text
null | inputs
dict | prediction
null | prediction_agent
null | annotation
list | annotation_agent
null | multi_label
bool 1
class | explanation
null | id
stringlengths 1
5
| metadata
null | status
stringclasses 2
values | event_timestamp
null | metrics
null |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
null |
{
"abstract": " In this research, we investigate the nonlinear energy transmission phenomenon\nin a reconfigurable and adaptable metastable modular metastructure. Numerical\nstudies on a 1D metastable chain uncover that when the driving frequency is\nwithin the stopband of the periodic structure, there exists a threshold input\namplitude, beyond which sudden increase in the energy transmission can be\nobserved. This onset of transmission is due to nonlinear instability and is\nknown as supratransmission. We show that due to spatial asymmetry of\nstrategically configured constituents, such transmission thresholds could shift\nconsiderably when the structure is excited from different ends and therefore\nenabling the non-reciprocal energy transmission. We discover that the critical\nthreshold amplitude can be predicted analytically using a localized\nnonlinear-linear model combining harmonic balancing and transfer matrix\nanalyses. Additionally, influences of important parameters on the change of\nthreshold amplitude are investigated to provide insight on synthesizing systems\nwith desired non-reciprocal characteristics. These investigations elucidate the\nrich and intricate dynamics achievable by nonlinearity, asymmetry, and\nmetastability, and provide new insights and opportunities to accomplish\nadaptable non-reciprocal wave energy transmission.\n",
"title": "On the wave propagation analysis and supratransmission prediction of a metastable modular metastructure for adaptive non-reciprocal energy transmission"
}
| null | null | null | null | true | null |
14801
| null |
Default
| null | null |
null |
{
"abstract": " Many modern unsupervised or semi-supervised machine learning algorithms rely\non Bayesian probabilistic models. These models are usually intractable and thus\nrequire approximate inference. Variational inference (VI) lets us approximate a\nhigh-dimensional Bayesian posterior with a simpler variational distribution by\nsolving an optimization problem. This approach has been successfully used in\nvarious models and large-scale applications. In this review, we give an\noverview of recent trends in variational inference. We first introduce standard\nmean field variational inference, then review recent advances focusing on the\nfollowing aspects: (a) scalable VI, which includes stochastic approximations,\n(b) generic VI, which extends the applicability of VI to a large class of\notherwise intractable models, such as non-conjugate models, (c) accurate VI,\nwhich includes variational models beyond the mean field approximation or with\natypical divergences, and (d) amortized VI, which implements the inference over\nlocal latent variables with inference networks. Finally, we provide a summary\nof promising future research directions.\n",
"title": "Advances in Variational Inference"
}
| null | null | null | null | true | null |
14802
| null |
Default
| null | null |
null |
{
"abstract": " In a multi-agent system, transitioning from a centralized to a distributed\ndecision-making strategy can introduce vulnerability to adversarial\nmanipulation. We study the potential for adversarial manipulation in a class of\ngraphical coordination games where the adversary can pose as a friendly agent\nin the game, thereby influencing the decision-making rules of a subset of\nagents. The adversary's influence can cascade throughout the system, indirectly\ninfluencing other agents' behavior and significantly impacting the emergent\ncollective behavior. The main results in this paper focus on characterizing\nconditions under which the adversary's local influence can dramatically impact\nthe emergent global behavior, e.g., destabilize efficient Nash equilibria.\n",
"title": "Security Against Impersonation Attacks in Distributed Systems"
}
| null | null | null | null | true | null |
14803
| null |
Default
| null | null |
null |
{
"abstract": " Large-batch training approaches have enabled researchers to utilize\nlarge-scale distributed processing and greatly accelerate deep-neural net (DNN)\ntraining. For example, by scaling the batch size from 256 to 32K, researchers\nhave been able to reduce the training time of ResNet50 on ImageNet from 29\nhours to 2.2 minutes (Ying et al., 2018). In this paper, we propose a new\napproach called linear-epoch gradual-warmup (LEGW) for better large-batch\ntraining. With LEGW, we are able to conduct large-batch training for both CNNs\nand RNNs with the Sqrt Scaling scheme. LEGW enables Sqrt Scaling scheme to be\nuseful in practice and as a result we achieve much better results than the\nLinear Scaling learning rate scheme. For LSTM applications, we are able to\nscale the batch size by a factor of 64 without losing accuracy and without\ntuning the hyper-parameters. For CNN applications, LEGW is able to achieve the\nsame accuracy even as we scale the batch size to 32K. LEGW works better than\nprevious large-batch auto-tuning techniques. LEGW achieves a 5.3X average\nspeedup over the baselines for four LSTM-based applications on the same\nhardware. We also provide some theoretical explanations for LEGW.\n",
"title": "Large-Batch Training for LSTM and Beyond"
}
| null | null | null | null | true | null |
14804
| null |
Default
| null | null |
null |
{
"abstract": " Standard deep learning systems require thousands or millions of examples to\nlearn a concept, and cannot integrate new concepts easily. By contrast, humans\nhave an incredible ability to do one-shot or few-shot learning. For instance,\nfrom just hearing a word used in a sentence, humans can infer a great deal\nabout it, by leveraging what the syntax and semantics of the surrounding words\ntells us. Here, we draw inspiration from this to highlight a simple technique\nby which deep recurrent networks can similarly exploit their prior knowledge to\nlearn a useful representation for a new word from little data. This could make\nnatural language processing systems much more flexible, by allowing them to\nlearn continually from the new words they encounter.\n",
"title": "One-shot and few-shot learning of word embeddings"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14805
| null |
Validated
| null | null |
null |
{
"abstract": " We propose a novel method called robust kernel principal component analysis\n(RKPCA) to decompose a partially corrupted matrix as a sparse matrix plus a\nhigh or full-rank matrix whose columns are drawn from a nonlinear\nlow-dimensional latent variable model. RKPCA can be applied to many problems\nsuch as noise removal and subspace clustering and is so far the only\nunsupervised nonlinear method robust to sparse noises. We also provide\ntheoretical guarantees for RKPCA. The optimization of RKPCA is challenging\nbecause it involves nonconvex and indifferentiable problems simultaneously. We\npropose two nonconvex optimization algorithms for RKPCA: alternating direction\nmethod of multipliers with backtracking line search and proximal linearized\nminimization with adaptive step size. Comparative studies on synthetic data and\nnature images corroborate the effectiveness and superiority of RKPCA in noise\nremoval and robust subspace clustering.\n",
"title": "Exactly Robust Kernel Principal Component Analysis"
}
| null | null | null | null | true | null |
14806
| null |
Default
| null | null |
null |
{
"abstract": " We analyse families of codes for classical data transmission over quantum\nchannels that have both a vanishing probability of error and a code rate\napproaching capacity as the code length increases. To characterise the\nfundamental tradeoff between decoding error, code rate and code length for such\ncodes we introduce a quantum generalisation of the moderate deviation analysis\nproposed by Altug and Wagner as well as Polyanskiy and Verdu. We derive such a\ntradeoff for classical-quantum (as well as image-additive) channels in terms of\nthe channel capacity and the channel dispersion, giving further evidence that\nthe latter quantity characterises the necessary backoff from capacity when\ntransmitting finite blocks of classical data. To derive these results we also\nstudy asymmetric binary quantum hypothesis testing in the moderate deviations\nregime. Due to the central importance of the latter task, we expect that our\ntechniques will find further applications in the analysis of other quantum\ninformation processing tasks.\n",
"title": "Moderate deviation analysis for classical communication over quantum channels"
}
| null | null | null | null | true | null |
14807
| null |
Default
| null | null |
null |
{
"abstract": " We study the training process of Deep Neural Networks (DNNs) from the Fourier\nanalysis perspective. Our starting point is a Frequency Principle (F-Principle)\n--- DNNs initialized with small parameters often fit target functions from low\nto high frequencies --- which was first proposed by Xu et al. (2018) and\nRahaman et al. (2018) on synthetic datasets. In this work, we first show the\nuniversality of the F-Principle by demonstrating this phenomenon on\nhigh-dimensional benchmark datasets, such as MNIST and CIFAR10. Then, based on\nexperiments, we show that the F-Principle provides insight into both the\nsuccess and failure of DNNs in different types of problems. Based on the\nF-Principle, we further propose that DNN can be adopted to accelerate the\nconvergence of low frequencies for scientific computing problems, in which most\nof the conventional methods (e.g., Jacobi method) exhibit the opposite\nconvergence behavior --- faster convergence for higher frequencies. Finally, we\nprove a theorem for DNNs of one hidden layer as a first step towards a\nmathematical explanation of the F-Principle. Our work indicates that the\nF-Principle with Fourier analysis is a promising approach to the study of DNNs\nbecause it seems ubiquitous, applicable, and explainable.\n",
"title": "Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks"
}
| null | null | null | null | true | null |
14808
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we explore the theoretical boundaries of planning in a setting\nwhere no model of the agent's actions is given. Instead of an action model, a\nset of successfully executed plans are given and the task is to generate a plan\nthat is safe, i.e., guaranteed to achieve the goal without failing. To this\nend, we show how to learn a conservative model of the world in which actions\nare guaranteed to be applicable. This conservative model is then given to an\noff-the-shelf classical planner, resulting in a plan that is guaranteed to\nachieve the goal. However, this reduction from a model-free planning to a\nmodel-based planning is not complete: in some cases a plan will not be found\neven when such exists. We analyze the relation between the number of observed\nplans and the likelihood that our conservative approach will indeed fail to\nsolve a solvable problem. Our analysis show that the number of trajectories\nneeded scales gracefully.\n",
"title": "Efficient, Safe, and Probably Approximately Complete Learning of Action Models"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14809
| null |
Validated
| null | null |
null |
{
"abstract": " Let $f\\colon M \\to M$ be a uniformly quasiregular self-mapping of a compact,\nconnected, and oriented Riemannian $n$-manifold $M$ without boundary, $n\\ge 2$.\nWe show that, for $k \\in \\{0,\\ldots, n\\}$, the induced homomorphism $f^* \\colon\nH^k(M;\\mathbb{R}) \\to H^k(M;\\mathbb{R})$, where $H^k(M;\\mathbb{R})$ is the\n$k$:th singular cohomology of $M$, is complex diagonalizable and the\neigenvalues of $f^*$ have modulus $(\\mathrm{deg}\\ f)^{k/n}$. As an application,\nwe obtain a degree restriction for uniformly quasiregular self-mappings of\nclosed manifolds. In the proof of the main theorem, we use a Sobolev--de Rham\ncohomology based on conformally invariant differential forms and an induced\npush-forward operator.\n",
"title": "Uniform cohomological expansion of uniformly quasiregular mappings"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14810
| null |
Validated
| null | null |
null |
{
"abstract": " Our method of density elimination is generalized to the non-commutative\nsubstructural logic GpsUL*. Then the standard completeness of GpsUL* follows as\na lemma by virtue of previous work by Metcalfe and Montagna. This result shows\nthat GpsUL* is the logic of pseudo-uninorms and their residua and answered the\nquestion posed by Prof. Metcalfe, Olivetti, Gabbay and Tsinakis.\n",
"title": "The logic of pseudo-uninorms and their residua"
}
| null | null | null | null | true | null |
14811
| null |
Default
| null | null |
null |
{
"abstract": " In this article we use linear algebra to improve the computational time for\nthe obtaining of Green's functions of linear differential equations with\nreflection (DER). This is achieved by decomposing both the `reduced' equation\n(the ODE associated to a given DER) and the corresponding two-point boundary\nconditions.\n",
"title": "Computation of Green's functions through algebraic decomposition of operators"
}
| null | null | null | null | true | null |
14812
| null |
Default
| null | null |
null |
{
