polygraf-ai/arxiv-acl-pubmed-abstracts-10k-sample

null

A Study of Syntactic Multi-Modality in Non-Autoregressive Machine Translation

null

It is difficult for non-autoregressive translation (NAT) models to capture the multi-modal distribution of target translations due to their conditional independence assumption, which is known as the {\textquotedblleft}multi-modality problem{\textquotedblright}, including the lexical multi-modality and the syntactic multi-modality. While the first one has been well studied, the syntactic multi-modality brings severe challenges to the standard cross entropy (XE) loss in NAT and is understudied. In this paper, we conduct a systematic study on the syntactic multi-modality problem. Specifically, we decompose it into short- and long-range syntactic multi-modalities and evaluate several recent NAT algorithms with advanced loss functions on both carefully designed synthesized datasets and real datasets. We find that the Connectionist Temporal Classification (CTC) loss and the Order-Agnostic Cross Entropy (OAXE) loss can better handle short- and long-range syntactic multi-modalities respectively. Furthermore, we take the best of both and design a new loss function to better handle the complicated syntactic multi-modality in real-world datasets. To facilitate practical usage, we provide a guide to using different loss functions for different kinds of syntactic multi-modality.

null

acl

inproceedings

zhang-etal-2022-study

Carpuat, Marine and de Marneffe, Marie-Catherine and Meza Ruiz, Ivan Vladimir

jul

2,022

Seattle, United States

Association for Computational Linguistics

https://aclanthology.org/2022.naacl-main.126/

Zhang, Kexun and Wang, Rui and Tan, Xu and Guo, Junliang and Ren, Yi and Qin, Tao and Liu, Tie-Yan

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

1747--1757

null

Template-free synthesis of hierarchical porous metal-organic frameworks.

null

A template-free synthesis of a hierarchical microporous-mesoporous metal-organic framework (MOF) of zinc(II) 2,5-dihydroxy-1,4-benzenedicarboxylate (Zn-MOF-74) is reported. The surface morphology and porosity of the bimodal materials can be modified by etching the pore walls with various synthesis solvents for different reaction times. This template-free strategy enables the preparation of stable frameworks with mesopores exceeding 15 nm, which was previously unattained in the synthesis of MOFs by the ligand-extension method.

null

pubmed

null

2,013

null

23,796,254

Journal of the American Chemical Society

1910.02919

Manan Tomar Mr.

Manan Tomar, Yonathan Efroni, Mohammad Ghavamzadeh

Multi-step Greedy Reinforcement Learning Algorithms

ICML 2020

null

cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Multi-step greedy policies have been extensively used in model-based reinforcement learning (RL), both when a model of the environment is available (e.g.,~in the game of Go) and when it is learned. In this paper, we explore their benefits in model-free RL, when employed using multi-step dynamic programming algorithms: $\kappa$-Policy Iteration ($\kappa$-PI) and $\kappa$-Value Iteration ($\kappa$-VI). These methods iteratively compute the next policy ($\kappa$-PI) and value function ($\kappa$-VI) by solving a surrogate decision problem with a shaped reward and a smaller discount factor. We derive model-free RL algorithms based on $\kappa$-PI and $\kappa$-VI in which the surrogate problem can be solved by any discrete or continuous action RL method, such as DQN and TRPO. We identify the importance of a hyper-parameter that controls the extent to which the surrogate problem is solved and suggest a way to set this parameter. When evaluated on a range of Atari and MuJoCo benchmark tasks, our results indicate that for the right range of $\kappa$, our algorithms outperform DQN and TRPO. This shows that our multi-step greedy algorithms are general enough to be applied over any existing RL algorithm and can significantly improve its performance.

