The\nsecond effect leads to rapid orbit circularization as it allows an excited\nf-mode to continuously absorb orbital energy as the orbit eccentricity\ndecreases. So without any explicit dissipation, other than the fact that the\nf-mode will damp nonlinearly when its amplitude reaches unity, the planet can\nbe transported from a few AU to ~ 0.2 AU in ~ 10^4 yrs. Such a rapid\ncircularization is equivalent to a dissipation factor Q ~ 1, and it explains\nthe observed deficit of super-eccentric Jovian planets. Lastly, the repeated\nf-mode breaking likely deposit energy and angular momentum in the outer\nenvelope, and avoid thermally ablating the planet.\nOverall, this work boosts the case for forming hot Jupiters through\nhigh-eccentricity secular migration.\n",
"title": "Diffusive Tidal Evolution for Migrating hot Jupiters"
}
| null | null | null | null | true | null |
4412
| null |
Default
| null | null |
null |
{
"abstract": " This note contains a reformulation of the Hodge index theorem within the\nframework of Atiyah's $L^2$-index theory. More precisely, given a compact\nKähler manifold $(M,h)$ of even complex dimension $2m$, we prove that\n$$\\sigma(M)=\\sum_{p,q=0}^{2m}(-1)^ph_{(2),\\Gamma}^{p,q}(M)$$ where $\\sigma(M)$\nis the signature of $M$ and $h_{(2),\\Gamma}^{p,q}(M)$ are the $L^2$-Hodge\nnumbers of $M$ with respect to a Galois covering having $\\Gamma$ as group of\nDeck transformations. Likewise we also prove an $L^2$-version of the\nFrölicher index theorem. Afterwards we give some applications of these two\ntheorems and finally we conclude this paper by collecting other properties of\nthe $L^2$-Hodge numbers.\n",
"title": "Von Neumann dimension, Hodge index theorem and geometric applications"
}
| null | null |
[
"Mathematics"
] | null | true | null |
4413
| null |
Validated
| null | null |
null |
{
"abstract": " We study the asymmetry in the two-point cross-correlation function of two\npopulations of galaxies focusing in particular on the relativistic effects that\ninclude the gravitational redshift. We derive the cross-correlation function on\nsmall and large scales using two different approaches: General Relativistic and\nNewtonian perturbation theory. Following recent work by Bonvin et al.,\nGaztanaga et al. and Croft, we calculate the dipole and the shell estimator\nwith the two procedures and we compare our results. We find that while General\nRelativistic Perturbation Theory (GRPT) is able to make predictions of\nrelativistic effects on very large, obviously linear scales (r > 50 Mpc/h), the\npresence of non-linearities physically occurring on much smaller scales (down\nto those describing galactic potential wells) can strongly affect the asymmetry\nestimators. These can lead to cancellations of the relativistic terms, and sign\nchanges in the estimators on scales up to r ~ 50 Mpc/h. On the other hand, with\nan appropriate non-linear gravitational potential, the results obtained using\nNewtonian theory can successfully describe the asymmetry on smaller, non-linear\nscales (r < 20 Mpc/h) where gravitational redshift is the dominant term. On\nlarger scales the asymmetry is much smaller in magnitude, and measurement is\nnot within reach of current observations. This is in agreement with the\nobservational results obtained by Gaztnaga et al. and the first detection of\nrelativistic effects (on (r < 20 Mpc/h) scales) by Alam et al.\n",
"title": "Relativistic asymmetries in the galaxy cross-correlation function"
}
| null | null | null | null | true | null |
4414
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we present an end-to-end automatic speech recognition system,\nwhich successfully employs subword units in a hybrid CTC-Attention based\nsystem. The subword units are obtained by the byte-pair encoding (BPE)\ncompression algorithm. Compared to using words as modeling units, using\ncharacters or subword units does not suffer from the out-of-vocabulary (OOV)\nproblem. Furthermore, using subword units further offers a capability in\nmodeling longer context than using characters. We evaluate different systems\nover the LibriSpeech 1000h dataset. The subword-based hybrid CTC-Attention\nsystem obtains 6.8% word error rate (WER) on the test_clean subset without any\ndictionary or external language model. This represents a significant\nimprovement (a 12.8% WER relative reduction) over the character-based hybrid\nCTC-Attention system.\n",
"title": "Hybrid CTC-Attention based End-to-End Speech Recognition using Subword Units"
}
| null | null | null | null | true | null |
4415
| null |
Default
| null | null |
null |
{
"abstract": " Fourier analysis and representation of circular distributions in terms of\ntheir Fourier coefficients, is quite commonly discussed and used for model-free\ninference such as testing uniformity and symmetry etc. in dealing with\n2-dimensional directions. However a similar discussion for spherical\ndistributions, which are used to model 3-dimensional directional data, has not\nbeen fully developed in the literature in terms of their harmonics. This paper,\nin what we believe is the first such attempt, looks at the probability\ndistributions on a unit sphere, through the perspective of spherical harmonics,\nanalogous to the Fourier analysis for distributions on a unit circle. Harmonic\nrepresentations of many currently used spherical models are presented and\ndiscussed. A very general family of spherical distributions is then introduced,\nspecial cases of which yield many known spherical models. Through the prism of\nharmonic analysis, one can look at the mean direction, dispersion, and various\nforms of symmetry for these models in a generic setting. Aspects of\ndistribution free inference such as estimation and large-sample tests for these\nsymmetries, are provided. The paper concludes with a real-data example\nanalyzing the longitudinal sunspot activity.\n",
"title": "Harmonic analysis and distribution-free inference for spherical distributions"
}
| null | null | null | null | true | null |
4416
| null |
Default
| null | null |
null |
{
"abstract": " A new method is presented for modelling the physical properties of galaxy\nclusters. Our technique moves away from the traditional approach of assuming\nspecific parameterised functional forms for the variation of physical\nquantities within the cluster, and instead allows for a 'free-form'\nreconstruction, but one for which the level of complexity is determined\nautomatically by the observational data and may depend on position within the\ncluster. This is achieved by representing each independent cluster property as\nsome interpolating or approximating function that is specified by a set of\ncontrol points, or 'nodes', for which the number of nodes, together with their\npositions and amplitudes, are allowed to vary and are inferred in a Bayesian\nmanner from the data. We illustrate our nodal approach in the case of a\nspherical cluster by modelling the electron pressure profile Pe(r) in analyses\nboth of simulated Sunyaev-Zel'dovich (SZ) data from the Arcminute MicroKelvin\nImager (AMI) and of real AMI observations of the cluster MACS J0744+3927 in the\nCLASH sample. We demonstrate that one may indeed determine the complexity\nsupported by the data in the reconstructed Pe(r), and that one may constrain\ntwo very important quantities in such an analysis: the cluster total volume\nintegrated Comptonisation parameter (Ytot) and the extent of the gas\ndistribution in the cluster (rmax). The approach is also well-suited to\ndetecting clusters in blind SZ surveys.\n",
"title": "Free-form modelling of galaxy clusters: a Bayesian and data-driven approach"
}
| null | null |
[
"Physics"
] | null | true | null |
4417
| null |
Validated
| null | null |
null |
{
"abstract": " Deep reinforcement learning (DRL) has shown incredible performance in\nlearning various tasks to the human level. However, unlike human perception,\ncurrent DRL models connect the entire low-level sensory input to the\nstate-action values rather than exploiting the relationship between and among\nentities that constitute the sensory input. Because of this difference, DRL\nneeds vast amount of experience samples to learn. In this paper, we propose a\nMulti-focus Attention Network (MANet) which mimics human ability to spatially\nabstract the low-level sensory input into multiple entities and attend to them\nsimultaneously. The proposed method first divides the low-level input into\nseveral segments which we refer to as partial states. After this segmentation,\nparallel attention layers attend to the partial states relevant to solving the\ntask. Our model estimates state-action values using these attended partial\nstates. In our experiments, MANet attains highest scores with significantly\nless experience samples. Additionally, the model shows higher performance\ncompared to the Deep Q-network and the single attention model as benchmarks.\nFurthermore, we extend our model to attentive communication model for\nperforming multi-agent cooperative tasks. In multi-agent cooperative task\nexperiments, our model shows 20% faster learning than existing state-of-the-art\nmodel.\n",
"title": "Multi-focus Attention Network for Efficient Deep Reinforcement Learning"
}
| null | null | null | null | true | null |
4418
| null |
Default
| null | null |
null |
{
"abstract": " We propose an Analytical method of Blind Separation (ABS) of cosmic\nmagnification from the intrinsic fluctuations of galaxy number density in the\nobserved galaxy number density distribution. The ABS method utilizes the\ndifferent dependences of the signal (cosmic magnification) and contamination\n(galaxy intrinsic clustering) on galaxy flux, to separate the two. It works\ndirectly on the measured cross galaxy angular power spectra between different\nflux bins. It determines/reconstructs the lensing power spectrum analytically,\nwithout assumptions of galaxy intrinsic clustering and cosmology. It is\nunbiased in the limit of infinite number of galaxies. In reality the lensing\nreconstruction accuracy depends on survey configurations, galaxy biases, and\nother complexities, due to finite number of galaxies and the resulting shot\nnoise fluctuations in the cross galaxy power spectra. We estimate its\nperformance (systematic and statistical errors) in various cases. We find that,\nstage IV dark energy surveys such as SKA and LSST are capable of reconstructing\nthe lensing power spectrum at $z\\simeq 1$ and $\\ell\\la 5000$ accurately. This\nlensing reconstruction only requires counting galaxies, and is therefore highly\ncomplementary to the cosmic shear measurement by the same surveys.\n",
"title": "Weak lensing power spectrum reconstruction by counting galaxies.-- I: the ABS method"
}
| null | null | null | null | true | null |
4419
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we study an optimal output consensus problem for a multi-agent\nnetwork with agents in the form of multi-input multi-output minimum-phase\ndynamics. Optimal output consensus can be taken as an extended version of the\nexisting output consensus problem for higher-order agents with an optimization\nrequirement, where the output variables of agents are driven to achieve a\nconsensus on the optimal solution of a global cost function. To solve this\nproblem, we first construct an optimal signal generator, and then propose an\nembedded control scheme by embedding the generator in the feedback loop. We\ngive two kinds of algorithms based on different available information along\nwith both state feedback and output feedback, and prove that these algorithms\nwith the embedded technique can guarantee the solvability of the problem for\nhigh-order multi-agent systems under standard assumptions.\n",
"title": "Optimal Output Consensus of High-Order Multi-Agent Systems with Embedded Technique"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
4420
| null |
Validated
| null | null |
null |
{
"abstract": " We propose a new dynamics for equilibrium selection of finite player discrete\nstrategy games. The dynamics is motivated by optimal transportation, and models\nindividual players' myopicity, greedy and uncertainty when making decisions.\nThe stationary measure of the dynamics provides each pure Nash equilibrium a\nprobability by which it is ranked. For potential games, its dynamical\nproperties are characterized by entropy and Fisher information.\n",
"title": "Equilibrium selection via Optimal transport"
}
| null | null | null | null | true | null |
4421
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a new method for medical diagnosis of neurodegenerative\ndiseases, such as Parkinson's, by extracting and using latent information from\ntrained Deep convolutional, or convolutional-recurrent Neural Networks (DNNs).\nIn particular, our approach adopts a combination of transfer learning, k-means\nclustering and k-Nearest Neighbour classification of deep neural network\nlearned representations to provide enriched prediction of the disease based on\nMRI and/or DaT Scan data. A new loss function is introduced and used in the\ntraining of the DNNs, so as to perform adaptation of the generated learned\nrepresentations between data from different medical environments. Results are\npresented using a recently published database of Parkinson's related\ninformation, which was generated and evaluated in a hospital environment.\n",