"abstract": " The concept of leader--follower (or Stackelberg) equilibrium plays a central\nrole in a number of real--world applications of game theory. While the case\nwith a single follower has been thoroughly investigated, results with multiple\nfollowers are only sporadic and the problem of designing and evaluating\ncomputationally tractable equilibrium-finding algorithms is still largely open.\nIn this work, we focus on the fundamental case where multiple followers play a\nNash equilibrium once the leader has committed to a strategy---as we\nillustrate, the corresponding equilibrium finding problem can be easily shown\nto be $\\mathcal{FNP}$--hard and not in Poly--$\\mathcal{APX}$ unless\n$\\mathcal{P} = \\mathcal{NP}$ and therefore it is one among the hardest problems\nto solve and approximate. We propose nonconvex mathematical programming\nformulations and global optimization methods to find both exact and approximate\nequilibria, as well as a heuristic black box algorithm. All the methods and\nformulations that we introduce are thoroughly evaluated computationally.\n",
"title": "Methods for finding leader--follower equilibria with multiple followers"
}
| null | null | null | null | true | null |
14813
| null |
Default
| null | null |
null |
{
"abstract": " It is widely perceived that the correlation effect may play an important role\nin several unconventional superconducting families, such as cuprate, iron-based\nand heavy-fermion superconductors. The application of high pressure can tune\nthe ground state properties and balance the localization and itineracy of\nelectrons in correlated systems, which may trigger unconventional\nsuperconductivity. Moreover, non-centrosymmetric structure may induce the spin\ntriplet pairing which is very rare in nature. Here, we report a new compound\nScZrCo1-${\\delta}$ crystallizing in the Ti2Ni structure with the space group of\nFD3-MS without a spatial inversion center. The resistivity of the material at\nambient pressure shows a bad metal and weak semiconducting behavior.\nFurthermore, specific heat and magnetic susceptibility measurements yield a\nrather large value of Wilson ratio ~4.47. Both suggest a ground state with\ncorrelation effect. By applying pressure, the up-going behavior of resistivity\nin lowering temperature at ambient pressure is suppressed and gradually it\nbecomes metallic. At a pressure of about 19.5 GPa superconductivity emerges. Up\nto 36.05 GPa, a superconducting transition at about 3.6 K with a quite high\nupper critical field is observed. Our discovery here provides a new platform\nfor investigating the relationship between correlation effect and\nsuperconductivity.\n",
"title": "Pressure Induced Superconductivity in the New Compound ScZrCo1-$δ$"
}
| null | null |
[
"Physics"
] | null | true | null |
14814
| null |
Validated
| null | null |
null |
{
"abstract": " While deep learning is remarkably successful on perceptual tasks, it was also\nshown to be vulnerable to adversarial perturbations of the input. These\nperturbations denote noise added to the input that was generated specifically\nto fool the system while being quasi-imperceptible for humans. More severely,\nthere even exist universal perturbations that are input-agnostic but fool the\nnetwork on the majority of inputs. While recent work has focused on image\nclassification, this work proposes attacks against semantic image segmentation:\nwe present an approach for generating (universal) adversarial perturbations\nthat make the network yield a desired target segmentation as output. We show\nempirically that there exist barely perceptible universal noise patterns which\nresult in nearly the same predicted segmentation for arbitrary inputs.\nFurthermore, we also show the existence of universal noise which removes a\ntarget class (e.g., all pedestrians) from the segmentation while leaving the\nsegmentation mostly unchanged otherwise.\n",
"title": "Universal Adversarial Perturbations Against Semantic Image Segmentation"
}
| null | null | null | null | true | null |
14815
| null |
Default
| null | null |
null |
{
"abstract": " Bayesian inference for models that have an intractable partition function is\nknown as a doubly intractable problem, where standard Monte Carlo methods are\nnot applicable. The past decade has seen the development of auxiliary variable\nMonte Carlo techniques (M{\\o}ller et al., 2006; Murray et al., 2006) for\ntackling this problem; these approaches being members of the more general class\nof pseudo-marginal, or exact-approximate, Monte Carlo algorithms (Andrieu and\nRoberts, 2009), which make use of unbiased estimates of intractable posteriors.\nEveritt et al. (2017) investigated the use of exact-approximate importance\nsampling (IS) and sequential Monte Carlo (SMC) in doubly intractable problems,\nbut focussed only on SMC algorithms that used data-point tempering. This paper\ndescribes SMC samplers that may use alternative sequences of distributions, and\ndescribes ways in which likelihood estimates may be improved adaptively as the\nalgorithm progresses, building on ideas from Moores et al. (2015). This\napproach is compared with a number of alternative algorithms for doubly\nintractable problems, including approximate Bayesian computation (ABC), which\nwe show is closely related to the method of M{\\o}ller et al. (2006).\n",
"title": "Marginal sequential Monte Carlo for doubly intractable models"
}
| null | null | null | null | true | null |
14816
| null |
Default
| null | null |
null |
{
"abstract": " The impact of neutral impurity scattering of electrons on the charge drift\nmobility in high purity n-type germanium crystals at 77 Kelvin is investigated.\nWe calculated the contributions from ionized impurity scattering, lattice\nscattering, and neutral impurity scattering to the total charge drift mobility\nusing theoretical models. The experimental data such as charge carrier\nconcentration, mobility and resistivity are measured by Hall Effect system at\n77 Kelvin. The neutral impurity concentration is derived from the Matthiessen's\nrule using the measured Hall mobility and ionized impurity concentration. The\nradial distribution of the neutral impurity concentration in the self-grown\ncrystals is determined. Consequently, we demonstrated that neutral impurity\nscattering is a significant contribution to the charge drift mobility, which\nhas a dependence on the concentration of neutral impurities in high purity\nn-type germanium crystal.\n",
"title": "The impact of neutral impurity concentration on charge drift mobility in n-type germanium"
}
| null | null |
[
"Physics"
] | null | true | null |
14817
| null |
Validated
| null | null |
null |
{
"abstract": " Batyrev constructed a family of Calabi-Yau hypersurfaces dual to the first\nChern class in toric Fano varieties. Using this construction, we introduce a\nfamily of Calabi-Yau manifolds whose SU-bordism classes generate the special\nunitary bordism ring\n$\\varOmega^{SU}\\otimes\\mathbb{Z}[\\frac{1}{2}]\\cong\\mathbb{Z}[\\frac{1}{2}][y_{i}\\colon\ni\\ge 2]$. We also describe explicit Calabi-Yau representatives for\nmultiplicative generators of the SU-bordism ring in low dimensions.\n",
"title": "Calabi-Yau hypersurfaces and SU-bordism"
}
| null | null | null | null | true | null |
14818
| null |
Default
| null | null |
null |
{
"abstract": " Transfer Learning (TL) aims to transfer knowledge acquired in one problem,\nthe source problem, onto another problem, the target problem, dispensing with\nthe bottom-up construction of the target model. Due to its relevance, TL has\ngained significant interest in the Machine Learning community since it paves\nthe way to devise intelligent learning models that can easily be tailored to\nmany different applications. As it is natural in a fast evolving area, a wide\nvariety of TL methods, settings and nomenclature have been proposed so far.\nHowever, a wide range of works have been reporting different names for the same\nconcepts. This concept and terminology mixture contribute however to obscure\nthe TL field, hindering its proper consideration. In this paper we present a\nreview of the literature on the majority of classification TL methods, and also\na distribution-based categorization of TL with a common nomenclature suitable\nto classification problems. Under this perspective three main TL categories are\npresented, discussed and illustrated with examples.\n",
"title": "Distribution-Based Categorization of Classifier Transfer Learning"
}
| null | null | null | null | true | null |
14819
| null |
Default
| null | null |
null |
{
"abstract": " We show that every periodic virtual knot can be realized as the closure of a\nperiodic virtual braid and use this to study the Alexander invariants of\nperiodic virtual knots. If $K$ is a $q$-periodic and almost classical knot, we\nshow that its quotient knot $K_*$ is also almost classical, and in the case\n$q=p^r$ is a prime power, we establish an analogue of Murasugi's congruence\nrelating the Alexander polynomials of $K$ and $K_*$ over the integers modulo\n$p$. This result is applied to the problem of determining the possible periods\nof a virtual knot $K$. One consequence is that if $K$ is an almost classical\nknot with a nontrivial Alexander polynomial, then it is $p$-periodic for only\nfinitely many primes $p$. Combined with parity and Manturov projection, our\nmethods provide conditions that a general virtual knot must satisfy in order to\nbe $q$-periodic.\n",
"title": "Alexander invariants of periodic virtual knots"
}
| null | null | null | null | true | null |
14820
| null |
Default
| null | null |
null |
{
"abstract": " Computing optimal transport distances such as the earth mover's distance is a\nfundamental problem in machine learning, statistics, and computer vision.\nDespite the recent introduction of several algorithms with good empirical\nperformance, it is unknown whether general optimal transport distances can be\napproximated in near-linear time. This paper demonstrates that this ambitious\ngoal is in fact achieved by Cuturi's Sinkhorn Distances. This result relies on\na new analysis of Sinkhorn iteration, which also directly suggests a new greedy\ncoordinate descent algorithm, Greenkhorn, with the same theoretical guarantees.\nNumerical simulations illustrate that Greenkhorn significantly outperforms the\nclassical Sinkhorn algorithm in practice.\n",
"title": "Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration"
}
| null | null | null | null | true | null |
14821
| null |
Default
| null | null |
null |
{
"abstract": " Women have become better represented in business, academia, and government\nover time, yet a dearth of women at the highest levels of leadership remains.\nSociologists have attributed the leaky progression of women through\nprofessional hierarchies to various cultural and psychological factors, such as\nself-segregation and bias. Here, we present a minimal mathematical model that\nreveals the relative role that bias and homophily (self-seeking) may play in\nthe ascension of women through professional hierarchies. Unlike previous\nmodels, our novel model predicts that gender parity is not inevitable, and\ndeliberate intervention may be required to achieve gender balance in several\nfields. To validate the model, we analyze a new database of gender\nfractionation over time for 16 professional hierarchies. We quantify the degree\nof homophily and bias in each professional hierarchy, and we propose specific\ninterventions to achieve gender parity more quickly.\n",
"title": "Mathematical model of gender bias and homophily in professional hierarchies"
}
| null | null | null | null | true | null |
14822
| null |
Default
| null | null |
null |
{
"abstract": " We establish the monotonicity property for the mass of non-pluripolar\nproducts on compact Kahler manifolds, and we initiate the study of complex\nMonge-Ampere type equations with prescribed singularity type. Using the\nvariational method of Berman-Boucksom-Guedj-Zeriahi we prove existence and\nuniqueness of solutions with small unbounded locus. We give applications to\nKahler-Einstein metrics with prescribed singularity, and we show that the\nlog-concavity property holds for non-pluripolar products with small unbounded\nlocus.\n",
"title": "Monotonicity of non-pluripolar products and complex Monge-Ampère equations with prescribed singularity"
}
| null | null | null | null | true | null |
14823
| null |
Default
| null | null |
null |
{
"abstract": " Creating tetrahedral meshes with anatomically accurate surfaces is critically\nimportant for a wide range of model-based neuroimaging modalities. However,\ncomputationally efficient brain meshing algorithms and software are greatly\nlacking. Here, we report a fully automated open-source software to rapidly\ncreate high-quality tetrahedral meshes from brain segmentations. Built upon\nvarious open-source meshing utilities, the proposed meshing workflow allows\nrobust generation of complex head and brain mesh models from multi-label\nvolumes, tissue probability maps, surface meshes and their combinations. The\nquality of the complex tissue boundaries is preserved through a surface-based\napproach, allowing fine-grained control over the sizes and quality of the mesh\nelements through explicit user-defined meshing criteria. The proposed meshing\npipeline is highly versatile and compatible with many commonly used brain\nanalysis tools, including SPM, FSL, FreeSurfer, and BrainSuite. With this\nmesh-generation pipeline, we demonstrate that one can generate 3D full-head\nmeshes that combine scalp, skull, cerebrospinal fluid, gray matter, white\nmatter, and air cavities with a typical processing time of less than 40\nseconds. This approach can also incorporate highly detailed cortical and white\nmatter surface meshes derived from FSL and FreeSurfer with tissue segmentation\ndata. Finally, a high-quality brain atlas mesh library for different age\ngroups, ranging from infants to elderlies, was built to demonstrate the\nrobustness of the proposed workflow, as well as to serve as a common platform\nfor simulation-based brain studies. Our open-source meshing software\n\"brain2mesh\" and the human brain atlas mesh library can be downloaded at\nthis http URL.\n",