[ { "created": "Mon, 7 Oct 2019 17:20:25 GMT", "version": "v1" }, { "created": "Mon, 14 Oct 2019 17:25:19 GMT", "version": "v2" }, { "created": "Mon, 13 Jul 2020 00:00:32 GMT", "version": "v3" } ]

2020-07-14

[ [ "Tomar", "Manan", "" ], [ "Efroni", "Yonathan", "" ], [ "Ghavamzadeh", "Mohammad", "" ] ]

Mon, 7 Oct 2019 17:20:25 GMT

2019-10-07T17:20:25Z

[ "cs", "stat" ]

arxiv

null

1803.05589

Mohammad Emtiyaz Khan

Wu Lin, Nicolas Hubacher, Mohammad Emtiyaz Khan

Variational Message Passing with Structured Inference Networks

Added a missing term in the gradient of the lower bound

ICLR 2018

null

stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Recent efforts on combining deep models with probabilistic graphical models are promising in providing flexible models that are also easy to interpret. We propose a variational message-passing algorithm for variational inference in such models. We make three contributions. First, we propose structured inference networks that incorporate the structure of the graphical model in the inference network of variational auto-encoders (VAE). Second, we establish conditions under which such inference networks enable fast amortized inference similar to VAE. Finally, we derive a variational message passing algorithm to perform efficient natural-gradient inference while retaining the efficiency of the amortized inference. By simultaneously enabling structured, amortized, and natural-gradient inference for deep structured models, our method simplifies and generalizes existing methods.

[ { "created": "Thu, 15 Mar 2018 04:26:24 GMT", "version": "v1" }, { "created": "Thu, 14 Jun 2018 13:21:32 GMT", "version": "v2" } ]

2018-06-15

[ [ "Lin", "Wu", "" ], [ "Hubacher", "Nicolas", "" ], [ "Khan", "Mohammad Emtiyaz", "" ] ]

Thu, 15 Mar 2018 04:26:24 GMT

2018-03-15T04:26:24Z

[ "stat" ]

arxiv

null

Low-dose drug combinations along molecular pathways could maximize therapeutic effectiveness while minimizing collateral adverse effects.

null

Increasing knowledge of molecular signaling processes has enabled the identification of drug targets that synergistically address multifactorial symptoms along several contributing pathways. The idea behind 'polypills' is that minor doses of pharmacodynamically interacting drugs would selectively achieve intended clinical effects. Analogously, monofactorial symptoms could be addressed vertically along their main pathway. Clinical selectivity follows from successive incomplete inhibitions of the pathological pathway at several steps. Here, we discuss and exemplify of how successive inhibitions in the prostaglandin E2 (PGE(2)) signaling pathway could achieve anti-inflammatory and analgesic effects while preserving physiological PGE(2) signaling in organs of major toxicity. Intentionally using intelligent low-dose drug combinations might provide an innovative therapeutic concept that directs combined small-drug effects towards a large, selective clinical effect with minor collateral damage.

null

pubmed

null

2,011

null

22,016,087

Drug discovery today

1807.03990

Pierre Berard

Pierre B\'erard (IF), Bernard Helffer (LMJL, LM-Orsay)

Sturm's theorem on the zeros of sums of eigenfunctions: Gelfand's strategy implemented

Comments: One subsection deleted. References added. Minor corrections

Moscow Mathematical Journal 20:1 (2020) 1--25 (January-March 2020)

null

math.AP math-ph math.MP math.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In the second section "Courant-Gelfand theorem" of his last published paper (Topological properties of eigenoscillations in mathematical physics, Proc. Steklov Institute Math. 273 (2011) 25--34), Arnold recounts Gelfand's strategy to prove that the zeros of any linear combination of the $n$ first eigenfunctions of the Sturm-Liouville problem $$-\, y"(s) + q(x)\, y(x) = \lambda\, y(x) \mbox{ in } ]0,1[\,, \mbox{ with } y(0)=y(1)=0\,,$$divide the interval into at most $n$ connected components, and concludes that "the lack of a published formal text with a rigorous proof \dots is still distressing." Inspired by Quantum mechanics, Gelfand's strategy consists in replacing the ana\-lysis of linear combinations of the $n$ first eigenfunctions by that of their Slater determinant which is the first eigenfunction of the associated $n$-particle operator acting on Fermions. In the present paper, we implement Gelfand's strategy, and give a complete proof of the above assertion. As a matter of fact, refining Gelfand's strategy, we prove a stronger property taking the multiplicity of zeros into account, a result which actually goes back to Sturm (1836).