"title": "Predicting Parkinson's Disease using Latent Information extracted from Deep Neural Networks"
}
| null | null | null | null | true | null |
4422
| null |
Default
| null | null |
null |
{
"abstract": " New results are added to the paper [4] about q-closed and solvable\nsesquilinear forms. The structure of the Banach space\n$\\mathcal{D}[||\\cdot||_\\Omega]$ defined on the domain $\\mathcal{D}$ of a\nq-closed sesquilinear form $\\Omega$ is unique up to isomorphism, and the\nadjoint of a sesquilinear form has the same property of q-closure or of\nsolvability. The operator associated to a solvable sesquilinear form is the\ngreatest which represents the form and it is self-adjoint if, and only if, the\nform is symmetric. We give more criteria of solvability for q-closed\nsesquilinear forms. Some of these criteria are related to the numerical range,\nand we analyse in particular the forms which are solvable with respect to inner\nproducts. The theory of solvable sesquilinear forms generalises those of many\nknown sesquilinear forms in literature.\n",
"title": "Representation Theorems for Solvable Sesquilinear Forms"
}
| null | null | null | null | true | null |
4423
| null |
Default
| null | null |
null |
{
"abstract": " This letter provides a simple but efficient technique, which allows each\ntransmitter of an interference network, to exchange local channel state\ninformation with the other transmitters. One salient feature of the proposed\ntechnique is that a transmitter only needs measurements of the signal power at\nits intended receiver to implement it, making direct inter-transmitter\nsignaling channels unnecessary. The key idea to achieve this is to use a\ntransient period during which the continuous power level of a transmitter is\ntaken to be the linear combination of the channel gains to be exchanged.\n",
"title": "Using Continuous Power Modulation for Exchanging Local Channel State Information"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4424
| null |
Validated
| null | null |
null |
{
"abstract": " The current and envisaged increase of cellular traffic poses new challenges\nto Mobile Network Operators (MNO), who must densify their Radio Access Networks\n(RAN) while maintaining low Capital Expenditure and Operational Expenditure to\nensure long-term sustainability. In this context, this paper analyses optimal\nclustering solutions based on Device-to-Device (D2D) communications to mitigate\npartially or completely the need for MNOs to carry out extremely dense RAN\ndeployments. Specifically, a low complexity algorithm that enables the creation\nof spectral efficient clusters among users from different cells, denoted as\nenhanced Clustering Optimization for Resources' Efficiency (eCORE) is\npresented. Due to the imbalance between uplink and downlink traffic, a\ncomplementary algorithm, known as Clustering algorithm for Load Balancing\n(CaLB), is also proposed to create non-spectral efficient clusters when they\nresult in a capacity increase. Finally, in order to alleviate the energy\noverconsumption suffered by cluster heads, the Clustering Energy Efficient\nalgorithm (CEEa) is also designed to manage the trade-off between the capacity\nenhancement and the early battery drain of some users. Results show that the\nproposed algorithms increase the network capacity and outperform existing\nsolutions, while, at the same time, CEEa is able to handle the cluster heads\nenergy overconsumption.\n",
"title": "Spectral Efficient and Energy Aware Clustering in Cellular Networks"
}
| null | null | null | null | true | null |
4425
| null |
Default
| null | null |
null |
{
"abstract": " Recurrent neural network (RNN) language models (LMs) and Long Short Term\nMemory (LSTM) LMs, a variant of RNN LMs, have been shown to outperform\ntraditional N-gram LMs on speech recognition tasks. However, these models are\ncomputationally more expensive than N-gram LMs for decoding, and thus,\nchallenging to integrate into speech recognizers. Recent research has proposed\nthe use of lattice-rescoring algorithms using RNNLMs and LSTMLMs as an\nefficient strategy to integrate these models into a speech recognition system.\nIn this paper, we evaluate existing lattice rescoring algorithms along with new\nvariants on a YouTube speech recognition task. Lattice rescoring using LSTMLMs\nreduces the word error rate (WER) for this task by 8\\% relative to the WER\nobtained using an N-gram LM.\n",
"title": "Lattice Rescoring Strategies for Long Short Term Memory Language Models in Speech Recognition"
}
| null | null | null | null | true | null |
4426
| null |
Default
| null | null |
null |
{
"abstract": " J. DeLoera-T. McAllister and K. D. Mulmuley-H. Narayanan-M. Sohoni\nindependently proved that determining the vanishing of Littlewood-Richardson\ncoefficients has strongly polynomial time computational complexity. Viewing\nthese as Schubert calculus numbers, we prove the generalization to the\nLittlewood-Richardson polynomials that control equivariant cohomology of\nGrassmannians. We construct a polytope using the edge-labeled tableau rule of\nH. Thomas-A. Yong. Our proof then combines a saturation theorem of D.\nAnderson-E. Richmond-A. Yong, a reading order independence property, and E.\nTardos' algorithm for combinatorial linear programming.\n",
"title": "Vanishing of Littlewood-Richardson polynomials is in P"
}
| null | null | null | null | true | null |
4427
| null |
Default
| null | null |
null |
{
"abstract": " In aspect-based sentiment analysis, most existing methods either focus on\naspect/opinion terms extraction or aspect terms categorization. However, each\ntask by itself only provides partial information to end users. To generate more\ndetailed and structured opinion analysis, we propose a finer-grained problem,\nwhich we call category-specific aspect and opinion terms extraction. This\nproblem involves the identification of aspect and opinion terms within each\nsentence, as well as the categorization of the identified terms. To this end,\nwe propose an end-to-end multi-task attention model, where each task\ncorresponds to aspect/opinion terms extraction for a specific category. Our\nmodel benefits from exploring the commonalities and relationships among\ndifferent tasks to address the data sparsity issue. We demonstrate its\nstate-of-the-art performance on three benchmark datasets.\n",
"title": "Multi-task memory networks for category-specific aspect and opinion terms co-extraction"
}
| null | null | null | null | true | null |
4428
| null |
Default
| null | null |
null |
{
"abstract": " I begin my discussion by summarizing the methodology proposed and new\ndistributional results on multivariate log-Gamma derived in the paper. Then, I\ndraw an interesting connection between their work with mean field variational\nBayes. Lastly, I make some comments on the simulation results and the\nperformance of the proposed Poisson multivariate spatio-temporal mixed effects\nmodel (P-MSTM).\n",
"title": "Discussion on Computationally Efficient Multivariate Spatio-Temporal Models for High-Dimensional Count-Valued Data by Bradley et al"
}
| null | null | null | null | true | null |
4429
| null |
Default
| null | null |
null |
{
"abstract": " We present the results of resonant x-ray scattering measurements and\nelectronic structure calculations on the monoarsenide FeAs. We elucidate\ndetails of the magnetic structure, showing the ratio of ellipticity of the spin\nhelix is larger than previously thought, at 2.58(3), and reveal both a\nright-handed chirality and an out of plane component of the magnetic moments in\nthe spin helix. We find that electronic structure calculations and analysis of\nthe spin-orbit interaction are able to qualitatively account for this canting.\n",
"title": "Elucidation of the helical spin structure of FeAs"
}
| null | null | null | null | true | null |
4430
| null |
Default
| null | null |
null |
{
"abstract": " Rapid popularity of Internet of Things (IoT) and cloud computing permits\nneuroscientists to collect multilevel and multichannel brain data to better\nunderstand brain functions, diagnose diseases, and devise treatments. To ensure\nsecure and reliable data communication between end-to-end (E2E) devices\nsupported by current IoT and cloud infrastructure, trust management is needed\nat the IoT and user ends. This paper introduces a Neuro-Fuzzy based\nBrain-inspired trust management model (TMM) to secure IoT devices and relay\nnodes, and to ensure data reliability. The proposed TMM utilizes node\nbehavioral trust and data trust estimated using Adaptive Neuro-Fuzzy Inference\nSystem and weighted-additive methods respectively to assess the nodes\ntrustworthiness. In contrast to the existing fuzzy based TMMs, the NS2\nsimulation results confirm the robustness and accuracy of the proposed TMM in\nidentifying malicious nodes in the communication network. With the growing\nusage of cloud based IoT frameworks in Neuroscience research, integrating the\nproposed TMM into the existing infrastructure will assure secure and reliable\ndata communication among the E2E devices.\n",
"title": "A Brain-Inspired Trust Management Model to Assure Security in a Cloud based IoT Framework for Neuroscience Applications"
}
| null | null | null | null | true | null |
4431
| null |
Default
| null | null |
null |
{
"abstract": " In environments with scarce resources, adopting the right search strategy can\nmake the difference between succeeding and failing, even between life and\ndeath. At different scales, this applies to molecular encounters in the cell\ncytoplasm, to animals looking for food or mates in natural landscapes, to\nrescuers during search-and-rescue operations in disaster zones, as well as to\ngenetic computer algorithms exploring parameter spaces. When looking for sparse\ntargets in a homogeneous environment, a combination of ballistic and diffusive\nsteps is considered optimal; in particular, more ballistic Lévy flights with\nexponent {\\alpha} <= 1 are generally believed to optimize the search process.\nHowever, most search spaces present complex topographies, with boundaries,\nbarriers and obstacles. What is the best search strategy in these more\nrealistic scenarios? Here we show that the topography of the environment\nsignificantly alters the optimal search strategy towards less ballistic and\nmore Brownian strategies. We consider an active particle performing a blind\nsearch in a two-dimensional space with steps drawn from a Lévy distribution\nwith exponent varying from {\\alpha} = 1 to {\\alpha} = 2 (Brownian). We\ndemonstrate that the optimal search strategy depends on the topography of the\nenvironment, with {\\alpha} assuming intermediate values in the whole range\nunder consideration. We interpret these findings in terms of a simple\ntheoretical model, and discuss their robustness to the addition of Brownian\ndiffusion to the searcher's motion. Our results are relevant for search\nproblems at different length scales, from animal and human foraging to\nmicroswimmers' taxis, to biochemical rates of reaction.\n",
"title": "The topography of the environment alters the optimal search strategy for active particles"
}
| null | null | null | null | true | null |
4432
| null |
Default
| null | null |
null |
{
"abstract": " The performance of Neural Network (NN)-based language models is steadily\nimproving due to the emergence of new architectures, which are able to learn\ndifferent natural language characteristics. This paper presents a novel\nframework, which shows that a significant improvement can be achieved by\ncombining different existing heterogeneous models in a single architecture.\nThis is done through 1) a feature layer, which separately learns different\nNN-based models and 2) a mixture layer, which merges the resulting model\nfeatures. In doing so, this architecture benefits from the learning\ncapabilities of each model with no noticeable increase in the number of model\nparameters or the training time. Extensive experiments conducted on the Penn\nTreebank (PTB) and the Large Text Compression Benchmark (LTCB) corpus showed a\nsignificant reduction of the perplexity when compared to state-of-the-art\nfeedforward as well as recurrent neural network architectures.\n",
"title": "A Neural Network Approach for Mixing Language Models"
}
| null | null | null | null | true | null |
4433
| null |
Default
| null | null |
null |
{
"abstract": " Covalent-organic frameworks (COFs) are intriguing platforms for designing\nfunctional molecular materials. Here, we present a computational study based on\nvan der Waals dispersion-corrected hybrid density functional theory\ncalculations to analyze the material properties of boroxine-linked and\ntriazine-linked intercalated-COFs. The effect of Fe atoms on the electronic\nband structures near the Fermi energy level of the intercalated-COFs have been\ninvestigated. The density of states (DOSs) computations have been performed to\nanalyze the material properties of these kind of intercalated-COFs. We predict\nthat COFs's electronic properties can be fine tuned by adding Fe atoms between\ntwo organic layers in their structures. The new COFs are predicted to be\nthermoelectric materials. These intercalated-COFs provide a new strategy to\ncreate thermoelectric materials within a rigid porous network in a highly\ncontrolled and predictable manner.\n",
"title": "Iron Intercalated Covalent-Organic Frameworks: First Crystalline Porous Thermoelectric Materials"