"title": "Fast and high-quality tetrahedral mesh generation from neuroanatomical scans"
}
| null | null | null | null | true | null |
14824
| null |
Default
| null | null |
null |
{
"abstract": " Dropout is a very effective way of regularizing neural networks.\nStochastically \"dropping out\" units with a certain probability discourages\nover-specific co-adaptations of feature detectors, preventing overfitting and\nimproving network generalization. Besides, Dropout can be interpreted as an\napproximate model aggregation technique, where an exponential number of smaller\nnetworks are averaged in order to get a more powerful ensemble. In this paper,\nwe show that using a fixed dropout probability during training is a suboptimal\nchoice. We thus propose a time scheduling for the probability of retaining\nneurons in the network. This induces an adaptive regularization scheme that\nsmoothly increases the difficulty of the optimization problem. This idea of\n\"starting easy\" and adaptively increasing the difficulty of the learning\nproblem has its roots in curriculum learning and allows one to train better\nmodels. Indeed, we prove that our optimization strategy implements a very\ngeneral curriculum scheme, by gradually adding noise to both the input and\nintermediate feature representations within the network architecture.\nExperiments on seven image classification datasets and different network\narchitectures show that our method, named Curriculum Dropout, frequently yields\nto better generalization and, at worst, performs just as well as the standard\nDropout method.\n",
"title": "Curriculum Dropout"
}
| null | null | null | null | true | null |
14825
| null |
Default
| null | null |
null |
{
"abstract": " This paper is a continuation of arXiv:1405.1707. We present certain new\napplications and generalizations of the free field realization of the twisted\nHeisenberg-Virasoro algebra ${\\mathcal H}$ at level zero.\nWe find explicit formulas for singular vectors in certain Verma modules. A\nfree field realization of self-dual modules for ${\\mathcal H}$ is presented by\ncombining a bosonic construction of Whittaker modules from arXiv:1409.5354 with\na construction of logarithmic modules for vertex algebras. As an application,\nwe prove that there exists a non-split self-extension of irreducible self-dual\nmodule which is a logarithmic module of rank two.\nWe construct a large family of logarithmic modules containing different types\nof highest weight modules as subquotients. We believe that these logarithmic\nmodules are related with projective covers of irreducible modules in a suitable\ncategory of ${\\mathcal H}$-modules.\n",
"title": "Self-dual and logarithmic representations of the twisted Heisenberg--Virasoro algebra at level zero"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14826
| null |
Validated
| null | null |
null |
{
"abstract": " We illustrate the advantages of distance weighted discrimination for\nclassification and feature extraction in a High Dimension Low Sample Size\n(HDLSS) situation. The HDLSS context is a gender classification problem of face\nimages in which the dimension of the data is several orders of magnitude larger\nthan the sample size. We compare distance weighted discrimination with Fisher's\nlinear discriminant, support vector machines, and principal component analysis\nby exploring their classification interpretation through insightful\nvisuanimations and by examining the classifiers' discriminant errors. This\nanalysis enables us to make new contributions to the understanding of the\ndrivers of human discrimination between males and females.\n",
"title": "Distance weighted discrimination of face images for gender classification"
}
| null | null | null | null | true | null |
14827
| null |
Default
| null | null |
null |
{
"abstract": " Accurate state estimation of large-scale lithium-ion battery packs is\nnecessary for the advanced control of batteries, which could potentially\nincrease their lifetime through e.g. reconfiguration. To tackle this problem,\nan enhanced reduced-order electrochemical model is used here. This model allows\nconsidering a wider operating range and thermal coupling between cells, the\nlatter turning out to be significant. The resulting nonlinear model is\nexploited for state estimation through unscented Kalman filters (UKF). A sensor\nnetwork composed of one sensor node per battery cell is deployed. Each sensor\nnode is equipped with a local UKF, which uses available local measurements\ntogether with additional information coming from neighboring sensor nodes. Such\nstate estimation scheme gives rise to a partition-based unscented Kalman filter\n(PUKF). The method is validated on data from a detailed simulator for a battery\npack comprised of six cells, with reconfiguration capabilities. The results\nshow that the distributed approach outperforms the centralized one in terms of\ncomputation time at the expense of a very low increase of mean-square\nestimation error.\n",
"title": "Partition-based Unscented Kalman Filter for Reconfigurable Battery Pack State Estimation using an Electrochemical Model"
}
| null | null | null | null | true | null |
14828
| null |
Default
| null | null |
null |
{
"abstract": " The goal of this paper is not to introduce a single algorithm or method, but\nto make theoretical steps towards fully understanding the training dynamics of\ngenerative adversarial networks. In order to substantiate our theoretical\nanalysis, we perform targeted experiments to verify our assumptions, illustrate\nour claims, and quantify the phenomena. This paper is divided into three\nsections. The first section introduces the problem at hand. The second section\nis dedicated to studying and proving rigorously the problems including\ninstability and saturation that arize when training generative adversarial\nnetworks. The third section examines a practical and theoretically grounded\ndirection towards solving these problems, while introducing new tools to study\nthem.\n",
"title": "Towards Principled Methods for Training Generative Adversarial Networks"
}
| null | null | null | null | true | null |
14829
| null |
Default
| null | null |
null |
{
"abstract": " This paper presents a new multitask learning framework that learns a shared\nrepresentation among the tasks, incorporating both task and feature clusters.\nThe jointly-induced clusters yield a shared latent subspace where task\nrelationships are learned more effectively and more generally than in\nstate-of-the-art multitask learning methods. The proposed general framework\nenables the derivation of more specific or restricted state-of-the-art\nmultitask methods. The paper also proposes a highly-scalable multitask learning\nalgorithm, based on the new framework, using conjugate gradient descent and\ngeneralized \\textit{Sylvester equations}. Experimental results on synthetic and\nbenchmark datasets show that the proposed method systematically outperforms\nseveral state-of-the-art multitask learning methods.\n",
"title": "Co-Clustering for Multitask Learning"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14830
| null |
Validated
| null | null |
null |
{
"abstract": " The fast iterative soft thresholding algorithm (FISTA) is used to solve\nconvex regularized optimization problems in machine learning. Distributed\nimplementations of the algorithm have become popular since they enable the\nanalysis of large datasets. However, existing formulations of FISTA communicate\ndata at every iteration which reduces its performance on modern distributed\narchitectures. The communication costs of FISTA, including bandwidth and\nlatency costs, is closely tied to the mathematical formulation of the\nalgorithm. This work reformulates FISTA to communicate data at every k\niterations and reduce data communication when operating on large data sets. We\nformulate the algorithm for two different optimization methods on the Lasso\nproblem and show that the latency cost is reduced by a factor of k while\nbandwidth and floating-point operation costs remain the same. The convergence\nrates and stability properties of the reformulated algorithms are similar to\nthe standard formulations. The performance of communication-avoiding FISTA and\nProximal Newton methods is evaluated on 1 to 1024 nodes for multiple benchmarks\nand demonstrate average speedups of 3-10x with scaling properties that\noutperform the classical algorithms.\n",
"title": "Avoiding Communication in Proximal Methods for Convex Optimization Problems"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14831
| null |
Validated
| null | null |
null |
{
"abstract": " We introduce a regression model for data on non-linear manifolds. The model\ndescribes the relation between a set of manifold valued observations, such as\nshapes of anatomical objects, and Euclidean explanatory variables. The approach\nis based on stochastic development of Euclidean diffusion processes to the\nmanifold. Defining the data distribution as the transition distribution of the\nmapped stochastic process, parameters of the model, the non-linear analogue of\ndesign matrix and intercept, are found via maximum likelihood. The model is\nintrinsically related to the geometry encoded in the connection of the\nmanifold. We propose an estimation procedure which applies the Laplace\napproximation of the likelihood function. A simulation study of the performance\nof the model is performed and the model is applied to a real dataset of Corpus\nCallosum shapes.\n",
"title": "Stochastic Development Regression on Non-Linear Manifolds"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14832
| null |
Validated
| null | null |
null |
{
"abstract": " A Convolutional Neural Network was used to predict kidney function in\npatients with chronic kidney disease from high-resolution digital pathology\nscans of their kidney biopsies. Kidney biopsies were taken from participants of\nthe NEPTUNE study, a longitudinal cohort study whose goal is to set up\ninfrastructure for observing the evolution of 3 forms of idiopathic nephrotic\nsyndrome, including developing predictors for progression of kidney disease.\nThe knowledge of future kidney function is desirable as it can identify\nhigh-risk patients and influence treatment decisions, reducing the likelihood\nof irreversible kidney decline.\n",
"title": "Prediction of Kidney Function from Biopsy Images Using Convolutional Neural Networks"
}
| null | null | null | null | true | null |
14833
| null |
Default
| null | null |
null |
{
"abstract": " We briefly review the recent results of constraining neutrino mass in\ndynamical dark energy models using cosmological observations and summarize the\nfindings. (i) In dynamical dark energy models, compared to $\\Lambda$CDM, the\nupper limit of $\\sum m_\\nu$ can become larger and can also become smaller. In\nthe cases of phantom and early phantom (i.e., the quintom evolving from $w<-1$\nto $w>-1$), the constraint on $\\sum m_\\nu$ becomes looser; but in the cases of\nquintessence and early quintessence (i.e., the quintom evolving from $w>-1$ to\n$w<-1$), the constraint on $\\sum m_\\nu$ becomes tighter. (ii) In the\nholographic dark energy (HDE) model, the tightest constraint on $\\sum m_\\nu$,\ni.e., $\\sum m_\\nu<0.105$ eV, is obtained, which is almost equal to the lower\nlimit of $\\sum m_\\nu$ of IH case. Thus, it is of great interest to find that\nthe future neutrino oscillation experiments would potentially offer a possible\nfalsifying scheme for the HDE model. (iii) The mass splitting of neutrinos can\ninfluence the cosmological fits. We find that the NH case fits the current\nobservations slightly better than the IH case, although the difference of\n$\\chi^2$ of the two cases is still not significant enough to definitely\ndistinguish the neutrino mass hierarchy.\n",
"title": "Weighing neutrinos in dynamical dark energy models"
}
| null | null | null | null | true | null |
14834
| null |
Default
| null | null |
null |
{
"abstract": " Supersymmetry plays an important role in superstring theory and particle\nphysics, but has never been observed in experiments. At certain quantum\ncritical points of condensed matter systems, the fermionic excitations are\ngapless due to the special electronic structure whereas the bosonic order\nparameter is automatically gapless, offering a promising platform to realize\nemergent supersymmetry by tuning a single parameter. Here, we study under what\ncircumstances can supersymmetry emerge in a quantum critical system. We\ndemonstrate that the Yukawa-type coupling between the gapless fermion and boson\nmay induce a number of highly nonlocal self-interacting terms in the effective\nfield theory of the boson. Only when such terms do not exist or are irrelevant,\ncould supersymmetry have the chance to be dynamically generated at low\nenergies. This strong constraint provides an important guidance for the\nexploration of emergent supersymmetry in various condensed matter systems, and\nalso should be carefully considered in the study of quantum critical behaviors.\n",
"title": "Can supersymmetry emerge at a quantum critical point?"