[ { "created": "Wed, 11 Jul 2018 08:20:01 GMT", "version": "v1" }, { "created": "Mon, 6 Aug 2018 08:54:04 GMT", "version": "v2" }, { "created": "Wed, 14 Nov 2018 07:25:12 GMT", "version": "v3" } ]

2020-03-12

[ [ "Bérard", "Pierre", "", "IF" ], [ "Helffer", "Bernard", "", "LMJL, LM-Orsay" ] ]

Wed, 11 Jul 2018 08:20:01 GMT

2018-07-11T08:20:01Z

[ "math" ]

arxiv

null

Designing magnetic organic lattices from high-spin polycyclic units.

null

This work addresses the conception of purely organic magnetic materials by properly bridging high-spin polycyclic hydrocarbons A and B, through covalent ligands L. The strategy varies two degrees of freedom that govern the magnetic character of the A-L--B sequence, namely, the bridge response to spin polarization and the relative signs of spin density on carbon atoms to which the bridge is attached. Topological prescriptions based on Ovchinnikov's rule are proposed to predict ground-state spin multiplicities of various A-L-B sets. The relevance of these guiding principles is essentially confirmed through DFT calculations on dimers connected by conjugated bridges. The transferability of interunit magnetic couplings to larger assemblies is further checked, the building blocks tending to maintain their high-spin character whatever the environment. Such local designs open the way to periodic lattices of ferromagnetic, antiferromagnetic, ferrimagnetic, or paramagnetic materials.

null

pubmed

null

2,011

null

22,021,220

Chemphyschem : a European journal of chemical physics and physical chemistry

null

Deep Neural Machine Translation with Weakly-Recurrent Units

null

Recurrent neural networks (RNNs) have represented for years the state of the art in neural machine translation. Recently, new architectures have been proposed, which can leverage parallel computation on GPUs better than classical RNNs. Faster training and inference combined with different sequence-to-sequence modeling also lead to performance improvements. While the new models completely depart from the original recurrent architecture, we decided to investigate how to make RNNs more efficient. In this work, we propose a new recurrent NMT architecture, called Simple Recurrent NMT, built on a class of fast and weakly-recurrent units that use layer normalization and multiple attentions. Our experiments on the WMT14 English-to-German and WMT16 English-Romanian benchmarks show that our model represents a valid alternative to LSTMs, as it can achieve better results at a significantly lower computational cost.

null

acl

inproceedings

di-gangi-federico-2018-deep

P{\'e}rez-Ortiz, Juan Antonio and S{\'a}nchez-Mart{\'i}nez, Felipe and Espl{\`a}-Gomis, Miquel and Popovi{\'c}, Maja and Rico, Celia and Martins, Andr{\'e} and Van den Bogaert, Joachim and Forcada, Mikel L.

may

2,018

Alicante, Spain

null

https://aclanthology.org/2018.eamt-main.12/

Di Gangi, Mattia A. and Federico, Marcello

Proceedings of the 21st Annual Conference of the European Association for Machine Translation

139--148

null

1412.1862

Paul Egr\'e

Paul \'Egr\'e and Paul Marty and Bryan Renne

Knowledge, Justification, and Adequate Reasons

v3 edits acknowledgments

The Review of Symbolic Logic, 14(3), 687-727. 2021

null

cs.LO cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Is knowledge definable as justified true belief ("JTB")? We argue that one can legitimately answer positively or negatively, depending on whether or not one's true belief is justified by what we call adequate reasons. To facilitate our argument we introduce a simple propositional logic of reason-based belief, and give an axiomatic characterization of the notion of adequacy for reasons. We show that this logic is sufficiently flexible to accommodate various useful features, including quantification over reasons. We use our framework to contrast two notions of JTB: one internalist, the other externalist. We argue that Gettier cases essentially challenge the internalist notion but not the externalist one. Our approach commits us to a form of infallibilism about knowledge, but it also leaves us with a puzzle, namely whether knowledge involves the possession of only adequate reasons, or leaves room for some inadequate reasons. We favor the latter position, which reflects a milder and more realistic version of infallibilism.