}
| null | null | null | null | true | null |
4434
| null |
Default
| null | null |
null |
{
"abstract": " Filters in a Convolutional Neural Network (CNN) contain model parameters\nlearned from enormous amounts of data. In this paper, we suggest to decompose\nconvolutional filters in CNN as a truncated expansion with pre-fixed bases,\nnamely the Decomposed Convolutional Filters network (DCFNet), where the\nexpansion coefficients remain learned from data. Such a structure not only\nreduces the number of trainable parameters and computation, but also imposes\nfilter regularity by bases truncation. Through extensive experiments, we\nconsistently observe that DCFNet maintains accuracy for image classification\ntasks with a significant reduction of model parameters, particularly with\nFourier-Bessel (FB) bases, and even with random bases. Theoretically, we\nanalyze the representation stability of DCFNet with respect to input\nvariations, and prove representation stability under generic assumptions on the\nexpansion coefficients. The analysis is consistent with the empirical\nobservations.\n",
"title": "DCFNet: Deep Neural Network with Decomposed Convolutional Filters"
}
| null | null | null | null | true | null |
4435
| null |
Default
| null | null |
null |
{
"abstract": " Gravitationally collapsed objects are known to be biased tracers of an\nunderlying density contrast. Using symmetry arguments, generalised biasing\nschemes have recently been developed to relate the halo density contrast\n$\\delta_h$ with the underlying density contrast $\\delta$, divergence of\nvelocity $\\theta$ and their higher-order derivatives. This is done by\nconstructing invariants such as $s, t, \\psi,\\eta$. We show how the generating\nfunction formalism in Eulerian standard perturbation theory (SPT) can be used\nto show that many of the additional terms based on extended Galilean and\nLifshitz symmetry actually do not make any contribution to the higher-order\nstatistics of biased tracers. Other terms can also be drastically simplified\nallowing us to write the vertices associated with $\\delta_h$ in terms of the\nvertices of $\\delta$ and $\\theta$, the higher-order derivatives and the bias\ncoefficients. We also compute the cumulant correlators (CCs) for two different\ntracer populations. These perturbative results are valid for tree-level\ncontributions but at an arbitrary order. We also take into account the\nstochastic nature bias in our analysis. Extending previous results of a local\npolynomial model of bias, we express the one-point cumulants ${\\cal S}_N$ and\ntheir two-point counterparts, the CCs i.e. ${\\cal C}_{pq}$, of biased tracers\nin terms of that of their underlying density contrast counterparts. As a\nby-product of our calculation we also discuss the results using approximations\nbased on Lagrangian perturbation theory (LPT).\n",
"title": "Symmetries, Invariants and Generating Functions: Higher-order Statistics of Biased Tracers"
}
| null | null | null | null | true | null |
4436
| null |
Default
| null | null |
null |
{
"abstract": " A relational structure ${\\mathbb X}$ is said to be reversible iff every\nbijective endomorphism $f:X\\rightarrow X$ is an automorphism. We define a\nsequence of non-zero cardinals $\\langle \\kappa_i :i\\in I\\rangle$ to be\nreversible iff each surjection $f :I\\rightarrow I$ such that $\\kappa_j\n=\\sum_{i\\in f^{-1}[\\{ j \\}]}\\kappa_i$, for all $j\\in I $, is a bijection, and\ncharacterize such sequences: either $\\langle \\kappa_i :i\\in I\\rangle$ is a\nfinite-to-one sequence, or $\\kappa_i\\in {\\mathbb N}$, for all $i\\in I$, $K:=\\{\nm\\in {\\mathbb N} : \\kappa_i =m $, for infinitely many $i\\in I \\}$ is a\nnon-empty independent set, and $\\gcd (K)$ divides at most finitely many\nelements of the set $\\{ \\kappa_i :i\\in I \\}$. We isolate a class of binary\nstructures such that a structure from the class is reversible iff the sequence\nof cardinalities of its connectivity components is reversible. In particular,\nwe characterize reversible equivalence relations, reversible posets which are\ndisjoint unions of cardinals $\\leq \\omega$, and some similar structures. In\naddition, we show that a poset with linearly ordered connectivity components is\nreversible, if the corresponding sequence of cardinalities is reversible and,\nusing this fact, detect a wide class of examples of reversible posets and\ntopological spaces.\n",
"title": "Reversible Sequences of Cardinals, Reversible Equivalence Relations, and Similar Structures"
}
| null | null | null | null | true | null |
4437
| null |
Default
| null | null |
null |
{
"abstract": " The difficulty of large scale monitoring of app markets affects our\nunderstanding of their dynamics. This is particularly true for dimensions such\nas app update frequency, control and pricing, the impact of developer actions\non app popularity, as well as coveted membership in top app lists. In this\npaper we perform a detailed temporal analysis on two datasets we have collected\nfrom the Google Play Store, one consisting of 160,000 apps and the other of\n87,223 newly released apps. We have monitored and collected data about these\napps over more than 6 months. Our results show that a high number of these apps\nhave not been updated over the monitoring interval. Moreover, these apps are\ncontrolled by a few developers that dominate the total number of app downloads.\nWe observe that infrequently updated apps significantly impact the median app\nprice. However, a changing app price does not correlate with the download\ncount. Furthermore, we show that apps that attain higher ranks have better\nstability in top app lists. We show that app market analytics can help detect\nemerging threat vectors, and identify search rank fraud and even malware.\nFurther, we discuss the research implications of app market analytics on\nimproving developer and user experiences.\n",
"title": "A Longitudinal Study of Google Play"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4438
| null |
Validated
| null | null |
null |
{
"abstract": " The recently developed variational autoencoders (VAEs) have proved to be an\neffective confluence of the rich representational power of neural networks with\nBayesian methods. However, most work on VAEs use a rather simple prior over the\nlatent variables such as standard normal distribution, thereby restricting its\napplications to relatively simple phenomena. In this work, we propose\nhierarchical nonparametric variational autoencoders, which combines\ntree-structured Bayesian nonparametric priors with VAEs, to enable infinite\nflexibility of the latent representation space. Both the neural parameters and\nBayesian priors are learned jointly using tailored variational inference. The\nresulting model induces a hierarchical structure of latent semantic concepts\nunderlying the data corpus, and infers accurate representations of data\ninstances. We apply our model in video representation learning. Our method is\nable to discover highly interpretable activity hierarchies, and obtain improved\nclustering accuracy and generalization capacity based on the learned rich\nrepresentations.\n",
"title": "Nonparametric Variational Auto-encoders for Hierarchical Representation Learning"
}
| null | null | null | null | true | null |
4439
| null |
Default
| null | null |
null |
{
"abstract": " In this work, we describe a problem which we refer to as the \\textbf{Spotify\nproblem} and explore a potential solution in the form of what we call\ncorpus-compressed streaming schemes.\nInspired by the problem of constrained bandwidth during use of the popular\nSpotify application on mobile networks, the Spotify problem applies in any\nnumber of practical domains where devices may be periodically expected to\nexperience degraded communication or storage capacity. One obvious solution\ncandidate which comes to mind immediately is standard compression. Though\nobviously applicable, standard compression does not in any way exploit all\ncharacteristics of the problem; in particular, standard compression is\noblivious to the fact that a decoder has a period of virtually unrestrained\ncommunication. Towards applying compression in a manner which attempts to\nstretch the benefit of periods of higher communication capacity into periods of\nrestricted capacity, we introduce as a solution the idea of a corpus-compressed\nstreaming scheme.\nThis report begins with a formal definition of a corpus-compressed streaming\nscheme. Following a discussion of how such schemes apply to the Spotify\nproblem, we then give a survey of specific corpus-compressed scheming schemes\nguided by an exploration of different measures of description complexity within\nthe Chomsky hierarchy of languages.\n",
"title": "Corpus-compressed Streaming and the Spotify Problem"
}
| null | null | null | null | true | null |
4440
| null |
Default
| null | null |
null |
{
"abstract": " If M is a smooth compact connected Riemannian manifold, let P(M) denote the\nWasserstein space of probability measures on M. We describe a geometric\nconstruction of parallel transport of some tangent cones along geodesics in\nP(M). We show that when everything is smooth, the geometric parallel transport\nagrees with earlier formal calculations.\n",
"title": "An intrinsic parallel transport in Wasserstein space"
}
| null | null |
[
"Mathematics"
] | null | true | null |
4441
| null |
Validated
| null | null |
null |
{
"abstract": " Power spectrum estimation is an important tool in many applications, such as\nthe whitening of noise. The popular multitaper method enjoys significant\nsuccess, but fails for short signals with few samples. We propose a statistical\nmodel where a signal is given by a random linear combination of fixed, yet\nunknown, stochastic sources. Given multiple such signals, we estimate the\nsubspace spanned by the power spectra of these fixed sources. Projecting\nindividual power spectrum estimates onto this subspace increases estimation\naccuracy. We provide accuracy guarantees for this method and demonstrate it on\nsimulated and experimental data from cryo-electron microscopy.\n",
"title": "Factor Analysis for Spectral Estimation"
}
| null | null |
[
"Mathematics",
"Statistics"
] | null | true | null |
4442
| null |
Validated
| null | null |
null |
{
"abstract": " Evaluating human brain potentials during watching different images can be\nused for memory evaluation, information retrieving, guilty-innocent\nidentification and examining the brain response. In this study, the effects of\nwatching images, with different levels of familiarity, on subjects'\nElectroencephalogram (EEG) have been studied. Three different groups of images\nwith three familiarity levels of \"unfamiliar\", \"familiar\" and \"very familiar\"\nhave been considered for this study. EEG signals of 21 subjects (14 men) were\nrecorded. After signal acquisition, pre-processing, including noise and\nartifact removal, were performed on epochs of data. Features, including\nspatial-statistical, wavelet, frequency and harmonic parameters, and also\ncorrelation between recording channels, were extracted from the data. Then, we\nevaluated the efficiency of the extracted features by using p-value and also an\northogonal feature selection method (combination of Gram-Schmitt method and\nFisher discriminant ratio) for feature dimensional reduction. As the final step\nof feature selection, we used 'add-r take-away l' method for choosing the most\ndiscriminative features. For data classification, including all two-class and\nthree-class cases, we applied Support Vector Machine (SVM) on the extracted\nfeatures. The correct classification rates (CCR) for \"unfamiliar-familiar\",\n\"unfamiliar-very familiar\" and \"familiar-very familiar\" cases were 85.6%,\n92.6%, and 70.6%, respectively. The best results of classifications were\nobtained in pre-frontal and frontal regions of brain. Also, wavelet, frequency\nand harmonic features were among the most discriminative features. Finally, in\nthree-class case, the best CCR was 86.8%.\n",
"title": "Effects of Images with Different Levels of Familiarity on EEG"
}
| null | null | null | null | true | null |
4443
| null |
Default
| null | null |
null |
{
"abstract": " Geophysical inversion should ideally produce geologically realistic\nsubsurface models that explain the available data. Multiple-point statistics is\na geostatistical approach to construct subsurface models that are consistent\nwith site-specific data, but also display the same type of patterns as those\nfound in a training image. The training image can be seen as a conceptual model\nof the subsurface and is used as a non-parametric model of spatial variability.\nInversion based on multiple-point statistics is challenging due to high\nnonlinearity and time-consuming geostatistical resimulation steps that are\nneeded to create new model proposals. We propose an entirely new model proposal\nmechanism for geophysical inversion that is inspired by texture synthesis in\ncomputer vision. Instead of resimulating pixels based on higher-order patterns\nin the training image, we identify a suitable patch of the training image that\nreplace a corresponding patch in the current model without breaking the\npatterns found in the training image, that is, remaining consistent with the\ngiven prior. We consider three cross-hole ground-penetrating radar examples in\nwhich the new model proposal mechanism is employed within an extended\nMetropolis Markov chain Monte Carlo (MCMC) inversion. The model proposal step\nis about 40 times faster than state-of-the-art multiple-point statistics\nresimulation techniques, the number of necessary MCMC steps is lower and the\nquality of the final model realizations is of similar quality. The model\nproposal mechanism is presently limited to 2-D fields, but the method is\ngeneral and can be applied to a wide range of subsurface settings and\ngeophysical data types.\n",