}
| null | null | null | null | true | null |
14835
| null |
Default
| null | null |
null |
{
"abstract": " Using the Tridiagonal Representation Approach, we obtain solutions (energy\nspectrum and corresponding wavefunctions) for a new five-parameter potential\nbox with inverse square singularity at the boundaries.\n",
"title": "Five-parameter potential box with inverse square singular boundaries"
}
| null | null |
[
"Physics"
] | null | true | null |
14836
| null |
Validated
| null | null |
null |
{
"abstract": " In this paper we systematically explore questions of succinctness in modal\nlogics employed in spatial reasoning. We show that the closure operator,\ndespite being less expressive, is exponentially more succinct than the\nlimit-point operator, and that the $\\mu$-calculus is exponentially more\nsuccinct than the equally-expressive tangled limit operator. These results hold\nfor any class of spaces containing at least one crowded metric space or\ncontaining all spaces based on ordinals below $\\omega^\\omega$, with the usual\nlimit operator. We also show that these results continue to hold even if we\nenrich the less succinct language with the universal modality.\n",
"title": "Succinctness in subsystems of the spatial mu-calculus"
}
| null | null | null | null | true | null |
14837
| null |
Default
| null | null |
null |
{
"abstract": " By applying measurements of the dielectric constants and relative length\nchanges to the dimerized molecular conductor\n$\\kappa$-(BEDT-TTF)$_2$Hg(SCN)$_2$Cl, we provide evidence for order-disorder\ntype electronic ferroelectricity which is driven by charge order within the\n(BEDT-TTF)$_2$ dimers and stabilized by a coupling to the anions. According to\nour density functional theory calculations, this material is characterized by a\nmoderate strength of dimerization. This system thus bridges the gap between\nstrongly dimerized materials, often approximated as dimer-Mott systems at 1/2\nfilling, and non- or weakly dimerized systems at 1/4 filling exhibiting charge\norder. Our results indicate that intra-dimer charge degrees of freedom are of\nparticular importance in correlated $\\kappa$-(BEDT-TTF)$_2$X salts and can\ncreate novel states, such as electronically-driven multiferroicity or\ncharge-order-induced quasi-1D spin liquids.\n",
"title": "Evidence for electronically-driven ferroelectricity in the family of strongly correlated dimerized BEDT-TTF molecular conductors"
}
| null | null | null | null | true | null |
14838
| null |
Default
| null | null |
null |
{
"abstract": " Samples of two characteristic semiconductor sensor materials, silicon and\ngermanium, have been irradiated with neutrons produced at the RP-10 Nuclear\nResearch Reactor at 4.5 MW. Their radionuclides photon spectra have been\nmeasured with high resolution gamma spectroscopy, quantifying four\nradioisotopes ($^{28}$Al, $^{29}$Al for Si and $^{75}$Ge and $^{77}$Ge for Ge).\nWe have compared the radionuclides production and their emission spectrum data\nwith Monte Carlo simulation results from FLUKA. Thus we have tested FLUKA's low\nenergy neutron library (ENDF/B-VIIR) and decay photon scoring with respect to\nthe activation of these semiconductors. We conclude that FLUKA is capable of\npredicting relative photon peak amplitudes, with gamma intensities greater than\n1%, of produced radionuclides with an average uncertainty of 13%. This work\nallows us to estimate the corresponding systematic error on neutron activation\nsimulation studies of these sensor materials.\n",
"title": "Testing FLUKA on neutron activation of Si and Ge at nuclear research reactor using gamma spectroscopy"
}
| null | null | null | null | true | null |
14839
| null |
Default
| null | null |
null |
{
"abstract": " Databases are widespread, yet extracting relevant data can be difficult.\nWithout substantial domain knowledge, multivariate search queries often return\nsparse or uninformative results. This paper introduces an approach for\nsearching structured data based on probabilistic programming and nonparametric\nBayes. Users specify queries in a probabilistic language that combines standard\nSQL database search operators with an information theoretic ranking function\ncalled predictive relevance. Predictive relevance can be calculated by a fast\nsparse matrix algorithm based on posterior samples from CrossCat, a\nnonparametric Bayesian model for high-dimensional, heterogeneously-typed data\ntables. The result is a flexible search technique that applies to a broad class\nof information retrieval problems, which we integrate into BayesDB, a\nprobabilistic programming platform for probabilistic data analysis. This paper\ndemonstrates applications to databases of US colleges, global macroeconomic\nindicators of public health, and classic cars. We found that human evaluators\noften prefer the results from probabilistic search to results from a standard\nbaseline.\n",
"title": "Probabilistic Search for Structured Data via Probabilistic Programming and Nonparametric Bayes"
}
| null | null | null | null | true | null |
14840
| null |
Default
| null | null |
null |
{
"abstract": " The Dominative $p$-Laplace Operator is introduced. This operator is a\nrelative to the $p$-Laplacian, but with the distinguishing property of being\nsublinear. It explains the superposition principle in the $p$-Laplace Equation.\n",
"title": "Superposition of p-superharmonic functions"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14841
| null |
Validated
| null | null |
null |
{
"abstract": " We show that every free amalgamation class of finite structures with\nrelations and (symmetric) partial functions is a Ramsey class when enriched by\na free linear ordering of vertices. This is a common strengthening of the\nNešetřil-Rödl Theorem and the second and third authors' Ramsey\ntheorem for finite models (that is, structures with both relations and\nfunctions). We also find subclasses with the ordering property. For languages\nwith relational symbols and unary functions we also show the extension property\nfor partial automorphisms (EPPA) of free amalgamation classes. These general\nresults solve several conjectures and provide an easy Ramseyness test for many\nclasses of structures.\n",
"title": "Ramsey properties and extending partial automorphisms for classes of finite structures"
}
| null | null |
[
"Computer Science",
"Mathematics"
] | null | true | null |
14842
| null |
Validated
| null | null |
null |
{
"abstract": " Elliptically contoured distributions generalize the multivariate normal\ndistributions in such a way that the density generators need not be\nexponential. However, as the name suggests, elliptically contoured\ndistributions remain to be restricted in that the similar density contours\nought to be elliptical. Kamiya, Takemura and Kuriki [Star-shaped distributions\nand their generalizations, Journal of Statistical Planning and Inference 138\n(2008), 3429--3447] proposed star-shaped distributions, for which the density\ncontours are allowed to be boundaries of arbitrary similar star-shaped sets. In\nthe present paper, we propose a nonparametric estimator of the shape of the\ndensity contours of star-shaped distributions, and prove its strong consistency\nwith respect to the Hausdorff distance. We illustrate our estimator by\nsimulation.\n",
"title": "Estimation of the shape of the density contours of star-shaped distributions"
}
| null | null | null | null | true | null |
14843
| null |
Default
| null | null |
null |
{
"abstract": " The classification of shapes is of great interest in diverse areas ranging\nfrom medical imaging to computer vision and beyond. While many statistical\nframeworks have been developed for the classification problem, most are\nstrongly tied to early formulations of the problem - with an object to be\nclassified described as a vector in a relatively low-dimensional Euclidean\nspace. Statistical shape data have two main properties that suggest a need for\na novel approach: (i) shapes are inherently infinite dimensional with strong\ndependence among the positions of nearby points, and (ii) shape space is not\nEuclidean, but is fundamentally curved. To accommodate these features of the\ndata, we work with the square-root velocity function of the curves to provide a\nuseful formal description of the shape, pass to tangent spaces of the manifold\nof shapes at different projection points which effectively separate shapes for\npairwise classification in the training data, and use principal components\nwithin these tangent spaces to reduce dimensionality. We illustrate the impact\nof the projection point and choice of subspace on the misclassification rate\nwith a novel method of combining pairwise classifiers.\n",
"title": "Aggregated Pairwise Classification of Statistical Shapes"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14844
| null |
Validated
| null | null |
null |
{
"abstract": " We present a clustering analysis of a sample of 238 Ly{$\\alpha$}-emitters at\nredshift 3<z<6 from the MUSE-Wide survey. This survey mosaics extragalactic\nlegacy fields with 1h MUSE pointings to detect statistically relevant samples\nof emission line galaxies. We analysed the first year observations from\nMUSE-Wide making use of the clustering signal in the line-of-sight direction.\nThis method relies on comparing pair-counts at close redshifts for a fixed\ntransverse distance and thus exploits the full potential of the redshift range\ncovered by our sample. A clear clustering signal with a correlation length of\nr0 = 2.9(+1.0/-1.1) Mpc (comoving) is detected. Whilst this result is based on\nonly about a quarter of the full survey size, it already shows the immense\npotential of MUSE for efficiently observing and studying the clustering of\nLy{$\\alpha$}-emitters.\n",
"title": "The MUSE-Wide survey: Detection of a clustering signal from Lyman-α-emitters at 3<z<6"
}
| null | null | null | null | true | null |
14845
| null |
Default
| null | null |
null |
{
"abstract": " This work exploits the logical foundation of session types to determine what\nkind of type discipline for the pi-calculus can exactly capture, and is\ncaptured by, lambda-calculus behaviours. Leveraging the proof theoretic content\nof the soundness and completeness of sequent calculus and natural deduction\npresentations of linear logic, we develop the first mutually inverse and fully\nabstract processes-as-functions and functions-as-processes encodings between a\npolymorphic session pi-calculus and a linear formulation of System F. We are\nthen able to derive results of the session calculus from the theory of the\nlambda-calculus: (1) we obtain a characterisation of inductive and coinductive\nsession types via their algebraic representations in System F; and (2) we\nextend our results to account for value and process passing, entailing strong\nnormalisation.\n",
"title": "On Polymorphic Sessions and Functions: A Tale of Two (Fully Abstract) Encodings"
}
| null | null | null | null | true | null |
14846
| null |
Default
| null | null |
null |
{
"abstract": " We study radiative neutrino pair emission in deexcitation process of atoms\ntaking into account coherence effect in a macroscopic target system. In the\ncourse of preparing the coherent initial state to enhance the rate, a spatial\nphase factor is imprinted in the macroscopic target. It is shown that this\ninitial spatial phase changes the kinematics of the radiative neutrino pair\nemission. We investigate effects of the initial spatial phase in the photon\nspectrum of the process. It turns out that the initial spatial phase provides\nus significant improvements in exploring neutrino physics such as the\nDirac-Majorana distinction and the cosmic neutrino background.\n",
"title": "Effects of initial spatial phase in radiative neutrino pair emission"
}
| null | null | null | null | true | null |
14847
| null |
Default
| null | null |
null |
{
"abstract": " The MDL two-part coding $ \\textit{index of resolvability} $ provides a\nfinite-sample upper bound on the statistical risk of penalized likelihood\nestimators over countable models. However, the bound does not apply to\nunpenalized maximum likelihood estimation or procedures with exceedingly small\npenalties. In this paper, we point out a more general inequality that holds for\narbitrary penalties. In addition, this approach makes it possible to derive\nexact risk bounds of order $1/n$ for iid parametric models, which improves on\nthe order $(\\log n)/n$ resolvability bounds. We conclude by discussing\nimplications for adaptive estimation.\n",
"title": "Finite-sample risk bounds for maximum likelihood estimation with arbitrary penalties"
}
| null | null | null | null | true | null |
14848
| null |
Default
| null | null |
null |
{
"abstract": " The current work combines the Cluster Dynamics (CD) technique and\nCALPHAD-based precipitation modeling to address the second phase precipitation\nin cold-worked (CW) 316 stainless steels (SS) under irradiation at 300-400 C.\nCD provides the radiation enhanced diffusion and dislocation evolution as\ninputs for the precipitation model. The CALPHAD-based precipitation model\ntreats the nucleation, growth and coarsening of precipitation processes based\non classical nucleation theory and evolution equations, and simulates the\ncomposition, size and size distribution of precipitate phases. We benchmark the\nmodel against available experimental data at fast reactor conditions (9.4 x\n10^-7 dpa/s and 390 C) and then use the model to predict the phase instability\nof CW 316 SS under light water reactor (LWR) extended life conditions (7 x\n10^-8 dpa/s and 275 C). The model accurately predicts the gamma-prime (Ni3Si)\nprecipitation evolution under fast reactor conditions and that the formation of\nthis phase is dominated by radiation enhanced segregation. The model also\npredicts a carbide volume fraction that agrees well with available experimental\ndata from a PWR reactor but is much higher than the volume fraction observed in\nfast reactors. We propose that radiation enhanced dissolution and/or carbon\ndepletion at sinks that occurs at high flux could be the main sources of this\ninconsistency. The integrated model predicts ~1.2% volume fraction for carbide\nand ~3.0% volume fraction for gamma-prime for typical CW 316 SS (with 0.054\nwt.% carbon) under LWR extended life conditions. This work provides valuable\ninsights into the magnitudes and mechanisms of precipitation in irradiated CW\n316 SS for nuclear applications.\n",
"title": "Integrated Modeling of Second Phase Precipitation in Cold-Worked 316 Stainless Steels under Irradiation"
}
| null | null | null | null | true | null |
14849
| null |
Default
| null | null |
null |
{