[ { "created": "Thu, 4 Dec 2014 23:21:12 GMT", "version": "v1" }, { "created": "Tue, 20 Jan 2015 05:05:05 GMT", "version": "v2" }, { "created": "Wed, 20 May 2015 13:54:36 GMT", "version": "v3" }, { "created": "Sun, 26 Dec 2021 17:52:28 GMT", "version": "v4" } ]

2021-12-28

[ [ "Égré", "Paul", "" ], [ "Marty", "Paul", "" ], [ "Renne", "Bryan", "" ] ]

Thu, 4 Dec 2014 23:21:12 GMT

2014-12-04T23:21:12Z

[ "cs" ]

arxiv

null

1312.5412

Taichi Kiwaki Mr

Taichi Kiwaki, Takaki Makino, Kazuyuki Aihara

Approximated Infomax Early Stopping: Revisiting Gaussian RBMs on Natural Images

9 pages with 1 page appendix

null

stat.ML cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We pursue an early stopping technique that helps Gaussian Restricted Boltzmann Machines (GRBMs) to gain good natural image representations in terms of overcompleteness and data fitting. GRBMs are widely considered as an unsuitable model for natural images because they gain non-overcomplete representations which include uniform filters that do not represent useful image features. We have recently found that GRBMs once gain and subsequently lose useful filters during their training, contrary to this common perspective. We attribute this phenomenon to a tradeoff between overcompleteness of GRBM representations and data fitting. To gain GRBM representations that are overcomplete and fit data well, we propose a measure for GRBM representation quality, approximated mutual information, and an early stopping technique based on this measure. The proposed method boosts performance of classifiers trained on GRBM representations.

[ { "created": "Thu, 19 Dec 2013 05:37:50 GMT", "version": "v1" }, { "created": "Thu, 2 Jan 2014 16:45:15 GMT", "version": "v2" }, { "created": "Mon, 6 Jan 2014 20:07:09 GMT", "version": "v3" } ]

2014-01-07

[ [ "Kiwaki", "Taichi", "" ], [ "Makino", "Takaki", "" ], [ "Aihara", "Kazuyuki", "" ] ]

Thu, 19 Dec 2013 05:37:50 GMT

2013-12-19T05:37:50Z

[ "cs", "stat" ]

arxiv

null

1809.06018

Xi Zhang

Xi Sheryl Zhang, Jingyuan Chou, Fei Wang

Integrative Analysis of Patient Health Records and Neuroimages via Memory-based Graph Convolutional Network

null

cs.LG stat.ML

http://creativecommons.org/licenses/by/4.0/

With the arrival of the big data era, more and more data are becoming readily available in various real-world applications and those data are usually highly heterogeneous. Taking computational medicine as an example, we have both Electronic Health Records (EHR) and medical images for each patient. For complicated diseases such as Parkinson's and Alzheimer's, both EHR and neuroimaging information are very important for disease understanding because they contain complementary aspects of the disease. However, EHR and neuroimage are completely different. So far the existing research has been mainly focusing on one of them. In this paper, we proposed a framework, Memory-Based Graph Convolution Network (MemGCN), to perform integrative analysis with such multi-modal data. Specifically, GCN is used to extract useful information from the patients' neuroimages. The information contained in the patient EHRs before the acquisition of each brain image is captured by a memory network because of its sequential nature. The information contained in each brain image is combined with the information read out from the memory network to infer the disease state at the image acquisition timestamp. To further enhance the analytical power of MemGCN, we also designed a multi-hop strategy that allows multiple reading and updating on the memory can be performed at each iteration. We conduct experiments using the patient data from the Parkinson's Progression Markers Initiative (PPMI) with the task of classification of Parkinson's Disease (PD) cases versus controls. We demonstrate that superior classification performance can be achieved with our proposed framework, comparing with existing approaches involving a single type of data.