"title": "Image synthesis with graph cuts: a fast model proposal mechanism in probabilistic inversion"
}
| null | null | null | null | true | null |
4444
| null |
Default
| null | null |
null |
{
"abstract": " We provide a new proof of the super duality equivalence between infinite-rank\nparabolic BGG categories of general linear Lie (super) algebras conjectured by\nCheng and Wang and first proved by Cheng and Lam. We do this by establishing a\nnew uniqueness theorem for tensor product categorifications motivated by work\nof Brundan, Losev, and Webster. Moreover we show that these BGG categories have\nKoszul graded lifts and super duality can be lifted to a graded equivalence.\n",
"title": "Graded super duality for general linear Lie superalgebras"
}
| null | null |
[
"Mathematics"
] | null | true | null |
4445
| null |
Validated
| null | null |
null |
{
"abstract": " In recent years, car makers and tech companies have been racing towards self\ndriving cars. It seems that the main parameter in this race is who will have\nthe first car on the road. The goal of this paper is to add to the equation two\nadditional crucial parameters. The first is standardization of safety assurance\n--- what are the minimal requirements that every self-driving car must satisfy,\nand how can we verify these requirements. The second parameter is scalability\n--- engineering solutions that lead to unleashed costs will not scale to\nmillions of cars, which will push interest in this field into a niche academic\ncorner, and drive the entire field into a \"winter of autonomous driving\". In\nthe first part of the paper we propose a white-box, interpretable, mathematical\nmodel for safety assurance, which we call Responsibility-Sensitive Safety\n(RSS). In the second part we describe a design of a system that adheres to our\nsafety assurance requirements and is scalable to millions of cars.\n",
"title": "On a Formal Model of Safe and Scalable Self-driving Cars"
}
| null | null | null | null | true | null |
4446
| null |
Default
| null | null |
null |
{
"abstract": " An electrically-controllable, solid-state, reversible device for sourcing and\nsinking alkali vapor is presented. When placed inside an alkali vapor cell,\nboth an increase and decrease of the rubidium vapor density by a factor of two\nare demonstrated through laser absorption spectroscopy on 10 to 15 s time\nscales. The device requires low voltage (5 V), low power (<3.4 mW peak power),\nand low energy (<10.7 mJ per 10 s pulse). The absence of oxygen emission during\noperation is shown through residual gas analysis, indicating Rb is not lost\nthrough chemical reaction but rather by ion transport through the designed\nchannel. This device is of interest for atomic physics experiments and, in\nparticular, for portable cold-atom systems where dynamic control of alkali\nvapor density can enable advances in science and technology.\n",
"title": "A Low-power Reversible Alkali Atom Source"
}
| null | null |
[
"Physics"
] | null | true | null |
4447
| null |
Validated
| null | null |
null |
{
"abstract": " Global pairwise network alignment (GPNA) aims to find a one-to-one node\nmapping between two networks that identifies conserved network regions. GPNA\nalgorithms optimize node conservation (NC) and edge conservation (EC). NC\nquantifies topological similarity between nodes. Graphlet-based degree vectors\n(GDVs) are a state-of-the-art topological NC measure. Dynamic GDVs (DGDVs) were\nused as a dynamic NC measure within the first-ever algorithms for GPNA of\ntemporal networks: DynaMAGNA++ and DynaWAVE. The latter is superior for larger\nnetworks. We recently developed a different graphlet-based measure of temporal\nnode similarity, graphlet-orbit transitions (GoTs). Here, we use GoTs instead\nof DGDVs as a new dynamic NC measure within DynaWAVE, resulting in a new\napproach, GoT-WAVE.\nOn synthetic networks, GoT-WAVE improves DynaWAVE's accuracy by 25% and speed\nby 64%. On real networks, when optimizing only dynamic NC, each method is\nsuperior ~50% of the time. While DynaWAVE benefits more from also optimizing\ndynamic EC, only GoT-WAVE can support directed edges. Hence, GoT-WAVE is a\npromising new temporal GPNA algorithm, which efficiently optimizes dynamic NC.\nFuture work on better incorporating dynamic EC may yield further improvements.\n",
"title": "GoT-WAVE: Temporal network alignment using graphlet-orbit transitions"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
4448
| null |
Validated
| null | null |
null |
{
"abstract": " We present a quantu spin liquid state in a spin-1/2 honeycomb lattice with\nrandomness in the exchange interaction. That is, we successfully introduce\nrandomness into the organic radial-based complex and realize a random-singlet\n(RS) state. All magnetic and thermodynamic experimental results indicate the\nliquid-like behaviors, which are consistent with those expected in the RS\nstate. These results demonstrate that the randomness or inhomogeneity in the\nactual systems stabilize the RS state and yield liquid-like behavior.\n",
"title": "Randomness-induced quantum spin liquid on honeycomb lattice"
}
| null | null |
[
"Physics"
] | null | true | null |
4449
| null |
Validated
| null | null |
null |
{
"abstract": " We show that for any singular dominant integral weight $\\lambda$ of a complex\nsemisimple Lie algebra $\\mathfrak{g}$, the endomorphism algebra $B$ of any\nprojective-injective module of the parabolic BGG category\n$\\mathcal{O}_\\lambda^{\\mathfrak{p}}$ is a symmetric algebra (as conjectured by\nKhovanov) extending the results of Mazorchuk and Stroppel for the regular\ndominant integral weight. Moreover, the endomorphism algebra $B$ is equipped\nwith a homogeneous (non-degenerate) symmetrizing form. In the appendix, there\nis a short proof due to K. Coulembier and V. Mazorchuk showing that the\nendomorphism algebra $B_\\lambda^{\\mathfrak{p}}$ of the basic\nprojective-injective module of $\\mathcal{O}_\\lambda^{\\mathfrak{p}}$ is a\nsymmetric algebra.\n",
"title": "Symmetric structure for the endomorphism algebra of projective-injective module in parabolic category"
}
| null | null |
[
"Mathematics"
] | null | true | null |
4450
| null |
Validated
| null | null |
null |
{
"abstract": " The article is devoted to the investigation of representation of rational\nnumbers by Cantor series. Necessary and sufficient conditions for a rational\nnumber to be representable by a positive Cantor series are formulated for the\ncase of an arbitrary sequence $(q_k)$ and some its corollaries are considered.\nResults of this article were presented by the author of this article on the\nInternational Conference on Algebra dedicated to 100th anniversary of S. M.\nChernikov (www.researchgate.net/publication/311415815,\nwww.researchgate.net/publication/301849984). This investigation was also\npresented in some reports (links to the reports:\nwww.researchgate.net/publication/303736670,\nwww.researchgate.net/publication/303720573, etc.).\n",
"title": "Cantor series and rational numbers"
}
| null | null | null | null | true | null |
4451
| null |
Default
| null | null |
null |
{
"abstract": " Many real-world systems are profitably described as complex networks that\ngrow over time. Preferential attachment and node fitness are two simple growth\nmechanisms that not only explain certain structural properties commonly\nobserved in real-world systems, but are also tied to a number of applications\nin modeling and inference. While there are statistical packages for estimating\nvarious parametric forms of the preferential attachment function, there is no\nsuch package implementing non-parametric estimation procedures. The\nnon-parametric approach to the estimation of the preferential attachment\nfunction allows for comparatively finer-grained investigations of the\n`rich-get-richer' phenomenon that could lead to novel insights in the search to\nexplain certain nonstandard structural properties observed in real-world\nnetworks. This paper introduces the R package PAFit, which implements\nnon-parametric procedures for estimating the preferential attachment function\nand node fitnesses in a growing network, as well as a number of functions for\ngenerating complex networks from these two mechanisms. The main computational\npart of the package is implemented in C++ with OpenMP to ensure scalability to\nlarge-scale networks. We first introduce the main functionalities of PAFit\nthrough simulated examples, and then use the package to analyze a collaboration\nnetwork between scientists in the field of complex networks. The results\nindicate the joint presence of `rich-get-richer' and `fit-get-richer' phenomena\nin the collaboration network. The estimated attachment function is observed to\nbe near-linear, which we interpret as meaning that the chance an author gets a\nnew collaborator is proportional to their current number of collaborators.\nFurthermore, the estimated author fitnesses reveal a host of familiar faces\nfrom the complex networks community among the field's topmost fittest network\nscientists.\n",
"title": "PAFit: an R Package for the Non-Parametric Estimation of Preferential Attachment and Node Fitness in Temporal Complex Networks"
}
| null | null | null | null | true | null |
4452
| null |
Default
| null | null |
null |
{
"abstract": " Given a semi-Riemannian $4$-manifold $(M,g)$ with two distinguished vector\nfields satisfying properties determined by their shear, twist and various Lie\nbracket relations, a family of Kähler metrics $g_K$ is constructed, defined\non an open set in $M$, which coincides with $M$ in many typical examples. Under\ncertain conditions $g$ and $g_K$ share various properties, such as a Killing\nvector field or a vector field with a geodesic flow. In some cases the Kähler\nmetrics are complete. The Ricci and scalar curvatures of $g_K$ are computed\nunder certain assumptions in terms of data associated to $g$. Many examples are\ndescribed, including classical spacetimes in warped products, for instance de\nSitter spacetime, as well as gravitational plane waves, metrics of Petrov type\n$D$ such as Kerr and NUT metrics, and metrics for which $g_K$ is an SKR metric.\nFor the latter an inverse ansatz is described, constructing $g$ from the SKR\nmetric.\n",
"title": "Kähler metrics via Lorentzian Geometry in dimension four"
}
| null | null | null | null | true | null |
4453
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we compute the number of z-classes (conjugacy classes of\ncentralizers of elements) in the symmetric group S_n, when n is greater or\nequal to 3 and alternating group A_n, when n is greater or equal to 4. It turns\nout that the difference between the number of conjugacy classes and the number\nof z-classes for S_n is determined by those restricted partitions of n-2 in\nwhich 1 and 2 do not appear as its part. And, in the case of alternating\ngroups, it is determined by those restricted partitions of n-3 which has all\nits parts distinct, odd and in which 1 (and 2) does not appear as its part,\nalong with an error term. The error term is given by those partitions of n\nwhich have each of its part distinct, odd and perfect square. Further, we prove\nthat the number of rational-valued irreducible complex characters for A_n is\nsame as the number of conjugacy classes which are rational.\n",
"title": "z-Classes and Rational Conjugacy Classes in Alternating Groups"
}
| null | null | null | null | true | null |
4454
| null |
Default
| null | null |
null |
{
"abstract": " This paper proposes a practical approach for automatic sleep stage\nclassification based on a multi-level feature learning framework and Recurrent\nNeural Network (RNN) classifier using heart rate and wrist actigraphy derived\nfrom a wearable device. The feature learning framework is designed to extract\nlow- and mid-level features. Low-level features capture temporal and frequency\ndomain properties and mid-level features learn compositions and structural\ninformation of signals. Since sleep staging is a sequential problem with\nlong-term dependencies, we take advantage of RNNs with Bidirectional Long\nShort-Term Memory (BLSTM) architectures for sequence data learning. To simulate\nthe actual situation of daily sleep, experiments are conducted with a resting\ngroup in which sleep is recorded in resting state, and a comprehensive group in\nwhich both resting sleep and non-resting sleep are included.We evaluate the\nalgorithm based on an eight-fold cross validation to classify five sleep stages\n(W, N1, N2, N3, and REM). The proposed algorithm achieves weighted precision,\nrecall and F1 score of 58.0%, 60.3%, and 58.2% in the resting group and 58.5%,\n61.1%, and 58.5% in the comprehensive group, respectively. Various comparison\nexperiments demonstrate the effectiveness of feature learning and BLSTM. We\nfurther explore the influence of depth and width of RNNs on performance. Our\nmethod is specially proposed for wearable devices and is expected to be\napplicable for long-term sleep monitoring at home. Without using too much prior\ndomain knowledge, our method has the potential to generalize sleep disorder\ndetection.\n",