"abstract": " One dimensional hybrid systems play an important role in the search for\ntopological superconductivity. Nevertheless, all one dimensional hybrid systems\nso far have been externally defined. Here we show that one-dimensional domain\nwall in a nematic superconductor can serve as an emergent hybrid system in the\npresence of spin-orbit coupling. As a concrete setting we study the domain wall\nbetween nematic domains in FeSe, which is well established to be a nematic\nsuperconductor. We first show on the symmetry grounds that spin-triplet pairing\ncan be induced at the domain wall by constructing a Ginzburg-Landau theory. We\nthen demonstrate using Bogoliubov-de Gennes approach that such nematic domain\nwall supports zero energy bound states which would satisfy Majorana condition.\nWell-known existence of these domain walls at relatively high temperatures,\nwhich can in principle be located and investigated with scanning tunneling\nmicroscopy, presents new opportunities for a search for realization of Majorana\nbound states.\n",
"title": "Emergent topological superconductivity at nematic domain wall of FeSe"
}
| null | null | null | null | true | null |
14850
| null |
Default
| null | null |
null |
{
"abstract": " Pearson correlation and mutual information based complex networks of the\nday-to-day returns of US S&P500 stocks between 1985 and 2015 have been\nconstructed in order to investigate the mutual dependencies of the stocks and\ntheir nature. We show that both networks detect qualitative differences\nespecially during (recent) turbulent market periods thus indicating strongly\nfluctuating interconnections between the stocks of different companies in\nchanging economic environments. A measure for the strength of nonlinear\ndependencies is derived using surrogate data and leads to interesting\nobservations during periods of financial market crises. In contrast to the\nexpectation that dependencies reduce mainly to linear correlations during\ncrises we show that (at least in the 2008 crisis) nonlinear effects are\nsignificantly increasing. It turns out that the concept of centrality within a\nnetwork could potentially be used as some kind of an early warning indicator\nfor abnormal market behavior as we demonstrate with the example of the 2008\nsubprime mortgage crisis. Finally, we apply a Markowitz mean variance portfolio\noptimization and integrate the measure of nonlinear dependencies to scale the\ninvestment exposure. This leads to significant outperformance as compared to a\nfully invested portfolio.\n",
"title": "Linear and nonlinear market correlations: characterizing financial crises and portfolio optimization"
}
| null | null | null | null | true | null |
14851
| null |
Default
| null | null |
null |
{
"abstract": " We construct an explicit projective bimodule resolution for the Leavitt path\nalgebra of a row-finite quiver. We prove that the Leavitt path algebra of a\nrow-countable quiver has Hochschild cohomolgical dimension at most one, that\nis, it is quasi-free in the sense of Cuntz-Quillen. The construction of the\nresolution relies on an explicit derivation of the Leavitt path algebra.\n",
"title": "An explicit projective bimodule resolution of a Leavitt path algebra"
}
| null | null | null | null | true | null |
14852
| null |
Default
| null | null |
null |
{
"abstract": " Modern deep neural networks (DNNs) spend a large amount of their execution\ntime computing convolutions. Winograd's minimal algorithm for small\nconvolutions can greatly reduce the number of arithmetic operations. However, a\nlarge reduction in floating point (FP) operations in these algorithms can\nresult in poor numeric accuracy. In this paper we analyse the FP error and\nprove boundaries on the error. We show that the \"modified\" algorithm gives a\nsignificantly better accuracy of the result. We propose several methods for\nreducing FP error of these algorithms. Minimal convolution algorithms depend on\nthe selection of several numeric \\textit{points} that have a large impact on\nthe accuracy of the result. We propose a canonical evaluation ordering that\nboth reduces FP error and the size of the search space based on Huffman coding.\nWe study point selection experimentally, and find empirically good points. We\nalso identify the main factors that associated with sets of points that result\nin a low error. In addition, we explore other methods to reduce FP error,\nincluding mixed-precision convolution, and pairwise addition across DNN\nchannels. Using our methods we can significantly reduce FP error for a given\nblock size, which allows larger block sizes and reduced computation.\n",
"title": "Error Analysis and Improving the Accuracy of Winograd Convolution for Deep Neural Networks"
}
| null | null | null | null | true | null |
14853
| null |
Default
| null | null |
null |
{
"abstract": " Structured prediction is ubiquitous in applications of machine learning such\nas knowledge extraction and natural language processing. Structure often can be\nformulated in terms of logical constraints. We consider the question of how to\nperform efficient active learning in the presence of logical constraints among\nvariables inferred by different classifiers. We propose several methods and\nprovide theoretical results that demonstrate the inappropriateness of employing\nuncertainty guided sampling, a commonly used active learning method.\nFurthermore, experiments on ten different datasets demonstrate that the methods\nsignificantly outperform alternatives in practice. The results are of practical\nsignificance in situations where labeled data is scarce.\n",
"title": "Active Learning amidst Logical Knowledge"
}
| null | null | null | null | true | null |
14854
| null |
Default
| null | null |
null |
{
"abstract": " As mentioned by Schwartz (1974) and Cokelet (1977), it was failed to gain\nconvergent results of limiting Stokes' waves in extremely shallow water by\nmeans of perturbation methods even with the aid of extrapolation techniques\nsuch as Padé approximant. Especially, it is extremely difficult for\ntraditional analytic/numerical approaches to present the wave profile of\nlimiting waves with a sharp crest of $120^\\circ$ included angle first mentioned\nby Stokes in 1880s. Thus, traditionally, different wave models are used for\nwaves in different water depths. In this paper, by means of the homotopy\nanalysis method (HAM), an analytic approximation method for highly nonlinear\nequations, we successfully gain convergent results (and especially the wave\nprofiles) of the limiting Stokes' waves with this kind of sharp crest in\narbitrary water depth, even including solitary waves of extreme form in\nextremely shallow water, without using any extrapolation techniques. Therefore,\nin the frame of the HAM, the Stokes' wave can be used as a unified theory for\nall kinds of waves, including periodic waves in deep and intermediate depth,\ncnoidal waves in shallow water and solitary waves in extremely shallow water.\n",
"title": "On the Limiting Stokes' Wave of Extreme Height in Arbitrary Water Depth"
}
| null | null | null | null | true | null |
14855
| null |
Default
| null | null |
null |
{
"abstract": " The effects of high pressure on the crystal structure of orthorhombic (Pnma)\nperovskite type cerium scandate have been studied in situ under high pressure\nby means of synchrotron x-ray powder diffraction, using a diamond anvil cell.\nWe have found that the perovskite type crystal structure remains stable up to\n40 GPa, the highest pressure reached in the experiments. The evolution of\nunit-cell parameters with pressure has indicated an anisotropic compression.\nThe room-temperature pressure-volume equation of state is obtained from the\nexperiments. From the evolution of microscopic structural parameters like bond\ndistances and coordination polyhedra of cerium and scandium, the macroscopic\nbehavior of CeScO3 under compression has been explained and reasoned for its\nlarge pressure stability. The reported results are discussed in comparison with\nhigh-pressure results from other perovskites.\n",
"title": "Pressure impact on the stability and distortion of the crystal structure of CeScO3"
}
| null | null | null | null | true | null |
14856
| null |
Default
| null | null |
null |
{
"abstract": " The aim of this research is to design and implementation of cloud based\nlearning environment for separate division of the university. The analysis of\nexisting approaches to the construction of cloud based learning environments,\nthe formation of requirements cloud based learning tools, the selection on the\nbasis of these requirements, cloud ICT training and pilot their use for\nbuilding cloud based learning environment for separate division of the\nuniversity with the use of open source software and resources its own IT\ninfrastructure of the institution. Results of the study is planned to\ngeneralize to develop recommendations for the design of cloud based environment\nof high school.\n",
"title": "The system of cloud oriented learning tools as an element of educational and scientific environment of high school"
}
| null | null |
[
"Computer Science"
] | null | true | null |
14857
| null |
Validated
| null | null |
null |
{
"abstract": " We use integrated-light spectroscopic observations to measure metallicities\nand chemical abundances for two extragalactic young massive star clusters\n(NGC1313-379 and NGC1705-1). The spectra were obtained with the X-Shooter\nspectrograph on the ESO Very Large Telescope. We compute synthetic\nintegrated-light spectra, based on colour-magnitude diagrams for the brightest\nstars in the clusters from Hubble Space Telescope photometry and theoretical\nisochrones. Furthermore, we test the uncertainties arising from the use of\nColour Magnitude Diagram (CMD) +Isochrone method compared to an Isochrone-Only\nmethod. The abundances of the model spectra are iteratively adjusted until the\nbest fit to the observations is obtained. In this work we mainly focus on the\noptical part of the spectra. We find metallicities of [Fe/H] = $-$0.84 $\\pm$\n0.07 and [Fe/H] = $-$0.78 $\\pm$ 0.10 for NGC1313-379 and NGC1705-1,\nrespectively. We measure [$\\alpha$/Fe]=$+$0.06 $\\pm$ 0.11 for NGC1313-379 and a\nsuper-solar [$\\alpha$/Fe]=$+$0.32 $\\pm$ 0.12 for NGC1705-1. The roughly solar\n[$\\alpha$/Fe] ratio in NGC1313-379 resembles those for young stellar\npopulations in the Milky Way (MW) and the Magellanic Clouds, whereas the\nenhanced [$\\alpha$/Fe] ratio in NGC1705-1 is similar to that found for the\ncluster NGC1569-B by previous studies. Such super-solar [$\\alpha$/Fe] ratios\nare also predicted by chemical evolution models that incorporate the bursty\nstar formation histories of these dwarf galaxies. Furthermore, our\n$\\alpha$-element abundances agree with abundance measurements from H II regions\nin both galaxies. In general we derive Fe-peak abundances similar to those\nobserved in the MW and Large Magellanic Cloud (LMC) for both young massive\nclusters. For these elements, however, we recommend higher-resolution\nobservations to improve the Fe-peak abundance measurements.\n",
"title": "Chemical abundances of two extragalactic young massive clusters"
}
| null | null | null | null | true | null |
14858
| null |
Default
| null | null |
null |
{
"abstract": " The recent series 5 of the ASP system clingo provides generic means to\nenhance basic Answer Set Programming (ASP) with theory reasoning capabilities.\nWe instantiate this framework with different forms of linear constraints,\ndiscuss the respective implementations, and present techniques of how to use\nthese constraints in a reactive context. More precisely, we introduce\nextensions to clingo with difference and linear constraints over integers and\nreals, respectively, and realize them in complementary ways. Finally, we\nempirically evaluate the resulting clingo derivatives clingo[dl] and clingo[lp]\non common fragments and contrast them to related ASP systems.\nThis paper is under consideration for acceptance in TPLP.\n",
"title": "Clingo goes Linear Constraints over Reals and Integers"
}
| null | null | null | null | true | null |
14859
| null |
Default
| null | null |
null |
{
"abstract": " Flexible duplex is proposed to adapt to the channel and traffic asymmetry for\nfuture wireless networks. In this paper, we propose two novel algorithms within\nthe flexible duplex framework for joint uplink and downlink resource allocation\nin multi-cell scenario, named SAFP and RMDI, based on the awareness of\ninterference coupling among wireless links. Numerical results show significant\nperformance gain over the baseline system with fixed uplink/downlink resource\nconfiguration, and over the dynamic TDD scheme that independently adapts the\nconfiguration to time-varying traffic volume in each cell. The proposed\nalgorithms achieve two-fold increase when compared with the baseline scheme,\nmeasured by the worst-case quality of service satisfaction level, under a low\nlevel of traffic asymmetry. The gain is more significant when the traffic is\nhighly asymmetric, as it achieves three-fold increase.\n",
"title": "Dynamic Uplink/Downlink Resource Management in Flexible Duplex-Enabled Wireless Networks"
}
| null | null | null | null | true | null |
14860
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we establish the Carleman estimates for forward and backward\nstochastic fourth order Schrödinger equations, on basis of which, we can\nobtain the observability, unique continuation property and the exact\ncontrollability for the forward and backward stochastic fourth order\nSchrödinger equations.\n",
"title": "Carleman estimates for forward and backward stochastic fourth order Schrödinger equations and their applications"
}
| null | null | null | null | true | null |
14861
| null |
Default
| null | null |
null |
{
"abstract": " Doubts have been expressed in a comment (Eur. J. Phys., 39, 018001, 2018),\nabout the tenability of the formulation for radiative losses in our recent\npublished work (Eur. J. Phys., 37, 045210, 2016). We provide our reply to the\ncomment. In particular, it is pointed out that one need to clearly distinguish\nbetween the rate of the energy-momentum being carried by the electromagnetic\nradiation to far-off space, and that of the mechanical energy-momentum losses\nbeing incurred by the radiating charge. It is also demonstrated that while the\nPoynting flux is always positive through a spherical surface centred on the\nretarded position of the charge, it could surprisingly be negative through a\nsurface centred on the \"present\" position of the charge. It is further shown\nthat the mysterious Schott term, hitherto thought in literature to arise from\nsome acceleration-dependent energy in fields, is actually nothing but the\ndifference in rate of change of energy in self-fields of the charge between the\nretarded and present times.\n",
"title": "Reply to comment on `Poynting flux in the neighbourhood of a point charge in arbitrary motion and the radiative power losses'"