[ { "created": "Mon, 17 Sep 2018 04:45:39 GMT", "version": "v1" }, { "created": "Sun, 7 Oct 2018 14:02:12 GMT", "version": "v2" }, { "created": "Sun, 17 Mar 2019 06:06:25 GMT", "version": "v3" }, { "created": "Tue, 7 May 2019 04:11:54 GMT", "version": "v4" } ]

2019-05-08

[ [ "Zhang", "Xi Sheryl", "" ], [ "Chou", "Jingyuan", "" ], [ "Wang", "Fei", "" ] ]

Mon, 17 Sep 2018 04:45:39 GMT

2018-09-17T04:45:39Z

[ "cs", "stat" ]

arxiv

null

2202.09065

Bikram Banerjee

Bikram Banerjee, Goutam Mukherjee

A note on cohomology of Clifford algebras

Comments are welcome

null

math.AT

http://creativecommons.org/licenses/by/4.0/

In this article we construct a cochain complex of a complex Clifford algebra with coefficients in itself in a combinatorial fashion and we call the corresponding cohomology by {\it Clifford cohomology.} We show that {\it Clifford cohomology} controls the deformation of a complex Clifford algebra and can classify them up to Morita equivalence. We also study Hochschild cohomology groups and formal deformations of the algebra of smooth sections of a complex Clifford algebra bundle over an even dimensional orientable Riemannian manifold $M$ which admits a $Spin^{c}$ structure.

[ { "created": "Fri, 18 Feb 2022 08:11:49 GMT", "version": "v1" }, { "created": "Sat, 30 Jul 2022 07:18:50 GMT", "version": "v2" }, { "created": "Fri, 16 Dec 2022 03:36:14 GMT", "version": "v3" } ]

2022-12-19

[ [ "Banerjee", "Bikram", "" ], [ "Mukherjee", "Goutam", "" ] ]

Fri, 18 Feb 2022 08:11:49 GMT

2022-02-18T08:11:49Z

[ "math" ]

arxiv

null

2204.00936

Sergei Ivanov

Sergei Ivanov, Daniil Mamaev, Anya Nordskova

Banach's isometric subspace problem in dimension four

25 pages, v2: minor corrections and text improvements

Invent. math. 233, 1393-1425 (2023)

10.1007/s00222-023-01197-2

null

math.MG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We prove that if all intersections of a convex body $B\subset\mathbb R^4$ with 3-dimensional linear subspaces are linearly equivalent then $B$ is a centered ellipsoid. This gives an affirmative answer to the case $n=3$ of the following question by Banach from 1932: Is a normed vector space $V$ whose $n$-dimensional linear subspaces are all isometric, for a fixed $2 \le n< \dim V$, necessarily Euclidean? The dimensions $n=3$ and $\dim V=4$ is the first case where the question was unresolved. Since the $3$-sphere is parallelizable, known global topological methods do not help in this case. Our proof employs a differential geometric approach.

[ { "created": "Sat, 2 Apr 2022 20:15:51 GMT", "version": "v1" }, { "created": "Sun, 26 Nov 2023 14:46:46 GMT", "version": "v2" } ]

2023-11-28

[ [ "Ivanov", "Sergei", "" ], [ "Mamaev", "Daniil", "" ], [ "Nordskova", "Anya", "" ] ]

Sat, 2 Apr 2022 20:15:51 GMT

2022-04-02T20:15:51Z

[ "math" ]

arxiv

null

1701.06586

Andre Lessa

Federico Ambrogi, Sabine Kraml, Suchita Kulkarni, Ursula Laa, Andre Lessa, Veronika Magerl, Jory Sonneveld, Michael Traub, Wolfgang Waltenberger

SModelS v1.1 user manual

The manual is compatible with SModelS version 1.1.1. v4: Version to appear in Computer Physics Communications