"title": "Sleep Stage Classification Based on Multi-level Feature Learning and Recurrent Neural Networks via Wearable Device"
}
| null | null | null | null | true | null |
4455
| null |
Default
| null | null |
null |
{
"abstract": " Martin David Kruskal was one of the most versatile theoretical physicists of\nhis generation and is distinguished for his enduring work in several different\nareas, most notably plasma physics, a memorable detour into relativity, and his\npioneering work in nonlinear waves. In the latter, together with Norman\nZabusky, he invented the concept of the soliton and, with others, developed its\napplication to classes of partial differential equations of physical\nsignificance.\n",
"title": "Martin David Kruskal: a biographical memoir"
}
| null | null | null | null | true | null |
4456
| null |
Default
| null | null |
null |
{
"abstract": " We provide a unified framework to compute the stationary distribution of any\nfinite irreducible Markov chain or equivalently of any irreducible random walk\non a finite semigroup $S$. Our methods use geometric finite semigroup theory\nvia the Karnofsky-Rhodes and the McCammond expansions of finite semigroups with\nspecified generators; this does not involve any linear algebra. The original\nTsetlin library is obtained by applying the expansions to $P(n)$, the set of\nall subsets of an $n$ element set. Our set-up generalizes previous\ngroundbreaking work involving left-regular bands (or $\\mathscr{R}$-trivial\nbands) by Brown and Diaconis, extensions to $\\mathscr{R}$-trivial semigroups by\nAyyer, Steinberg, Thiéry and the second author, and important recent work by\nChung and Graham. The Karnofsky-Rhodes expansion of the right Cayley graph of\n$S$ in terms of generators yields again a right Cayley graph. The McCammond\nexpansion provides normal forms for elements in the expanded $S$. Using our\nprevious results with Silva based on work by Berstel, Perrin, Reutenauer, we\nconstruct (infinite) semaphore codes on which we can define Markov chains.\nThese semaphore codes can be lumped using geometric semigroup theory. Using\nnormal forms and associated Kleene expressions, they yield formulas for the\nstationary distribution of the finite Markov chain of the expanded $S$ and the\noriginal $S$. Analyzing the normal forms also provides an estimate on the\nmixing time.\n",
"title": "Unified theory for finite Markov chains"
}
| null | null | null | null | true | null |
4457
| null |
Default
| null | null |
null |
{
"abstract": " Hedonic games are meant to model how coalitions of people form and break\napart in the real world. However, it is difficult to run simulations when\neverything must be done by hand on paper. We present an online software that\nallows fast and visual simulation of several types of hedonic games.\nthis http URL\n",
"title": "A Simulator for Hedonic Games"
}
| null | null | null | null | true | null |
4458
| null |
Default
| null | null |
null |
{
"abstract": " We introduce the $k$-banded Cholesky prior for estimating a high-dimensional\nbandable precision matrix via the modified Cholesky decomposition. The bandable\nassumption is imposed on the Cholesky factor of the decomposition. We obtained\nthe P-loss convergence rate under the spectral norm and the matrix\n$\\ell_{\\infty}$ norm and the minimax lower bounds. Since the P-loss convergence\nrate (Lee and Lee (2017)) is stronger than the posterior convergence rate, the\nrates obtained are also posterior convergence rates. Furthermore, when the true\nprecision matrix is a $k_0$-banded matrix with some finite $k_0$, the obtained\nP-loss convergence rates coincide with the minimax rates. The established\nconvergence rates are slightly slower than the minimax lower bounds, but these\nare the fastest rates for bandable precision matrices among the existing\nBayesian approaches. A simulation study is conducted to compare the performance\nto the other competitive estimators in various scenarios.\n",
"title": "Estimating Large Precision Matrices via Modified Cholesky Decomposition"
}
| null | null | null | null | true | null |
4459
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we consider the stochastic Langevin equation with additive\nnoises, which possesses both conformal symplectic geometric structure and\nergodicity. We propose a methodology of constructing high weak order conformal\nsymplectic schemes by converting the equation into an equivalent autonomous\nstochastic Hamiltonian system and modifying the associated generating function.\nTo illustrate this approach, we construct a specific second order numerical\nscheme, and prove that its symplectic form dissipates exponentially. Moreover,\nfor the linear case, the proposed scheme is also shown to inherit the\nergodicity of the original system, and the temporal average of the numerical\nsolution is a proper approximation of the ergodic limit over long time.\nNumerical experiments are given to verify these theoretical results.\n",
"title": "High order conformal symplectic and ergodic schemes for stochastic Langevin equation via generating functions"
}
| null | null | null | null | true | null |
4460
| null |
Default
| null | null |
null |
{
"abstract": " Crowdsourcing is an important avenue for collecting machine learning data,\nbut crowdsourcing can go beyond simple data collection by employing the\ncreativity and wisdom of crowd workers. Yet crowd participants are unlikely to\nbe experts in statistics or predictive modeling, and it is not clear how well\nnon-experts can contribute creatively to the process of machine learning. Here\nwe study an end-to-end crowdsourcing algorithm where groups of non-expert\nworkers propose supervised learning problems, rank and categorize those\nproblems, and then provide data to train predictive models on those problems.\nProblem proposal includes and extends feature engineering because workers\npropose the entire problem, not only the input features but also the target\nvariable. We show that workers without machine learning experience can\ncollectively construct useful datasets and that predictive models can be\nlearned on these datasets. In our experiments, the problems proposed by workers\ncovered a broad range of topics, from politics and current events to problems\ncapturing health behavior, demographics, and more. Workers also favored\nquestions showing positively correlated relationships, which has interesting\nimplications given many supervised learning methods perform as well with strong\nnegative correlations. Proper instructions are crucial for non-experts, so we\nalso conducted a randomized trial to understand how different instructions may\ninfluence the types of problems proposed by workers. In general, shifting the\nfocus of machine learning tasks from designing and training individual\npredictive models to problem proposal allows crowdsourcers to design\nrequirements for problems of interest and then guide workers towards\ncontributing to the most suitable problems.\n",
"title": "Crowd ideation of supervised learning problems"
}
| null | null |
[
"Statistics"
] | null | true | null |
4461
| null |
Validated
| null | null |
null |
{
"abstract": " The Weihrauch degrees and strong Weihrauch degrees are partially ordered\nstructures representing degrees of unsolvability of various mathematical\nproblems. Their study has been widely applied in computable analysis,\ncomplexity theory, and more recently, also in computable combinatorics. We\nanswer an open question about the algebraic structure of the strong Weihrauch\ndegrees, by exhibiting a join operation that turns these degrees into a\nlattice. Previously, the strong Weihrauch degrees were only known to form a\nlower semi-lattice. We then show that unlike the Weihrauch degrees, which are\nknown to form a distributive lattice, the lattice of strong Weihrauch degrees\nis not distributive. Therefore, the two structures are not isomorphic.\n",
"title": "Joins in the strong Weihrauch degrees"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
4462
| null |
Validated
| null | null |
null |
{
"abstract": " Long short-term memory (LSTM) is normally used in recurrent neural network\n(RNN) as basic recurrent unit. However,conventional LSTM assumes that the state\nat current time step depends on previous time step. This assumption constraints\nthe time dependency modeling capability. In this study, we propose a new\nvariation of LSTM, advanced LSTM (A-LSTM), for better temporal context\nmodeling. We employ A-LSTM in weighted pooling RNN for emotion recognition. The\nA-LSTM outperforms the conventional LSTM by 5.5% relatively. The A-LSTM based\nweighted pooling RNN can also complement the state-of-the-art emotion\nclassification framework. This shows the advantage of A-LSTM.\n",
"title": "Advanced LSTM: A Study about Better Time Dependency Modeling in Emotion Recognition"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
4463
| null |
Validated
| null | null |
null |
{
"abstract": " We introduce the abstract notion of a closed necklical set in order to\ndescribe a functorial combinatorial model of the free loop fibration $\\Omega\nY\\rightarrow \\Lambda Y\\rightarrow Y$ over the geometric realization $Y=|X|$ of\na path connected simplicial set $X.$ In particular, to any path connected\nsimplicial set $X$ we associate a closed necklical set\n$\\widehat{\\mathbf{\\Lambda}}X$ such that its geometric realization\n$|\\widehat{\\mathbf{\\Lambda}}X|$, a space built out of gluing \"freehedrical\" and\n\"cubical\" cells, is homotopy equivalent to the free loop space $\\Lambda Y$ and\nthe differential graded module of chains $C_*(\\widehat{\\mathbf{\\Lambda}}X)$\ngeneralizes the coHochschild chain complex of the chain coalgebra $C_\\ast(X).$\n",
"title": "A combinatorial model for the free loop fibration"
}
| null | null | null | null | true | null |
4464
| null |
Default
| null | null |
null |
{
"abstract": " We give a classification and complete algebraic description of groups\nallowing only finitely many (left multiplication invariant) circular orders. In\nparticular, they are all solvable groups with a specific semi-direct product\ndecomposition. This allows us to also show that the space of circular orders of\nany group is either finite or uncountable. As a special case and first step, we\nshow that the space of circular orderings of an infinite Abelian group has no\nisolated points, hence is homeomorphic to a cantor set.\n",
"title": "On the number of circular orders on a group"
}
| null | null | null | null | true | null |
4465
| null |
Default
| null | null |
null |
{
"abstract": " Quantum Cognition has delivered a number of models for semantic memory, but\nto date these have tended to assume pure states and projective measurement.\nHere we relax these assumptions. A quantum inspired model of human word\nassociation experiments will be extended using a density matrix representation\nof human memory and a POVM based upon non-ideal measurements. Our formulation\nallows for a consideration of key terms like measurement and contextuality\nwithin a rigorous modern approach. This approach both provides new conceptual\nadvances and suggests new experimental protocols.\n",
"title": "Preparation and Measurement in Quantum Memory Models"
}
| null | null | null | null | true | null |
4466
| null |
Default
| null | null |
null |
{
"abstract": " The purpose of this note is to provide a detailed proof of Nazarov's\ninequality stated in Lemma A.1 in Chernozhukov, Chetverikov, and Kato (2017,\nAnnals of Probability).\n",
"title": "Detailed proof of Nazarov's inequality"
}
| null | null | null | null | true | null |
4467
| null |
Default
| null | null |
null |
{
"abstract": " The Lanczos method is one of the standard approaches for computing a few\neigenpairs of a large, sparse, symmetric matrix. It is typically used with\nrestarting to avoid unbounded growth of memory and computational requirements.\nThick-restart Lanczos is a popular restarted variant because of its simplicity\nand numerically robustness. However, convergence can be slow for highly\nclustered eigenvalues so more effective restarting techniques and the use of\npreconditioning is needed. In this paper, we present a thick-restart\npreconditioned Lanczos method, TRPL+K, that combines the power of locally\noptimal restarting (+K) and preconditioning techniques with the efficiency of\nthe thick-restart Lanczos method. TRPL+K employs an inner-outer scheme where\nthe inner loop applies Lanczos on a preconditioned operator while the outer\nloop augments the resulting Lanczos subspace with certain vectors from the\nprevious restart cycle to obtain eigenvector approximations with which it thick\nrestarts the outer subspace. We first identify the differences from various\nrelevant methods in the literature. Then, based on an optimization perspective,\nwe show an asymptotic global quasi-optimality of a simplified TRPL+K method\ncompared to an unrestarted global optimal method. Finally, we present extensive\nexperiments showing that TRPL+K either outperforms or matches other\nstate-of-the-art eigenmethods in both matrix-vector multiplications and\ncomputational time.\n",