}
| null | null | null | null | true | null |
14862
| null |
Default
| null | null |
null |
{
"abstract": " A notion of delegated causality is introduced. This subtle kind of causality\nis dual to interventional causality. Delegated causality elucidates the causal\nrole of dynamical systems at the \"edge of chaos\", explicates evident cases of\ndownward causation, and relates emergent phenomena to Godel's incompleteness\ntheorem. Apparently rich implications are noticed in biology and Chinese\nphilosophy.\n",
"title": "Delegated Causality of Complex Systems"
}
| null | null |
[
"Physics"
] | null | true | null |
14863
| null |
Validated
| null | null |
null |
{
"abstract": " Resonating valence bond (RVB) theory of high Tc superconductivity, an\nelectron correlation based mechanism, began as an insightful response by\nAnderson, to Bednorz and Muller's discovery of high Tc superconductivity in\ncuprates in late 1986. Shortly a theoretical framework for quantum spin liquids\nand superconductivity was developed. This theory adresses a formidable strong\ncoupling quantum manybody problem, in modern times. It is built on certain key\nexperimental facts: i) survival of a dynamical Mott localization in a metallic\nstate, ii) proliferation of bond singlets and iii) absence of fermi liquid\nquasi particles. After summarising RVB theory I will provide an aerial view of,\nmostly, new superconductors where I believe that, to a large degree RVB\nmechanism is at work and indicate prospects for even higher Tc's.\n",
"title": "Resonating Valence Bond Theory of Superconductivity: Beyond Cuprates"
}
| null | null | null | null | true | null |
14864
| null |
Default
| null | null |
null |
{
"abstract": " In this paper, we consider numerical approximations of a hydrodynamically\ncoupled phase field diblock copolymer model, in which the free energy contains\na kinetic potential, a gradient entropy, a Ginzburg-Landau double well\npotential, and a long range nonlocal type potential. We develop a set of second\norder time marching schemes for this system using the \"Invariant Energy\nQuadratization\" approach for the double well potential, the projection method\nfor the Navier-Stokes equation, and a subtle implicit-explicit treatment for\nthe stress and convective term. The resulting schemes are linear and lead to\nsymmetric positive definite systems at each time step, thus they can be\nefficiently solved. We further prove that these schemes are unconditionally\nenergy stable. Various numerical experiments are performed to validate the\naccuracy and energy stability of the proposed schemes.\n",
"title": "Efficient and accurate numerical schemes for a hydrodynamically coupled phase field diblock copolymer model"
}
| null | null | null | null | true | null |
14865
| null |
Default
| null | null |
null |
{
"abstract": " The Klein-Kramers equation, governing the Brownian motion of a classical\nparticle in quantum environment under the action of an arbitrary external\npotential, is derived. Quantum temperature and friction operators are\nintroduced and at large friction the corresponding Smoluchowski equation is\nobtained. Introducing the Bohm quantum potential, this Smoluchowski equation is\nextended to describe the Brownian motion of a quantum particle in quantum\nenvironment.\n",
"title": "Brownian Motion of a Classical Particle in Quantum Environment"
}
| null | null | null | null | true | null |
14866
| null |
Default
| null | null |
null |
{
"abstract": " Learning an optimal policy from a multi-modal reward function is a\nchallenging problem in reinforcement learning (RL). Hierarchical RL (HRL)\ntackles this problem by learning a hierarchical policy, where multiple option\npolicies are in charge of different strategies corresponding to modes of a\nreward function and a gating policy selects the best option for a given\ncontext. Although HRL has been demonstrated to be promising, current\nstate-of-the-art methods cannot still perform well in complex real-world\nproblems due to the difficulty of identifying modes of the reward function. In\nthis paper, we propose a novel method called hierarchical policy search via\nreturn-weighted density estimation (HPSDE), which can efficiently identify the\nmodes through density estimation with return-weighted importance sampling. Our\nproposed method finds option policies corresponding to the modes of the return\nfunction and automatically determines the number and the location of option\npolicies, which significantly reduces the burden of hyper-parameters tuning.\nThrough experiments, we demonstrate that the proposed HPSDE successfully learns\noption policies corresponding to modes of the return function and that it can\nbe successfully applied to a challenging motion planning problem of a redundant\nrobotic manipulator.\n",
"title": "Hierarchical Policy Search via Return-Weighted Density Estimation"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14867
| null |
Validated
| null | null |
null |
{
"abstract": " Several Fourier transformations of functions of one and two variables are\nevaluated and then used to derive some integral and series identities. It is\nshown that certain two- dimensional Mordell integrals factorize into product of\ntwo integrals and that the square of the absolute value of the Mordell integral\ncan be reduced to a single one-dimensional integral. Some connections to\nelliptic functions and lattice sums are discussed.\n",
"title": "Two-dimensional Fourier transformations and Mordell integrals"
}
| null | null | null | null | true | null |
14868
| null |
Default
| null | null |
null |
{
"abstract": " Shunt FACTS devices, such as, a Static Var Compensator (SVC), are capable of\nproviding local reactive power compensation. They are widely used in the\nnetwork to reduce the real power loss and improve the voltage profile. This\npaper proposes a planning model based on mixed integer conic programming (MICP)\nto optimally allocate SVCs in the transmission network considering load\nuncertainty. The load uncertainties are represented by a number of scenarios.\nReformulation and linearization techniques are utilized to transform the\noriginal non-convex model into a convex second order cone programming (SOCP)\nmodel. Numerical case studies based on the IEEE 30-bus system demonstrate the\neffectiveness of the proposed planning model.\n",
"title": "Optimal Allocation of Static Var Compensator via Mixed Integer Conic Programming"
}
| null | null | null | null | true | null |
14869
| null |
Default
| null | null |
null |
{
"abstract": " The established spin splitting in monolayer (ML) of transition metal\ndichalcogenides (TMDs) that is caused by inversion symmetry breaking is\ndictated by mirror symmetry operations to exhibit fully out-of-plane direction\nof spin polarization. Through first-principles density functional theory\ncalculations, we show that polarity-induced mirror symmetry breaking leads to\nnew sizable spin splitting having in-plane spin polarization. These splittings\nare effectively controlled by tuning the polarity using biaxial strain.\nFurthermore, the admixtures of the out-of-plane and in-plane spin-polarized\nstates in the strained polar systems are identified, which is expected to\ninfluence the spin relaxation through the Dyakonov-Perel mechanism. Our study\nclarified that the polarity-induced mirror symmetry breaking plays an important\nrole in controlling the spin splitting and spin relaxation in the TMDs ML,\nwhich is useful for designing future spintronic devices.\n",
"title": "Polarity tuning of spin-orbit-induced spin splitting in two-dimensional transition metal dichalcogenides semiconductors"
}
| null | null | null | null | true | null |
14870
| null |
Default
| null | null |
null |
{
"abstract": " Completeness of a dynamic priority scheduling scheme is of fundamental\nimportance for the optimal control of queues in areas as diverse as computer\ncommunications, communication networks, supply chains and manufacturing\nsystems. Our first main contribution is to identify the mean waiting time\ncompleteness as a unifying aspect for four different dynamic priority\nscheduling schemes by proving their completeness and equivalence in 2-class\nM/G/1 queue. These dynamic priority schemes are earliest due date based, head\nof line priority jump, relative priority, and probabilistic priority.\nIn our second main contribution, we characterize the optimal scheduling\npolicies for the case studies in different domains by exploiting the\ncompleteness of above dynamic priority schemes. The major theme of second main\ncontribution is resource allocation/optimal control in revenue management\nproblems for contemporary systems such as cloud computing, high-performance\ncomputing, etc., where congestion is inherent. Using completeness and\ntheoretically tractable nature of relative priority policy, we study the impact\nof approximation in a fairly generic data network utility framework. We\nintroduce the notion of min-max fairness in multi-class queues and show that a\nsimple global FCFS policy is min-max fair. Next, we re-derive the celebrated\n$c/\\rho$ rule for 2-class M/G/1 queues by an elegant argument and also simplify\na complex joint pricing and scheduling problem for a wider class of scheduling\npolicies.\n",
"title": "Some parametrized dynamic priority policies for 2-class M/G/1 queues: completeness and applications"
}
| null | null | null | null | true | null |
14871
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we obtain a description of the Grothendieck group of complex\nvector bundles over the classifying space of a p-local finite group in terms of\nrepresentation rings of subgroups of its Sylow. We also prove a stable elements\nformula for generalized cohomological invariants of p-local finite groups,\nwhich is used to show the existence of unitary embeddings of p-local finite\ngroups. Finally, we show that the augmentation map for the cochains of the\nclassifying space of a p-local finite group is Gorenstein in the sense of\nDwyer-Greenlees-Iyengar and obtain some consequences about the cohomology ring\nof these classifying spaces.\n",
"title": "Vector bundles over classifying spaces of p-local finite groups and Benson-Carlson duality"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14872
| null |
Validated
| null | null |
null |
{
"abstract": " Colletotrichum represent a genus of fungal species primarily known as plant\npathogens with severe economic impacts in temperate, subtropical and tropical\nclimates Consensus taxonomy and classification systems for Colletotrichum\nspecies have been undergoing revision as high resolution genomic data becomes\navailable. Here we propose an alternative annotation that provides a complete\nsequence for a Colletotrichum YPT1 gene homolog using the whole genome shotgun\nsequence of Colletotrichum incanum isolated from soybean crops in Illinois,\nUSA.\n",
"title": "Identification of a complete YPT1 Rab GTPase sequence from the fungal pathogen Colletotrichum incanum"
}
| null | null | null | null | true | null |
14873
| null |
Default
| null | null |
null |
{
"abstract": " The space of based loops in $SL_n(\\mathbb{C})$, also known as the affine\nGrassmannian of $SL_n(\\mathbb{C})$, admits an $\\mathbb{E}_2$ or fusion product.\nWork of Mitchell and Richter proves that this based loop space stably splits as\nan infinite wedge sum. We prove that the Mitchell--Richter splitting is\ncoherently multiplicative, but not $\\mathbb{E}_2$. Nonetheless, we show that\nthe splitting becomes $\\mathbb{E}_2$ after base-change to complex cobordism.\nOur proof of the $\\mathbb{A}_\\infty$ splitting involves on the one hand an\nanalysis of the multiplicative properties of Weiss calculus, and on the other a\nuse of Beilinson--Drinfeld Grassmannians to verify a conjecture of Mahowald and\nRichter. Other results are obtained by explicit, obstruction-theoretic\ncomputations.\n",
"title": "Multiplicative Structure in the Stable Splitting of $ΩSL_n(\\mathbb{C})$"
}
| null | null | null | null | true | null |
14874
| null |
Default
| null | null |
null |
{
"abstract": " We study Le Potier's strange duality conjecture on a rational surface. We\nfocus on the strange duality map $SD_{c_n^r,L}$ which involves the moduli space\nof rank $r$ sheaves with trivial first Chern class and second Chern class $n$,\nand the moduli space of 1-dimensional sheaves with determinant $L$ and Euler\ncharacteristic 0. We show there is an exact sequence relating the map\n$SD_{c_r^r,L}$ to $SD_{c^{r-1}_{r},L}$ and $SD_{c_r^r,L\\otimes K_X}$ for all\n$r\\geq1$ under some conditions on $X$ and $L$ which applies to a large number\nof cases on $\\p^2$ or Hirzebruch surfaces . Also on $\\mathbb{P}^2$ we show that\nfor any $r>0$, $SD_{c^r_r,dH}$ is an isomorphism for $d=1,2$, injective for\n$d=3$ and moreover $SD_{c_3^3,rH}$ and $SD_{c_3^2,rH}$ are injective. At the\nend we prove that the map $SD_{c_n^2,L}$ ($n\\geq2$) is an isomorphism for\n$X=\\mathbb{P}^2$ or Fano rational ruled surfaces and $g_L=3$, and hence so is\n$SD_{c_3^3,L}$ as a corollary of our main result.\n",
"title": "Strange duality on rational surfaces II: higher rank cases"
}
| null | null | null | null | true | null |
14875
| null |
Default
| null | null |
null |
{
"abstract": " Graph signals offer a very generic and natural representation for data that\nlives on networks or irregular structures. The actual data structure is however\noften unknown a priori but can sometimes be estimated from the knowledge of the\napplication domain. If this is not possible, the data structure has to be\ninferred from the mere signal observations. This is exactly the problem that we\naddress in this paper, under the assumption that the graph signals can be\nrepresented as a sparse linear combination of a few atoms of a structured graph\ndictionary. The dictionary is constructed on polynomials of the graph\nLaplacian, which can sparsely represent a general class of graph signals\ncomposed of localized patterns on the graph. We formulate a graph learning\nproblem, whose solution provides an ideal fit between the signal observations\nand the sparse graph signal model. As the problem is non-convex, we propose to\nsolve it by alternating between a signal sparse coding and a graph update step.\nWe provide experimental results that outline the good graph recovery\nperformance of our method, which generally compares favourably to other recent\nnetwork inference algorithms.\n",
"title": "Graph learning under sparsity priors"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14876
| null |
Validated
| null | null |
null |
{