Comput. Phys. Comm. 227 (2018) 72-98

10.1016/j.cpc.2018.02.007

LPSC17001, HEPHY-PUB 981/17

hep-ph hep-ex

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

SModelS is an automatised tool for the interpretation of simplified model results from the LHC. It allows to decompose models of new physics obeying a Z2 symmetry into simplified model components, and to compare these against a large database of experimental results. The first release of SModelS, v1.0, used only cross section upper limit maps provided by the experimental collaborations. In this new release, v1.1, we extend the functionality of SModelS to efficiency maps. This increases the constraining power of the software, as efficiency maps allow to combine contributions to the same signal region from different simplified models. Other new features of version 1.1 include likelihood and chi-square calculations, extended information on the topology coverage, an extended database of experimental results as well as major speed upgrades for both the code and the database. We describe in detail the concepts and procedures used in SModelS, explaining in particular how upper limits and efficiency map results are dealt with in parallel. Detailed instructions for code usage are also provided.

[ { "created": "Mon, 23 Jan 2017 19:01:12 GMT", "version": "v1" }, { "created": "Fri, 24 Feb 2017 20:34:47 GMT", "version": "v2" }, { "created": "Tue, 27 Jun 2017 21:44:48 GMT", "version": "v3" }, { "created": "Wed, 7 Feb 2018 16:32:56 GMT", "version": "v4" } ]

2018-03-19

[ [ "Ambrogi", "Federico", "" ], [ "Kraml", "Sabine", "" ], [ "Kulkarni", "Suchita", "" ], [ "Laa", "Ursula", "" ], [ "Lessa", "Andre", "" ], [ "Magerl", "Veronika", "" ], [ "Sonneveld", "Jory", "" ], [ "Traub", "Michael", "" ], [ "Waltenberger", "Wolfgang", "" ] ]

Mon, 23 Jan 2017 19:01:12 GMT

2017-01-23T19:01:12Z

[ "physics" ]

arxiv

null

Human macrophage responses to clinical isolates from the Mycobacterium tuberculosis complex discriminate between ancient and modern lineages.

null

The aim of the present study was to determine whether there is a correlation between phylogenetic relationship and inflammatory response amongst a panel of clinical isolates representative of the global diversity of the human Mycobacterium tuberculosis Complex (MTBC). Measurement of cytokines from infected human peripheral blood monocyte-derived macrophages revealed a wide variation in the response to different strains. The same pattern of high or low response to individual strains was observed for different pro-inflammatory cytokines and chemokines, and was conserved across multiple human donors. Although each major phylogenetic lineage of MTBC included strains inducing a range of cytokine responses, we found that overall inflammatory phenotypes differed significantly across lineages. In particular, comparison of evolutionarily modern lineages demonstrated a significant skewing towards lower early inflammatory response. The differential response to ancient and modern lineages observed using GM-CSF derived macrophages was also observed in autologous monocyte-derived dendritic cells and murine bone marrow-derived macrophages, but not in human unfractionated peripheral blood mononuclear cells. We hypothesize that the reduced immune responses to modern lineages contribute to more rapid disease progression and transmission, which might be a selective advantage in the context of expanding human populations. In addition to the lineage effects, the large strain-to-strain variation in innate immune responses elicited by MTBC will need to be considered in tuberculosis vaccine development.

null

pubmed

null

2,011

null

21,408,618

PLoS pathogens

physics/0607268

Miquel Montero

Miquel Montero and Jaume Masoliver

Mean Exit Time and Survival Probability within the CTRW Formalism

10 pages, 2 figures, revtex4; corrected typos, to appear in the APFA5 proceedings

Eur. Phys. J. B 57, 181-185 (2007)