"title": "TRPL+K: Thick-Restart Preconditioned Lanczos+K Method for Large Symmetric Eigenvalue Problems"
}
| null | null | null | null | true | null |
4468
| null |
Default
| null | null |
null |
{
"abstract": " Eigenstates of fully many-body localized (FMBL) systems are described by\nquasilocal operators $\\tau_i^z$ (l-bits), which are conserved exactly under\nHamiltonian time evolution. The algebra of the operators $\\tau_i^z$ and\n$\\tau_i^x$ associated with l-bits ($\\boldsymbol{\\tau}_i$) completely defines\nthe eigenstates and the matrix elements of local operators between eigenstates\nat all energies. We develop a non-perturbative construction of the full set of\nl-bit algebras in the many-body localized phase for the canonical model of MBL.\nOur algorithm to construct the Pauli-algebra of l-bits combines exact\ndiagonalization and a tensor network algorithm developed for efficient\ndiagonalization of large FMBL Hamiltonians. The distribution of localization\nlengths of the l-bits is evaluated in the MBL phase and used to characterize\nthe MBL-to-thermal transition.\n",
"title": "Behavior of l-bits near the many-body localization transition"
}
| null | null | null | null | true | null |
4469
| null |
Default
| null | null |
null |
{
"abstract": " Network modeling has become increasingly popular for analyzing genomic data,\nto aid in the interpretation and discovery of possible mechanistic components\nand therapeutic targets. However, genomic-scale networks are high-dimensional\nmodels and are usually estimated from a relatively small number of samples.\nTherefore, their usefulness is hampered by estimation instability. In addition,\nthe complexity of the models is controlled by one or more penalization (tuning)\nparameters where small changes to these can lead to vastly different networks,\nthus making interpretation of models difficult. This necessitates the\ndevelopment of techniques to produce robust network models accompanied by\nestimation quality assessments.\nWe introduce Resampling of Penalized Estimates (ROPE): a novel statistical\nmethod for robust network modeling. The method utilizes resampling-based\nnetwork estimation and integrates results from several levels of penalization\nthrough a constrained, over-dispersed beta-binomial mixture model. ROPE\nprovides robust False Discovery Rate (FDR) control of network estimates and\neach edge is assigned a measure of validity, the q-value, corresponding to the\nFDR-level for which the edge would be included in the network model. We apply\nROPE to several simulated data sets as well as genomic data from The Cancer\nGenome Atlas. We show that ROPE outperforms state-of-the-art methods in terms\nof FDR control and robust performance across data sets. We illustrate how to\nuse ROPE to make a principled model selection for which genomic associations to\nstudy further. ROPE is available as an R package on CRAN.\n",
"title": "ROPE: high-dimensional network modeling with robust control of edge FDR"
}
| null | null | null | null | true | null |
4470
| null |
Default
| null | null |
null |
{
"abstract": " Automated vehicles can change the society by improved safety, mobility and\nfuel efficiency. However, due to the higher cost and change in business model,\nover the coming decades, the highly automated vehicles likely will continue to\ninteract with many human-driven vehicles. In the past, the control/design of\nthe highly automated (robotic) vehicles mainly considers safety and efficiency\nbut failed to address the \"driving culture\" of surrounding human-driven\nvehicles. Thus, the robotic vehicles may demonstrate behaviors very different\nfrom other vehicles. We study this \"driving etiquette\" problem in this paper.\nAs the first step, we report the key behavior parameters of human driven\nvehicles derived from a large naturalistic driving database. The results can be\nused to guide future algorithm design of highly automated vehicles or to\ndevelop realistic human-driven vehicle behavior model in simulations.\n",
"title": "Developing Robot Driver Etiquette Based on Naturalistic Human Driving Behavior"
}
| null | null | null | null | true | null |
4471
| null |
Default
| null | null |
null |
{
"abstract": " In recent years, many new and interesting models of successful online\nbusiness have been developed, including competitive models such as auctions,\nwhere the product price tends to rise, and group-buying, where users cooperate\nobtaining a dynamic price that tends to go down. We propose the e-fair as a\nbusiness model for social commerce, where both sellers and buyers are grouped\nto maximize benefits. e-Fairs extend the group-buying model aggregating demand\nand supply for price optimization as well as consolidating shipments and\noptimize withdrawals for guaranteeing additional savings. e-Fairs work upon\nmultiple dimensions: time to aggregate buyers, their geographical distribution,\nprice/quantity curves provided by sellers, and location of withdrawal points.\nWe provide an analytical model for time and spatial optimization and simulate\nrealistic scenarios using both real purchase data from an Italian marketplace\nand simulated ones. Experimental results demonstrate the potentials offered by\ne-fairs and show benefits for all the involved actors.\n",
"title": "e-Fair: Aggregation in e-Commerce for Exploiting Economies of Scale"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4472
| null |
Validated
| null | null |
null |
{
"abstract": " We consider the Gierer-Meinhardt system with small inhibitor diffusivity,\nvery small activator diffusivity and a precursor inhomogeneity.\nFor any given positive integer k we construct a spike cluster consisting of\n$k$ spikes which all approach the same nondegenerate local minimum point of the\nprecursor inhomogeneity. We show that this spike cluster can be linearly\nstable. In particular, we show the existence of spike clusters for spikes\nlocated at the vertices of a polygon with or without centre. Further, the\ncluster without centre is stable for up to three spikes, whereas the cluster\nwith centre is stable for up to six spikes.\nThe main idea underpinning these stable spike clusters is the following: due\nto the small inhibitor diffusivity the interaction between spikes is repulsive,\nand the spikes are attracted towards the local minimum point of the precursor\ninhomogeneity. Combining these two effects can lead to an equilibrium of spike\npositions within the cluster such that the cluster is linearly stable.\n",
"title": "Stable spike clusters for the precursor Gierer-Meinhardt system in R2"
}
| null | null | null | null | true | null |
4473
| null |
Default
| null | null |
null |
{
"abstract": " In this article, we study subloci of solvable curves in $\\mathcal{M}_g$ which\nare contained in either a K3-surface or a quadric or a cubic surface. We give a\nbound on the dimension of such subloci. In the case of complete intersection\ngenus $g$ curves in a cubic surface, we show that a general such curve is\nsolvable.\n",
"title": "Solvability of curves on surfaces"
}
| null | null | null | null | true | null |
4474
| null |
Default
| null | null |
null |
{
"abstract": " We prove limit theorems for the super-replication cost of European options in\na Binomial model with transient price impact. We show that if the time step\ngoes to zero and the effective resilience between consecutive trading times\nremains constant then the limit of the super--replication prices coincide with\nthe scaling limit for temporary price impact with a modified market depth.\n",
"title": "Scaling Limits for Super--replication with Transient Price Impact"
}
| null | null |
[
"Quantitative Finance"
] | null | true | null |
4475
| null |
Validated
| null | null |
null |
{
"abstract": " The study of networks has witnessed an explosive growth over the past decades\nwith several ground-breaking methods introduced. A particularly interesting --\nand prevalent in several fields of study -- problem is that of inferring a\nfunction defined over the nodes of a network. This work presents a versatile\nkernel-based framework for tackling this inference problem that naturally\nsubsumes and generalizes the reconstruction approaches put forth recently by\nthe signal processing on graphs community. Both the static and the dynamic\nsettings are considered along with effective modeling approaches for addressing\nreal-world problems. The herein analytical discussion is complemented by a set\nof numerical examples, which showcase the effectiveness of the presented\ntechniques, as well as their merits related to state-of-the-art methods.\n",
"title": "Kernel-based Inference of Functions over Graphs"
}
| null | null | null | null | true | null |
4476
| null |
Default
| null | null |
null |
{
"abstract": " We present natural families of coordinate algebras of noncommutative products\nof Euclidean spaces. These coordinate algebras are quadratic ones associated\nwith an R-matrix which is involutive and satisfies the Yang-Baxter equations.\nAs a consequence they enjoy a list of nice properties, being regular of finite\nglobal dimension. Notably, we have eight-dimensional noncommutative euclidean\nspaces which are particularly well behaved and are deformations parametrised by\na two-dimensional sphere. Quotients include noncommutative seven-spheres as\nwell as noncommutative \"quaternionic tori\". There is invariance for an action\nof $SU(2) \\times SU(2)$ in parallel with the action of $U(1) \\times U(1)$ on a\n\"complex\" noncommutative torus which allows one to construct quaternionic toric\nnoncommutative manifolds. Additional classes of solutions are disjoint from the\nclassical case.\n",
"title": "Noncommutative products of Euclidean spaces"
}
| null | null | null | null | true | null |
4477
| null |
Default
| null | null |
null |
{
"abstract": " We observed the field of the Fermi source 3FGL J0838.8-2829 in optical and\nX-rays, initially motivated by the cataclysmic variable (CV) 1RXS\nJ083842.1-282723 that lies within its error circle. Several X-ray sources first\nclassified as CVs have turned out to be gamma-ray emitting millisecond pulsars\n(MSPs). We find that 1RXS J083842.1-282723 is in fact an unusual CV, a\nstream-fed asynchronous polar in which accretion switches between magnetic\npoles (that are $\\approx$120$^{\\circ}$ apart) when the accretion rate is at\nminimum. High-amplitude X-ray modulation at periods of 94.8$\\pm$0.4 minutes and\n14.7$\\pm$1.2 hr are seen. The former appears to be the spin period, while\nlatter is inferred to be one-third of the beat period between the spin and the\norbit, implying an orbital period of 98.3$\\pm$0.5 minutes. We also measure an\noptical emission-line spectroscopic period of 98.413$\\pm$0.004 minutes which is\nconsistent with the orbital period inferred from the X-rays. In any case, this\nsystem is unlikely to be the gamma-ray source. Instead, we find a fainter\nvariable X-ray and optical source, XMMU J083850.38-282756.8, that is modulated\non a time scale of hours in addition to exhibiting occasional sharp flares. It\nresembles the black widow or redback pulsars that have been discovered as\ncounterparts of Fermi sources, with the optical modulation due to heating of\nthe photosphere of a low-mass companion star by, in this case, an as-yet\nundetected MSP. We propose XMMU J083850.38-282756.8 as the MSP counterpart of\n3FGL J0838.8-2829.\n",
"title": "X-ray and Optical Study of the Gamma-ray Source 3FGL J0838.8$-$2829: Identification of a Candidate Millisecond Pulsar Binary and an Asynchronous Polar"
}
| null | null | null | null | true | null |
4478
| null |
Default
| null | null |
null |
{
"abstract": " Markov random fields (MRFs) find applications in a variety of machine\nlearning areas, while the inference and learning of such models are challenging\nin general. In this paper, we propose the Adversarial Variational Inference and\nLearning (AVIL) algorithm to solve the problems with a minimal assumption about\nthe model structure of an MRF. AVIL employs two variational distributions to\napproximately infer the latent variables and estimate the partition function,\nrespectively. The variational distributions, which are parameterized as neural\nnetworks, provide an estimate of the negative log likelihood of the MRF. On one\nhand, the estimate is in an intuitive form of approximate contrastive free\nenergy. On the other hand, the estimate is a minimax optimization problem,\nwhich is solved by stochastic gradient descent in an alternating manner. We\napply AVIL to various undirected generative models in a fully black-box manner\nand obtain better results than existing competitors on several real datasets.\n",
"title": "Adversarial Variational Inference and Learning in Markov Random Fields"
}
| null | null | null | null | true | null |
4479
| null |
Default
| null | null |
null |
{