"abstract": " We present an evaluation of several representative sampling-based and\noptimization-based motion planners, and then introduce an integrated motion\nplanning system which incorporates recent advances in trajectory optimization\ninto a sparse roadmap framework. Through experiments in 4 common application\nscenarios with 5000 test cases each, we show that optimization-based or\nsampling-based planners alone are not effective for realistic problems where\nfast planning times are required. To the best of our knowledge, this is the\nfirst work that presents such a systematic and comprehensive evaluation of\nstate-of-the-art motion planners, which are based on a significant amount of\nexperiments. We then combine different stand-alone planners with trajectory\noptimization. The results show that the combination of our sparse roadmap and\ntrajectory optimization provides superior performance over other standard\nsampling-based planners combinations. By using a multi-query roadmap instead of\ngenerating completely new trajectories for each planning problem, our approach\nallows for extensions such as persistent control policy information associated\nwith a trajectory across planning problems. Also, the sub-optimality resulting\nfrom the sparsity of roadmap, as well as the unexpected disturbances from the\nenvironment, can both be overcome by the real-time trajectory optimization\nprocess.\n",
"title": "Improving Trajectory Optimization using a Roadmap Framework"
}
| null | null | null | null | true | null |
14877
| null |
Default
| null | null |
null |
{
"abstract": " Many policy search algorithms have been proposed for robot learning and\nproved to be practical in real robot applications. However, there are still\nhyperparameters in the algorithms, such as the exploration rate, which requires\nmanual tuning. The existing methods to design the exploration rate manually or\nautomatically may not be general enough or hard to apply in the real robot. In\nthis paper, we propose a learning model to update the exploration rate\nadaptively. The overall algorithm is a combination of methods proposed by other\nresearchers. Smooth trajectories for the robot can be produced by the algorithm\nand the updated exploration rate maximizes the lower bound of the expected\nreturn. Our method is tested in the ball-in-cup problem. The results show that\nour method can receive the same learning outcome as the previous methods but\nwith fewer iterations.\n",
"title": "Smooth and Efficient Policy Exploration for Robot Trajectory Learning"
}
| null | null | null | null | true | null |
14878
| null |
Default
| null | null |
null |
{
"abstract": " The ideas that we forge creatively as individuals and groups build on one\nanother in a manner that is cumulative and adaptive, forming open-ended\nlineages across space and time. Thus, human culture is believed to evolve. The\npervasiveness of cross-domain creativity--as when a song inspires a\npainting--would appear indicative of discontinuities in cultural lineages.\nHowever, if what evolves through culture is our worldviews--the webs of\nthoughts, ideas, and attitudes that constitutes our way of seeing being in the\nworld--then the problem of discontinuities is solved. The state of a worldview\ncan be affected by information assimilated in one domain, and this\nchange-of-state can be expressed in another domain. In this view, the gesture,\nnarrative, or artifact that constitutes a specific creative act is not what is\nevolving; it is merely the external manifestation of the state of an evolving\nworldview. Like any evolutionary process, cultural evolution requires a balance\nbetween novelty, via the generation of variation, and continuity, via the\npreservation of variants that are adaptive. In cultural evolution, novelty is\ngenerated through creativity, and continuity is provided by social learning\nprocesses, e.g., imitation. Both the generative and imitative aspects of\ncultural evolution are affected by social media. We discuss the trajectory from\nsocial ideation to social innovation, focusing on the role of\nself-organization, renewal, and perspective-taking at the individual and social\ngroup level.\n",
"title": "Social Innovation and the Evolution of Creative, Sustainable Worldviews"
}
| null | null |
[
"Quantitative Biology"
] | null | true | null |
14879
| null |
Validated
| null | null |
null |
{
"abstract": " Elucidating the interaction between magnetic moments and itinerant carriers\nis an important step to spintronic applications. Here, we investigate magnetic\nand transport properties in d0 ferromagnetic SiC single crystals prepared by\npostimplantation pulsed laser annealing. Magnetic moments are contributed by\nthe p states of carbon atoms, but their magnetic circular dichroism is\ndifferent from that in semi-insulating SiC samples. The anomalous Hall effect\nand negative magnetoresistance indicate the influence of d0 spin order on free\ncarriers. The ferromagnetism is relatively weak in N-implanted SiC compared\nwith that in Al-implanted SiC after annealing. The results suggest that d0\nmagnetic moments and itinerant carriers can interact with each other, which\nwill facilitate the development of SiC spintronic devices with d0\nferromagnetism.\n",
"title": "Interaction between magnetic moments and itinerant carriers in d0 ferromagnetic SiC"
}
| null | null | null | null | true | null |
14880
| null |
Default
| null | null |
null |
{
"abstract": " Sylvester factor, an essential part of the asymptotic formula of Hardy and\nLittlewood which is the extended Goldbach conjecture, regarded as strongly\nmultiplicative arithmetic function, has several remarkable properties.\n",
"title": "Il Fattore di Sylvester"
}
| null | null | null | null | true | null |
14881
| null |
Default
| null | null |
null |
{
"abstract": " We study analytically and numerically envelope solitons (bright and gap\nsolitons) in a one-dimensional, nonlinear acoustic metamaterial, composed of an\nair-filled waveguide periodically loaded by clamped elastic plates. Based on\nthe transmission line approach, we derive a nonlinear dynamical lattice model\nwhich, in the continuum approximation, leads to a nonlinear, dispersive and\ndissipative wave equation. Applying the multiple scales perturbation method, we\nderive an effective lossy nonlinear Schrödinger equation and obtain\nanalytical expressions for bright and gap solitons. We also perform direct\nnumerical simulations to study the dissipation-induced dynamics of the bright\nand gap solitons. Numerical and analytical results, relying on the analytical\napproximations and perturbation theory for solions, are found to be in good\nagreement.\n",
"title": "Bright and Gap Solitons in Membrane-Type Acoustic Metamaterials"
}
| null | null | null | null | true | null |
14882
| null |
Default
| null | null |
null |
{
"abstract": " Decision making based on behavioral and neural observations of living systems\nhas been extensively studied in brain science, psychology, and other\ndisciplines. Decision-making mechanisms have also been experimentally\nimplemented in physical processes, such as single photons and chaotic lasers.\nThe findings of these experiments suggest that there is a certain common basis\nin describing decision making, regardless of its physical realizations. In this\nstudy, we propose a local reservoir model to account for choice-based learning\n(CBL). CBL describes decision consistency as a phenomenon where making a\ncertain decision increases the possibility of making that same decision again\nlater, which has been intensively investigated in neuroscience, psychology,\netc. Our proposed model is inspired by the viewpoint that a decision is\naffected by its local environment, which is referred to as a local reservoir.\nIf the size of the local reservoir is large enough, consecutive decision making\nwill not be affected by previous decisions, thus showing lower degrees of\ndecision consistency in CBL. In contrast, if the size of the local reservoir\ndecreases, a biased distribution occurs within it, which leads to higher\ndegrees of decision consistency in CBL. In this study, an analytical approach\non local reservoirs is presented, as well as several numerical demonstrations.\nFurthermore, a physical architecture for CBL based on single photons is\ndiscussed, and the effects of local reservoirs is numerically demonstrated.\nDecision consistency in human decision-making tasks and in recruiting empirical\ndata are evaluated based on local reservoir. In summary, the proposed local\nreservoir model paves a path toward establishing a foundation for computational\nmechanisms and the systematic analysis of decision making on different levels.\n",
"title": "Local reservoir model for choice-based learning"
}
| null | null | null | null | true | null |
14883
| null |
Default
| null | null |
null |
{
"abstract": " Inspired by mirror symmetry, we investigate some differential geometric\naspects of the space of Bridgeland stability conditions on a Calabi-Yau\ntriangulated category. The aim is to develop theory of Weil-Petersson geometry\non the stringy Kähler moduli space. A few basic examples are studied. In\nparticular, we identify our Weil-Petersson metric with the Bergman metric on a\nSiegel modular variety in the case of the self-product of an elliptic curve.\n",
"title": "Weil-Petersson geometry on the space of Bridgeland stability conditions"
}
| null | null |
[
"Mathematics"
] | null | true | null |
14884
| null |
Validated
| null | null |
null |
{
"abstract": " For sputter depth profiling often sample erosion by Ar+ ions is used. Only a\nhigh purity of the sputter gas and a low contamination level of the ion gun\navoids misleading depth profile measurements results. Here a new measurement\nprocedure is presented, which monitors these parameters. A Si sample is\nsputtered inside the instrument and then the surface concentration of the\nelements Ar, C, N and O is measured. Results of such measurements of an XPS\nmicroprobe PHI Quantum 2000, which were recorded over a period of 10 years, are\npresented.\n",
"title": "Analysis of a Sputtered Si Surface for Ar Sputter Gas Supply Purity Monitoring"
}
| null | null | null | null | true | null |
14885
| null |
Default
| null | null |
null |
{
"abstract": " Using a quantum wave packet simulation including the nuclear and electronic\ndegrees of freedom, we investigate the femtosecond and picosecond energy- and\nangle-resolved photoelectron spectra of the E($^1\\Sigma_g^+$) electronic state\nof Li$_2$. We find that the angular distributions of the emitted photoelectrons\ndepend strongly on the pulse duration in the regime of ultrashort laser pulses.\nThis effect is illustrated by the extraction of a time-dependent asymmetry\nparameter whose variation with pulse duration can be explained by an incoherent\naverage over different ion rotational quantum numbers. We then derive for the\nvariation of the asymmetry parameter a simple analytical formula, which can be\nused to extract the asymptotic CW asymmetry parameters of individual\ntransitions from measurements performed with ultra-short pulses.\n",
"title": "Extracting spectroscopic molecular parameters from short pulse photo-electron angular distributions"
}
| null | null | null | null | true | null |
14886
| null |
Default
| null | null |
null |
{
"abstract": " We propose a theoretical framework for thinking about score normalization,\nwhich confirms that normalization is not needed under (admittedly fragile)\nideal conditions. If, however, these conditions are not met, e.g. under\ndata-set shift between training and runtime, our theory reveals dependencies\nbetween scores that could be exploited by strategies such as score\nnormalization. Indeed, it has been demonstrated over and over experimentally,\nthat various ad-hoc score normalization recipes do work. We present a first\nattempt at using probability theory to design a generative score-space\nnormalization model which gives similar improvements to ZT-norm on the\ntext-dependent RSR 2015 database.\n",
"title": "A Generative Model for Score Normalization in Speaker Recognition"
}
| null | null | null | null | true | null |
14887
| null |
Default
| null | null |
null |
{
"abstract": " This work focuses on reliable detection and segmentation of bird\nvocalizations as recorded in the open field. Acoustic detection of avian sounds\ncan be used for the automatized monitoring of multiple bird taxa and querying\nin long-term recordings for species of interest. These tasks are tackled in\nthis work, by suggesting two approaches: A) First, DenseNets are applied to\nweekly labeled data to infer the attention map of the dataset (i.e. Salience\nand CAM). We push further this idea by directing attention maps to the YOLO v2\nDeepnet-based, detection framework to localize bird vocalizations. B) A deep\nautoencoder, namely the U-net, maps the audio spectrogram of bird vocalizations\nto its corresponding binary mask that encircles the spectral blobs of\nvocalizations while suppressing other audio sources. We focus solely on\nprocedures requiring minimum human attendance, suitable to scan massive volumes\nof data, in order to analyze them, evaluate insights and hypotheses and\nidentify patterns of bird activity. Hopefully, this approach will be valuable\nto researchers, conservation practitioners, and decision makers that need to\ndesign policies on biodiversity issues.\n",
"title": "Deep Networks tag the location of bird vocalisations on audio spectrograms"
}
| null | null | null | null | true | null |
14888
| null |
Default
| null | null |
null |
{
"abstract": " Telephone call centers offer a convenient communication channel between\nbusinesses and their customers. Efficient management of call centers needs\naccurate modeling of customer waiting behavior, which contains important\ninformation about customer patience (how long a customer is willing to wait)\nand service quality (how long a customer needs to wait to get served). Hazard\nfunctions offer dynamic characterization of customer waiting behavior, and\nprovide critical inputs for agent scheduling. Motivated by this application, we\ndevelop a two-way functional hazards (tF-Hazards) model to study customer\nwaiting behavior as a function of two timescales, waiting duration and the time\nof day that a customer calls in. The model stems from a two-way piecewise\nconstant hazard function, and imposes low-rank structure and smoothness on the\nhazard rates to enhance interpretability. We exploit an alternating direction\nmethod of multipliers (ADMM) algorithm to optimize a penalized likelihood\nfunction of the model. We carefully analyze the data from a US bank call\ncenter, and provide informative insights about customer patience and service\nquality patterns along waiting time and across different times of a day. The\nfindings provide primitive inputs for call center agent staffing and\nscheduling, as well as for call center practitioners to understand the effect\nof system protocols on customer waiting behavior.\n",