10.1140/epjb/e2007-00128-1

null

physics.soc-ph q-fin.ST

null

An intense research on financial market microstructure is presently in progress. Continuous time random walks (CTRWs) are general models capable to capture the small-scale properties that high frequency data series show. The use of CTRW models in the analysis of financial problems is quite recent and their potentials have not been fully developed. Here we present two (closely related) applications of great interest in risk control. In the first place, we will review the problem of modelling the behaviour of the mean exit time (MET) of a process out of a given region of fixed size. The surveyed stochastic processes are the cumulative returns of asset prices. The link between the value of the MET and the timescale of the market fluctuations of a certain degree is crystal clear. In this sense, MET value may help, for instance, in deciding the optimal time horizon for the investment. The MET is, however, one among the statistics of a distribution of bigger interest: the survival probability (SP), the likelihood that after some lapse of time a process remains inside the given region without having crossed its boundaries. The final part of the article is devoted to the study of this quantity. Note that the use of SPs may outperform the standard "Value at Risk" (VaR) method for two reasons: we can consider other market dynamics than the limited Wiener process and, even in this case, a risk level derived from the SP will ensure (within the desired quintile) that the quoted value of the portfolio will not leave the safety zone. We present some preliminary theoretical and applied results concerning this topic.

[ { "created": "Fri, 28 Jul 2006 17:40:55 GMT", "version": "v1" }, { "created": "Thu, 26 Oct 2006 13:07:08 GMT", "version": "v2" } ]

2008-12-02

[ [ "Montero", "Miquel", "" ], [ "Masoliver", "Jaume", "" ] ]

Fri, 28 Jul 2006 17:40:55 GMT

2006-07-28T17:40:55Z

[ "physics", "q-fin" ]

arxiv

null

Regenerative response following stab injury in the adult zebrafish telencephalon.

null

In contrast to mammals, the brain of the adult zebrafish has a remarkable ability to regenerate. In mammals, injuries induce proliferation of astrocytes and oligodendrocyte progenitors contributing to the formation of a glial scar. We analyzed the proliferation of glial cells and microglia in response to stab injury in the adult zebrafish telencephalon: Radial glial markers were up-regulated at the ventricle and co-expressed the proliferation nuclear antigen (PCNA). Microglia and oligodendrocyte progenitors accumulated transiently at the site of lesion. However, we could not find evidence of permanent scar formation. Parenchymal proliferation was almost negligible in comparison to the increase in proliferation at the ventricular zone. This suggests that most of the cellular material for regeneration is derived from regions of constitutive neurogenesis. Remarkably, the proliferative response is almost completely restricted to the lesioned hemisphere indicating that signals inducing regeneration remain mainly confined within the lesioned half of the telencephalon.

null

pubmed

null

2,011

null

22,016,188

Developmental dynamics : an official publication of the American Association of Anatomists

1909.09436

Miltiadis Allamanis

Hamel Husain, Ho-Hsiang Wu, Tiferet Gazit, Miltiadis Allamanis, Marc Brockschmidt

CodeSearchNet Challenge: Evaluating the State of Semantic Code Search

Updated evaluation numbers after fixing indexing bug

null

cs.LG cs.IR cs.SE stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Semantic code search is the task of retrieving relevant code given a natural language query. While related to other information retrieval tasks, it requires bridging the gap between the language used in code (often abbreviated and highly technical) and natural language more suitable to describe vague concepts and ideas. To enable evaluation of progress on code search, we are releasing the CodeSearchNet Corpus and are presenting the CodeSearchNet Challenge, which consists of 99 natural language queries with about 4k expert relevance annotations of likely results from CodeSearchNet Corpus. The corpus contains about 6 million functions from open-source code spanning six programming languages (Go, Java, JavaScript, PHP, Python, and Ruby). The CodeSearchNet Corpus also contains automatically generated query-like natural language for 2 million functions, obtained from mechanically scraping and preprocessing associated function documentation. In this article, we describe the methodology used to obtain the corpus and expert labels, as well as a number of simple baseline solutions for the task. We hope that CodeSearchNet Challenge encourages researchers and practitioners to study this interesting task further and will host a competition and leaderboard to track the progress on the challenge. We are also keen on extending CodeSearchNet Challenge to more queries and programming languages in the future.