"abstract": " Nonparametric estimation of mutual information is used in a wide range of\nscientific problems to quantify dependence between variables. The k-nearest\nneighbor (knn) methods are consistent, and therefore expected to work well for\nlarge sample size. These methods use geometrically regular local volume\nelements. This practice allows maximum localization of the volume elements, but\ncan also induce a bias due to a poor description of the local geometry of the\nunderlying probability measure. We introduce a new class of knn estimators that\nwe call geometric knn estimators (g-knn), which use more complex local volume\nelements to better model the local geometry of the probability measures. As an\nexample of this class of estimators, we develop a g-knn estimator of entropy\nand mutual information based on elliptical volume elements, capturing the local\nstretching and compression common to a wide range of dynamical systems\nattractors. A series of numerical examples in which the thickness of the\nunderlying distribution and the sample sizes are varied suggest that local\ngeometry is a source of problems for knn methods such as the\nKraskov-Stögbauer-Grassberger (KSG) estimator when local geometric effects\ncannot be removed by global preprocessing of the data. The g-knn method\nperforms well despite the manipulation of the local geometry. In addition, the\nexamples suggest that the g-knn estimators can be of particular relevance to\napplications in which the system is large, but data size is limited.\n",
"title": "Geometric k-nearest neighbor estimation of entropy and mutual information"
}
| null | null | null | null | true | null |
4480
| null |
Default
| null | null |
null |
{
"abstract": " We present laboratory spectra of the $3p$--$3d$ transitions in Fe$^{14+}$ and\nFe$^{15+}$ excited with a mono-energetic electron beam. In the energy dependent\nspectra obtained by sweeping the electron energy, resonant excitation is\nconfirmed as an intensity enhancement at specific electron energies. The\nexperimental results are compared with theoretical cross sections calculated\nbased on fully relativistic wave functions and the distorted-wave\napproximation. Comparisons between the experimental and theoretical results\nshow good agreement for the resonance strength. A significant discrepancy is,\nhowever, found for the non-resonant cross section in Fe$^{14+}$. %, which can\nbe considered as a fundamental cause of the line intensity ratio problem that\nhas often been found in both observatory and laboratory measurements. This\ndiscrepancy is considered to be the fundamental cause of the previously\nreported inconsistency of the model with the observed intensity ratio between\nthe $^3\\!P_2$ -- $^3\\!D_3$ and $^1\\!P_1$ -- $^1\\!D_2$ transitions.\n",
"title": "Resonant Electron Impact Excitation of 3d levels in Fe$^{14+}$ and Fe$^{15+}$"
}
| null | null | null | null | true | null |
4481
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we propose a novel method to register football broadcast video\nframes on the static top view model of the playing surface. The proposed method\nis fully automatic in contrast to the current state of the art which requires\nmanual initialization of point correspondences between the image and the static\nmodel. Automatic registration using existing approaches has been difficult due\nto the lack of sufficient point correspondences. We investigate an alternate\napproach exploiting the edge information from the line markings on the field.\nWe formulate the registration problem as a nearest neighbour search over a\nsynthetically generated dictionary of edge map and homography pairs. The\nsynthetic dictionary generation allows us to exhaustively cover a wide variety\nof camera angles and positions and reduce this problem to a minimal per-frame\nedge map matching procedure. We show that the per-frame results can be improved\nin videos using an optimization framework for temporal camera stabilization. We\ndemonstrate the efficacy of our approach by presenting extensive results on a\ndataset collected from matches of football World Cup 2014.\n",
"title": "Automated Top View Registration of Broadcast Football Videos"
}
| null | null | null | null | true | null |
4482
| null |
Default
| null | null |
null |
{
"abstract": " Fabrication of devices in industrial plants often includes undergoing quality\nassurance tests or tests that seek to determine some attributes or capacities\nof the device. For instance, in testing refrigeration compressors, we want to\nfind the true refrigeration capacity of the compressor being tested. Such test\n(also called an episode) may take up to four hours, being an actual hindrance\nto applying it to the total number of compressors produced. This work seeks to\nreduce the time spent on such industrial trials by employing Recurrent Neural\nNetworks (RNNs) as dynamical models for detecting when a test is entering the\nso-called steady-state region. Specifically, we use Reservoir Computing (RC)\nnetworks which simplify the learning of RNNs by speeding up training time and\nshowing convergence to a global optimum. Also, this work proposes a\nself-organized subspace projection method for RC networks which uses\ninformation from the beginning of the episode to define a cluster to which the\nepisode belongs to. This assigned cluster defines a particular binary input\nthat shifts the operating point of the reservoir to a subspace of trajectories\nfor the duration of the episode. This new method is shown to turn the RC model\nrobust in performance with respect to varying combination of reservoir\nparameters, such as spectral radius and leak rate, when compared to a standard\nRC network.\n",
"title": "Reservoir Computing for Detection of Steady State in Performance Tests of Compressors"
}
| null | null | null | null | true | null |
4483
| null |
Default
| null | null |
null |
{
"abstract": " We propose a fast and accurate numerical method for pricing European\nswaptions in multi-factor Gaussian term structure models. Our method can be\nused to accelerate the calibration of such models to the volatility surface.\nThe pricing of an interest rate option in such a model involves evaluating a\nmulti-dimensional integral of the payoff of the claim on a domain where the\npayoff is positive. In our method, we approximate the exercise boundary of the\nstate space by a hyperplane tangent to the maximum probability point on the\nboundary and simplify the multi-dimensional integration into an analytical\nform. The maximum probability point can be determined using the gradient\ndescent method. We demonstrate that our method is superior to previous methods\nby comparing the results to the price obtained by numerical integration.\n",
"title": "Fast swaption pricing in Gaussian term structure models"
}
| null | null |
[
"Quantitative Finance"
] | null | true | null |
4484
| null |
Validated
| null | null |
null |
{
"abstract": " In this article, we extend the conventional framework of\nconvolutional-Restricted-Boltzmann-Machine to learn highly abstract features\namong abitrary number of time related input maps by constructing a layer of\nmultiplicative units, which capture the relations among inputs. In many cases,\nmore than two maps are strongly related, so it is wise to make multiplicative\nunit learn relations among more input maps, in other words, to find the optimal\nrelational-order of each unit. In order to enable our machine to learn\nrelational order, we developed a reinforcement-learning method whose optimality\nis proven to train the network.\n",
"title": "Temporal-related Convolutional-Restricted-Boltzmann-Machine capable of learning relational order via reinforcement learning procedure?"
}
| null | null | null | null | true | null |
4485
| null |
Default
| null | null |
null |
{
"abstract": " A major obstacle to understanding neural coding and computation is the fact\nthat experimental recordings typically sample only a small fraction of the\nneurons in a circuit. Measured neural properties are skewed by interactions\nbetween recorded neurons and the \"hidden\" portion of the network. To properly\ninterpret neural data and determine how biological structure gives rise to\nneural circuit function, we thus need a better understanding of the\nrelationships between measured effective neural properties and the true\nunderlying physiological properties. Here, we focus on how the effective\nspatiotemporal dynamics of the synaptic interactions between neurons are\nreshaped by coupling to unobserved neurons. We find that the effective\ninteractions from a pre-synaptic neuron $r'$ to a post-synaptic neuron $r$ can\nbe decomposed into a sum of the true interaction from $r'$ to $r$ plus\ncorrections from every directed path from $r'$ to $r$ through unobserved\nneurons. Importantly, the resulting formula reveals when the hidden units\nhave---or do not have---major effects on reshaping the interactions among\nobserved neurons. As a particular example of interest, we derive a formula for\nthe impact of hidden units in random networks with \"strong\"\ncoupling---connection weights that scale with $1/\\sqrt{N}$, where $N$ is the\nnetwork size, precisely the scaling observed in recent experiments. With this\nquantitative relationship between measured and true interactions, we can study\nhow network properties shape effective interactions, which properties are\nrelevant for neural computations, and how to manipulate effective interactions.\n",
"title": "Predicting how and when hidden neurons skew measured synaptic interactions"
}
| null | null | null | null | true | null |
4486
| null |
Default
| null | null |
null |
{
"abstract": " The \"Loving AI\" project involves developing software enabling humanoid robots\nto interact with people in loving and compassionate ways, and to promote\npeople' self-understanding and self-transcendence. Currently the project\ncenters on the Hanson Robotics robot \"Sophia\" -- specifically, on supplying\nSophia with personality content and cognitive, linguistic, perceptual and\nbehavioral content aimed at enabling loving interactions supportive of human\nself-transcendence. In September 2017 a small pilot study was conducted,\ninvolving the Sophia robot leading human subjects through dialogues and\nexercises focused on meditation, visualization and relaxation. The pilot was an\napparent success, qualitatively demonstrating the viability of the approach and\nthe ability of appropriate human-robot interaction to increase human well-being\nand advance human consciousness.\n",
"title": "Humanoid Robots as Agents of Human Consciousness Expansion"
}
| null | null | null | null | true | null |
4487
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we study leveraging confidence information induced by\nadversarial training to reinforce adversarial robustness of a given\nadversarially trained model. A natural measure of confidence is $\\|F({\\bf\nx})\\|_\\infty$ (i.e. how confident $F$ is about its prediction?). We start by\nanalyzing an adversarial training formulation proposed by Madry et al.. We\ndemonstrate that, under a variety of instantiations, an only somewhat good\nsolution to their objective induces confidence to be a discriminator, which can\ndistinguish between right and wrong model predictions in a neighborhood of a\npoint sampled from the underlying distribution. Based on this, we propose\nHighly Confident Near Neighbor (${\\tt HCNN}$), a framework that combines\nconfidence information and nearest neighbor search, to reinforce adversarial\nrobustness of a base model. We give algorithms in this framework and perform a\ndetailed empirical study. We report encouraging experimental results that\nsupport our analysis, and also discuss problems we observed with existing\nadversarial training.\n",
"title": "Reinforcing Adversarial Robustness using Model Confidence Induced by Adversarial Training"
}
| null | null | null | null | true | null |
4488
| null |
Default
| null | null |
null |
{
"abstract": " The profitability of fraud in online systems such as app markets and social\nnetworks marks the failure of existing defense mechanisms. In this paper, we\npropose FraudSys, a real-time fraud preemption approach that imposes\nBitcoin-inspired computational puzzles on the devices that post online system\nactivities, such as reviews and likes. We introduce and leverage several novel\nconcepts that include (i) stateless, verifiable computational puzzles, that\nimpose minimal performance overhead, but enable the efficient verification of\ntheir authenticity, (ii) a real-time, graph-based solution to assign fraud\nscores to user activities, and (iii) mechanisms to dynamically adjust puzzle\ndifficulty levels based on fraud scores and the computational capabilities of\ndevices. FraudSys does not alter the experience of users in online systems, but\ndelays fraudulent actions and consumes significant computational resources of\nthe fraudsters. Using real datasets from Google Play and Facebook, we\ndemonstrate the feasibility of FraudSys by showing that the devices of honest\nusers are minimally impacted, while fraudster controlled devices receive daily\ncomputational penalties of up to 3,079 hours. In addition, we show that with\nFraudSys, fraud does not pay off, as a user equipped with mining hardware\n(e.g., AntMiner S7) will earn less than half through fraud than from honest\nBitcoin mining.\n",
"title": "Stateless Puzzles for Real Time Online Fraud Preemption"
}
| null | null |
[
"Computer Science"
] | null | true | null |
4489
| null |
Validated
| null | null |
null |
{