"title": "To Wait or Not to Wait: Two-way Functional Hazards Model for Understanding Waiting in Call Centers"
}
| null | null | null | null | true | null |
14889
| null |
Default
| null | null |
null |
{
"abstract": " Areal level spatial data are often large, sparse and may appear with\ngeographical shapes that are regular or irregular (e.g., postcode). Moreover,\nsometimes it is important to obtain predictive inference in regular or\nirregular areal shapes that is misaligned with the observed spatial areal\ngeographical boundary. For example, in a survey the respondents were asked\nabout their postcode, however for policy making purposes, researchers are often\ninterested to obtain information at the SA2. The statistical challenge is to\nobtain spatial prediction at the SA2s, where the SA2s may have overlapped\ngeographical boundaries with postcodes. The study is motivated by a practical\nsurvey data obtained from the Australian National University (ANU) Poll. Here\nthe main research question is to understand respondents' satisfaction level\nwith the way Australia is heading. The data are observed at 1,944 postcodes\namong the 2,516 available postcodes across Australia, and prediction is\nobtained at the 2,196 SA2s. The proposed method also explored through a\ngrid-based simulation study, where data have been observed in a regular grid\nand spatial prediction has been done in a regular grid that has a misaligned\ngeographical boundary with the first regular grid-set. The real-life example\nwith ANU Poll data addresses the situation of irregular geographical boundaries\nthat are misaligned, i.e., model fitted with postcode data and hence obtained\nprediction at the SA2. A comparison study is also performed to validate the\nproposed method. In this paper, a Gaussian model is constructed under Bayesian\nhierarchy. The novelty lies in the development of the basis function that can\naddress spatial sparsity and localised spatial structure. It can also address\nthe large-dimensional spatial data modelling problem by constructing knot based\nreduced-dimensional basis functions.\n",
"title": "Bayesian Gaussian models for interpolating large-dimensional data at misaligned areal units"
}
| null | null | null | null | true | null |
14890
| null |
Default
| null | null |
null |
{
"abstract": " The ECIR half-day workshop on Task-Based and Aggregated Search (TBAS) was\nheld in Barcelona, Spain on 1 April 2012. The program included a keynote talk\nby Professor Jarvelin, six full paper presentations, two poster presentations,\nand an interactive discussion among the approximately 25 participants. This\nreport overviews the aims and contents of the workshop and outlines the major\noutcomes.\n",
"title": "Report on TBAS 2012: Workshop on Task-Based and Aggregated Search"
}
| null | null | null | null | true | null |
14891
| null |
Default
| null | null |
null |
{
"abstract": " Determining the velocity distribution of halo stars is essential for\nestimating the mass of the Milky Way and for inferring its formation history.\nSince the stellar halo is a dynamically hot system, the velocity distribution\nof halo stars is well described by the 3-dimensional velocity dispersions\n$(\\sigma_r, \\sigma_\\theta, \\sigma_\\phi)$, or by the velocity anisotropy\nparameter $\\beta=1-(\\sigma_\\theta^2+\\sigma_\\phi^2)/(2\\sigma_r^2)$. Direct\nmeasurements of $(\\sigma_r, \\sigma_\\theta, \\sigma_\\phi)$ consistently suggest\n$\\beta =0.5$-$0.7$ for nearby halo stars. In contrast, the value of $\\beta$ at\nlarge Galactocentric radius $r$ is still controversial, since reliable proper\nmotion data are available for only a handful of stars. In the last decade,\nseveral authors have tried to estimate $\\beta$ for distant halo stars by\nfitting the observed line-of-sight velocities at each radius with simple\nvelocity distribution models (local fitting methods). Some results of local\nfitting methods imply $\\beta<0$ at $r \\gtrsim 20 \\;\\rm{kpc}$, which is\ninconsistent with recent predictions from cosmological simulations. Here we\nperform mock-catalogue analyses to show that the estimates of $\\beta$ based on\nlocal fitting methods are reliable only at $r \\leq 15 \\;\\rm{kpc}$ with the\ncurrent sample size ($\\sim10^3$ stars at a given radius). As $r$ increases, the\nline-of-sight velocity (corrected for the Solar reflex motion) becomes\nincreasingly closer to the Galactocentric radial velocity, so that it becomes\nincreasingly more difficult to estimate tangential velocity dispersion\n$(\\sigma_\\theta, \\sigma_\\phi)$ from line-of-sight velocity distribution. Our\nresults suggest that the forthcoming Gaia data will be crucial for\nunderstanding the velocity distribution of halo stars at $r \\geq 20\\;\\rm{kpc}$.\n",
"title": "Reliability of the measured velocity anisotropy of the Milky Way stellar halo"
}
| null | null | null | null | true | null |
14892
| null |
Default
| null | null |
null |
{
"abstract": " In this paper we study the asymptotic behavior of second-order uniformly\nelliptic operators on weighted Riemannian manifolds. We appeal to the notion of\n\\mbox{$H$-convergence} introduced by Murat and Tartar. In our main result we\nestablish an \\mbox{$H$-compactness} result that applies to elliptic operators\nwith measurable, uniformly elliptic coefficients on weighted Riemannian\nmanifolds. We further discuss the special case of \"locally periodic\"\ncoefficients and study the asymptotic behavior of the Laplace-Beltrami operator\non families of weighted manifolds obtained from a reference manifold by a\nconformal (rapidly oscillating) change of metrics.\n",
"title": "$H$-compactness of elliptic operators on weighted Riemannian Manifolds"
}
| null | null | null | null | true | null |
14893
| null |
Default
| null | null |
null |
{
"abstract": " Symbol-pair codes, introduced by Cassuto and Blaum [1], have been raised for\nsymbol-pair read channels. This new idea is motivated by the limitation of the\nreading process in high-density data storage technologies. Yaakobi et al. [8]\nintroduced codes for $b$-symbol read channels, where the read operation is\nperformed as a consecutive sequence of $b>2$ symbols. In this paper, we come up\nwith a method to compute the $b$-symbol-pair distance of two $n$-tuples, where\n$n$ is a positive integer. Also, we deal with the $b$-symbol-pair distances of\nsome kind of cyclic codes of length $p^e$ over $\\mathbb{F}_{p^m}$.\n",
"title": "$b$-symbol distance distribution of repeated-root cyclic codes"
}
| null | null | null | null | true | null |
14894
| null |
Default
| null | null |
null |
{
"abstract": " We present a baseline approach for cross-modal knowledge fusion. Different\nbasic fusion methods are evaluated on existing embedding approaches to show the\npotential of joining knowledge about certain concepts across modalities in a\nfused concept representation.\n",
"title": "Knowledge Fusion via Embeddings from Text, Knowledge Graphs, and Images"
}
| null | null |
[
"Computer Science",
"Statistics"
] | null | true | null |
14895
| null |
Validated
| null | null |
null |
{
"abstract": " From topology of the order parameter of the magnon condensate observed in\nyttrium-iron-garnet (YIG) magnetic films one must not expect energetic barriers\nmaking spin supercurrents metastable. But we show that some barriers of\ndynamical origin are possible nevertheless until the gradient of the phase\n(angle of spin precession) does not exceed the critical value (analog of the\nLandau critical velocity in superfluids). On the other hand, recently published\nclaims of experimental detection of spin superfluidity in YIG films and\nantiferromagnets are not justified, and spin superfluidity in magnetically\nordered solids has not yet been experimentally confirmed.\n",
"title": "Observation of spin superfluidity: YIG magnetic films and beyond"
}
| null | null | null | null | true | null |
14896
| null |
Default
| null | null |
null |
{
"abstract": " Van der Waals (vdW) heterostructures are receiving great attentions due to\ntheir intriguing properties and potentials in many research fields. The flow of\ncharge carriers in vdW heterostructures can be efficiently rectified by the\ninter-layer coupling between neighboring layers, offering a rich collection of\nfunctionalities and a mechanism for designing atomically thin devices.\nNevertheless, non-uniform contact in larger-area heterostructures reduces the\ndevice efficiency. In this work, ion irradiation had been verified as an\nefficient technique to enhance the contact and interlayer coupling in the newly\ndeveloped graphene/WSe2 hetero-structure with a large area of 10 mm x 10 mm.\nDuring the ion irradiation process, the morphology of monolayer graphene had\nbeen modified, promoting the contact with WSe2. Experimental evidences of the\ntunable interlayer electron transfer are displayed by investigation of\nphotoluminescence and ultrafast absorption of the irradiated heterostructure.\nBesides, we have found that in graphene/WSe2 heterostructure, graphene serves\nas a fast channel for the photo-excited carriers to relax in WSe2, and the\nnonlinear absorption of WSe2 could be effectively tuned by the carrier transfer\nprocess in graphene, enabling specific optical absorption of the\nheterostructure in comparison with separated graphene or WSe2. On the basis of\nthese new findings, by applying the ion beam modified graphene/WSe2\nheterostructure as a saturable absorber, Q-switched pulsed lasing with\noptimized performance has been realized in a Nd:YAG waveguide cavity. This work\npaves the way towards developing novel devices based on large-area\nheterostructures by using ion beam irradiation.\n",
"title": "Tuning of Interlayer Coupling in Large-Area Graphene/WSe2 van der Waals Heterostructure via Ion Irradiation: Optical Evidences and Photonic Applications"
}
| null | null | null | null | true | null |
14897
| null |
Default
| null | null |
null |
{
"abstract": " One of the long-standing challenges in Artificial Intelligence for learning\ngoal-directed behavior is to build a single agent which can solve multiple\ntasks. Recent progress in multi-task learning for goal-directed sequential\nproblems has been in the form of distillation based learning wherein a student\nnetwork learns from multiple task-specific expert networks by mimicking the\ntask-specific policies of the expert networks. While such approaches offer a\npromising solution to the multi-task learning problem, they require supervision\nfrom large expert networks which require extensive data and computation time\nfor training. In this work, we propose an efficient multi-task learning\nframework which solves multiple goal-directed tasks in an on-line setup without\nthe need for expert supervision. Our work uses active learning principles to\nachieve multi-task learning by sampling the harder tasks more than the easier\nones. We propose three distinct models under our active sampling framework. An\nadaptive method with extremely competitive multi-tasking performance. A\nUCB-based meta-learner which casts the problem of picking the next task to\ntrain on as a multi-armed bandit problem. A meta-learning method that casts the\nnext-task picking problem as a full Reinforcement Learning problem and uses\nactor critic methods for optimizing the multi-tasking performance directly. We\ndemonstrate results in the Atari 2600 domain on seven multi-tasking instances:\nthree 6-task instances, one 8-task instance, two 12-task instances and one\n21-task instance.\n",
"title": "Learning to Multi-Task by Active Sampling"
}
| null | null | null | null | true | null |
14898
| null |
Default
| null | null |
null |
{
"abstract": " We prove, under two sufficient conditions, that idealised models can have no\nadversarial examples. We discuss which idealised models satisfy our conditions,\nand show that idealised Bayesian neural networks (BNNs) satisfy these. We\ncontinue by studying near-idealised BNNs using HMC inference, demonstrating the\ntheoretical ideas in practice. We experiment with HMC on synthetic data derived\nfrom MNIST for which we know the ground-truth image density, showing that\nnear-perfect epistemic uncertainty correlates to density under image manifold,\nand that adversarial images lie off the manifold in our setting. This suggests\nwhy MC dropout, which can be seen as performing approximate inference, has been\nobserved to be an effective defence against adversarial examples in practice;\nWe highlight failure-cases of non-idealised BNNs relying on dropout, suggesting\na new attack for dropout models and a new defence as well. Lastly, we\ndemonstrate the defence on a cats-vs-dogs image classification task with a\nVGG13 variant.\n",
"title": "Sufficient Conditions for Idealised Models to Have No Adversarial Examples: a Theoretical and Empirical Study with Bayesian Neural Networks"
}
| null | null | null | null | true | null |
14899
| null |
Default
| null | null |
null |
{
"abstract": " Experimental data availability is a cornerstone for reproducibility in\nexperimental fracture mechanics, which is crucial to the scientific method.\nThis short communication focuses on the accessibility and long term\navailability of raw experimental data. The corresponding authors of the eleven\nmost cited papers, related to experimental fracture mechanics, for every year\nfrom 2000 up to 2016, were kindly asked about the availability of the raw\nexperimental data associated with each publication. For the 187 e-mails sent:\n22.46% resulted in outdated contact information, 57.75% of the authors did\nreceived our request and did not reply, and 19.79 replied to our request. The\navailability of data is generally low with only $11$ available data sets\n(5.9%). The authors identified two main issues for the lacking availability of\nraw experimental data. First, the ability to retrieve data is strongly attached\nto the the possibility to contact the corresponding author. This study suggests\nthat institutional e-mail addresses are insufficient means for obtaining\nexperimental data sets. Second, lack of experimental data is also due that\nsubmission and publication does not require to make the raw experimental data\navailable. The following solutions are proposed: (1) Requirement of unique\nidentifiers, like ORCID or ResearcherID, to detach the author(s) from their\ninstitutional e-mail address, (2) Provide DOIs, like Zenodo or Dataverse, to\nmake raw experimental data citable, and (3) grant providing organizations\nshould ensure that experimental data by public funded projects is available to\nthe public.\n",
"title": "Long term availability of raw experimental data in experimental fracture mechanics"
}
| null | null | null | null | true | null |
14900
| null |
Default
| null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.