"abstract": " Near-future electric distribution grids operation will have to rely on\ndemand-side flexibility, both by implementation of demand response strategies\nand by taking advantage of the intelligent management of increasingly common\nsmall-scale energy storage. The Home energy management system (HEMS), installed\nat low voltage residential clients, will play a crucial role on the flexibility\nprovision to both system operators and market players like aggregators.\nModeling and forecasting multi-period flexibility from residential prosumers,\nsuch as battery storage and electric water heater, while complying with\ninternal constraints (comfort levels, data privacy) and uncertainty is a\ncomplex task. This papers describes a computational method that is capable of\nefficiently learn and define the feasibility flexibility space from\ncontrollable resources connected to a HEMS. An Evolutionary Particle Swarm\nOptimization (EPSO) algorithm is adopted and reshaped to derive a set of\nfeasible temporal trajectories for the residential net-load, considering\nstorage, flexible appliances, and predefined costumer preferences, as well as\nload and photovoltaic (PV) forecast uncertainty. A support vector data\ndescription (SVDD) algorithm is used to build models capable of classifying\nfeasible and non-feasible HEMS operating trajectories upon request from an\noptimization/control algorithm operated by a DSO or market player.\n",
"title": "Multi-Period Flexibility Forecast for Low Voltage Prosumers"
}
| null | null | null | null | true | null |
4490
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents an approach to assess the economics of customer-sited\nenergy storage systems (ESSs) which are owned and operated by a customer. The\nESSs can participate in frequency regulation and spinning reserve markets, and\nare used to help the customer consume available renewable energy and reduce\nelectricity bill. A rolling-horizon approach is developed to optimize the\nservice schedule, and the resulting costs and revenues are used to assess\neconomics of the ESSs. The economic assessment approach is illustrated with\ncase studies, from which we obtain some new observations on profitability of\nthe customer- sited multi-use ESSs.\n",
"title": "Assessing the Economics of Customer-Sited Multi-Use Energy Storage"
}
| null | null | null | null | true | null |
4491
| null |
Default
| null | null |
null |
{
"abstract": " In this work we study the Thermodynamics of D-dimensional Schwarzschild-anti\nde Sitter (SAdS) black holes. The minimal Thermodynamics of the SAdS spacetime\nis briefly discussed, highlighting some of its strong points and shortcomings.\nThe minimal SAdS Thermodynamics is extended within a Hamiltonian approach, by\nmeans of the introduction of an additional degree of freedom. We demonstrate\nthat the cosmological constant can be introduced in the thermodynamic\ndescription of the SAdS black hole with a canonical transformation of the\nSchwarzschild problem, closely related to the introduction of an anti-de Sitter\nthermodynamic volume. The treatment presented is consistent, in the sense that\nit is compatible with the introduction of new thermodynamic potentials, and\nrespects the laws of black hole Thermodynamics. By demanding homogeneity of the\nthermodynamic variables, we are able to construct a new equation of state that\ncompletely characterizes the Thermodynamics of SAdS black holes. The treatment\nnaturally generates phenomenological constants that can be associated with\ndifferent boundary conditions in underlying microscopic theories. A whole new\nset of phenomena can be expected from the proposed generalization of SAdS\nThermodynamics.\n",
"title": "A Hamiltonian approach for the Thermodynamics of AdS black holes"
}
| null | null | null | null | true | null |
4492
| null |
Default
| null | null |
null |
{
"abstract": " The antiferromagnet (AFM) / ferromagnet (FM) interfaces are of central\nimportance in recently developed pure electric or ultrafast control of FM\nspins, where the underlying mechanisms remain unresolved. Here we report the\ndirect observation of Dzyaloshinskii Moriya interaction (DMI) across the AFM/FM\ninterface of IrMn/CoFeB thin films. The interfacial DMI is quantitatively\nmeasured from the asymmetric spin wave dispersion in the FM layer using\nBrillouin light scattering. The DMI strength is enhanced by a factor of 7 with\nincreasing IrMn layer thickness in the range of 1- 7.5 nm. Our findings provide\ndeeper insight into the coupling at AFM/FM interface and may stimulate new\ndevice concepts utilizing chiral spin textures such as magnetic skyrmions in\nAFM/FM heterostructures.\n",
"title": "Dzyaloshinskii Moriya interaction across antiferromagnet / ferromagnet interface"
}
| null | null | null | null | true | null |
4493
| null |
Default
| null | null |
null |
{
"abstract": " Blockage of pores by particles is found in many processes, including\nfiltration and oil extraction. We present filtration experiments through a\nlinear array of ten channels with one dimension which is sub-micron, through\nwhich a dilute dispersion of Brownian polystyrene spheres flows under the\naction of a fixed pressure drop. The growth rate of a clog formed by particles\nat a pore entrance systematically increases with the number of already\nsaturated (entirely clogged) pores, indicating that there is an interaction or\n\"cross-talk\" between the pores. This observation is interpreted based on a\nphenomenological model, stating that a diffusive redistribution of particles\noccurs along the membrane, from clogged to free pores. This one-dimensional\nmodel could be extended to two-dimensional membranes.\n",
"title": "Pore cross-talk in colloidal filtration"
}
| null | null | null | null | true | null |
4494
| null |
Default
| null | null |
null |
{
"abstract": " Monte Carlo (MC) simulations of transport in random porous networks indicate\nthat for high variances of the log-normal permeability distribution, the\ntransport of a passive tracer is non-Fickian. Here we model this non-Fickian\ndispersion in random porous networks using discrete temporal Markov models. We\nshow that such temporal models capture the spreading behavior accurately. This\nis true despite the fact that the slow velocities are strongly correlated in\ntime, and some studies have suggested that the persistence of low velocities\nwould render the temporal Markovian model inapplicable. Compared to previously\nproposed temporal stochastic differential equations with case specific drift\nand diffusion terms, the models presented here require fewer modeling\nassumptions. Moreover, we show that discrete temporal Markov models can be used\nto represent dispersion in unstructured networks, which are widely used to\nmodel porous media. A new method is proposed to extend the state space of\ntemporal Markov models to improve the model predictions in the presence of\nextremely low velocities in particle trajectories and extend the applicability\nof the model to higher temporal resolutions. Finally, it is shown that by\ncombining multiple transitions, temporal models are more efficient for\ncomputing particle evolution compared to correlated CTRW with spatial\nincrements that are equal to the lengths of the links in the network.\n",
"title": "Temporal Markov Processes for Transport in Porous Media: Random Lattice Networks"
}
| null | null | null | null | true | null |
4495
| null |
Default
| null | null |
null |
{
"abstract": " In-growth or post-deposition treatment of $Cu_{2}ZnSnS_{4}$ (CZTS) absorber\nlayer had led to improved photovoltaic efficiency, however, the underlying\nphysical mechanism of such improvements are less studied. In this study, the\nthermodynamics of Na and K related defects in CZTS are investigated from first\nprinciple approach using hybrid functional, with chemical potential of Na and K\nestablished from various phases of their polysulphides. Both Na and K\npredominantly substitute on Cu sites similar to their behavior in\n$Cu(In,Ga)Se_{2}$, in contrast to previous results using the generalized\ngradient approximation (GGA). All substitutional and interstitial defects are\nshown to be either shallow levels or highly energetically unfavorable. Defect\ncomplexing between Na and abundant intrinsic defects did not show possibility\nof significant incorporation enhancement or introducing deep n-type levels. The\npossible benefit of Na incorporation on enhancing photovoltaic efficiency is\ndiscussed. The negligible defect solubility of K in CZTS also suggests possible\nsurfactant candidate.\n",
"title": "Defect Properties of Na and K in Cu2ZnSnS4 from Hybrid Functional Calculation"
}
| null | null |
[
"Physics"
] | null | true | null |
4496
| null |
Validated
| null | null |
null |
{
"abstract": " A novel approach is introduced to a very widely occurring problem, providing\na complete, explicit resolution of it: minimisation of a convex quadratic under\na general quadratic, equality or inequality, constraint. Completeness comes via\nidentification of a set of mutually exclusive and exhaustive special cases.\nExplicitness, via algebraic expressions for each solution set. Throughout,\nunderlying geometry illuminates and informs algebraic development. In\nparticular, centrally to this new approach, affine equivalence is exploited to\nre-express the same problem in simpler coordinate systems. Overall, the\nanalysis presented provides insight into the diverse forms taken both by the\nproblem itself and its solution set, showing how each may be intrinsically\nunstable. Comparisons of this global, analytic approach with the, intrinsically\ncomplementary, local, computational approach of (generalised) trust region\nmethods point to potential synergies between them. Points of contact with\nsimultaneous diagonalisation results are noted.\n",
"title": "Explicit minimisation of a convex quadratic under a general quadratic constraint: a global, analytic approach"
}
| null | null | null | null | true | null |
4497
| null |
Default
| null | null |
null |
{
"abstract": " We define a notion of morphisms between open games, exploiting a surprising\nconnection between lenses in computer science and compositional game theory.\nThis extends the more intuitively obvious definition of globular morphisms as\nmappings between strategy profiles that preserve best responses, and hence in\nparticular preserve Nash equilibria. We construct a symmetric monoidal double\ncategory in which the horizontal 1-cells are open games, vertical 1-morphisms\nare lenses, and 2-cells are morphisms of open games. States (morphisms out of\nthe monoidal unit) in the vertical category give a flexible solution concept\nthat includes both Nash and subgame perfect equilibria. Products in the\nvertical category give an external choice operator that is reminiscent of\nproducts in game semantics, and is useful in practical examples. We illustrate\nthe above two features with a simple worked example from microeconomics, the\nmarket entry game.\n",
"title": "Morphisms of open games"
}
| null | null | null | null | true | null |
4498
| null |
Default
| null | null |
null |
{
"abstract": " A semi-relativistic density-functional theory that includes spin-orbit\ncouplings and Zeeman fields on equal footing with the electromagnetic\npotentials, is an appealing framework to develop a unified first-principles\ncomputational approach for non-collinear magnetism, spintronics, orbitronics,\nand topological states. The basic variables of this theory include the\nparamagnetic current and the spin-current density, besides the particle and the\nspin density, and the corresponding exchange-correlation (xc) energy functional\nis invariant under local U(1)$\\times$SU(2) gauge transformations. The xc-energy\nfunctional must be approximated to enable practical applications, but, contrary\nto the case of the standard density functional theory, finding simple\napproximations suited to deal with realistic atomistic inhomogeneities has been\na long-standing challenge. Here, we propose a way out of this impasse by\nshowing that approximate gauge-invariant functionals can be easily generated\nfrom existing approximate functionals of ordinary density-functional theory by\napplying a simple {\\it minimal substitution} on the kinetic energy density,\nwhich controls the short-range behavior of the exchange hole. Our proposal\nopens the way to the construction of approximate, yet non-empirical\nfunctionals, which do not assume weak inhomogeneity and should therefore have a\nwide range of applicability in atomic, molecular and condensed matter physics.\n",
"title": "U(1)$\\times$SU(2) Gauge Invariance Made Simple for Density Functional Approximations"
}
| null | null | null | null | true | null |
4499
| null |
Default
| null | null |
null |
{
"abstract": " We compute the $L^2$-Betti numbers of the free $C^*$-tensor categories, which\nare the representation categories of the universal unitary quantum groups\n$A_u(F)$. We show that the $L^2$-Betti numbers of the dual of a compact quantum\ngroup $G$ are equal to the $L^2$-Betti numbers of the representation category\n$Rep(G)$ and thus, in particular, invariant under monoidal equivalence. As an\napplication, we obtain several new computations of $L^2$-Betti numbers for\ndiscrete quantum groups, including the quantum permutation groups and the free\nwreath product groups. Finally, we obtain upper bounds for the first\n$L^2$-Betti number in terms of a generating set of a $C^*$-tensor category.\n",
"title": "L^2-Betti numbers of rigid C*-tensor categories and discrete quantum groups"
}
| null | null | null | null | true | null |
4500
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.