[ { "created": "Fri, 20 Sep 2019 11:52:45 GMT", "version": "v1" }, { "created": "Fri, 27 Sep 2019 09:21:21 GMT", "version": "v2" }, { "created": "Mon, 8 Jun 2020 09:09:28 GMT", "version": "v3" } ]

2020-06-09

[ [ "Husain", "Hamel", "" ], [ "Wu", "Ho-Hsiang", "" ], [ "Gazit", "Tiferet", "" ], [ "Allamanis", "Miltiadis", "" ], [ "Brockschmidt", "Marc", "" ] ]

Fri, 20 Sep 2019 11:52:45 GMT

2019-09-20T11:52:45Z

[ "cs", "stat" ]

arxiv

null

Metabolic profiling of Chinese tobacco leaf of different geographical origins by GC-MS.

null

Tobacco leaf obtained from different geographical areas in China was profiled using gas chromatography-mass spectrometry (GC-MS) coupled with multivariate data analyses. Hierarchical cluster analysis (HCA) and principal component analysis (PCA) showed that the tobacco metabolome was clearly dependent on geographical origins; climatic conditions, such as temperature and precipitation, imposed a greater impact on metabolite levels than the cultivars. By orthogonal partial least-squares-discrimination analysis (OPLS-DA), 20 metabolites that contributed to the discrimination were screened, including primary metabolites (sucrose, D-fructose, D-mannose, D-glucose, inositol, maleic acid, citric acid, malic acid, L-threonic acid, L-proline, L-phenylalanine), secondary metabolites (chlorogenic acid, α- and β-4,8,13-duvatriene-1,3-diol, nicotine, quinic acid), and four unknown metabolites. The results suggest that metabolic profiling using GC-MS combined with multivariate analysis can be used to discriminate tobacco leaf of different geographical origins and to provide potential indicators of tobacco origins.

null

pubmed

null

2,013

null

23,441,877

Journal of agricultural and food chemistry

2005.03889

Yong Xu Dr

Yong Xu, Meng Yu, Shi-Xiong Zhang, Lianwu Chen, Chao Weng, Jianming Liu, Dong Yu

Neural Spatio-Temporal Beamformer for Target Speech Separation

accepted to Interspeech2020, Demo: https://yongxuustc.github.io/mtmvdr/

null

eess.AS cs.SD

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Purely neural network (NN) based speech separation and enhancement methods, although can achieve good objective scores, inevitably cause nonlinear speech distortions that are harmful for the automatic speech recognition (ASR). On the other hand, the minimum variance distortionless response (MVDR) beamformer with NN-predicted masks, although can significantly reduce speech distortions, has limited noise reduction capability. In this paper, we propose a multi-tap MVDR beamformer with complex-valued masks for speech separation and enhancement. Compared to the state-of-the-art NN-mask based MVDR beamformer, the multi-tap MVDR beamformer exploits the inter-frame correlation in addition to the inter-microphone correlation that is already utilized in prior arts. Further improvements include the replacement of the real-valued masks with the complex-valued masks and the joint training of the complex-mask NN. The evaluation on our multi-modal multi-channel target speech separation and enhancement platform demonstrates that our proposed multi-tap MVDR beamformer improves both the ASR accuracy and the perceptual speech quality against prior arts.

[ { "created": "Fri, 8 May 2020 07:53:37 GMT", "version": "v1" }, { "created": "Mon, 11 May 2020 09:33:08 GMT", "version": "v2" }, { "created": "Thu, 14 May 2020 08:26:55 GMT", "version": "v3" }, { "created": "Mon, 18 May 2020 05:06:01 GMT", "version": "v4" }, { "created": "Fri, 31 Jul 2020 08:19:07 GMT", "version": "v5" } ]

2020-08-03

[ [ "Xu", "Yong", "" ], [ "Yu", "Meng", "" ], [ "Zhang", "Shi-Xiong", "" ], [ "Chen", "Lianwu", "" ], [ "Weng", "Chao", "" ], [ "Liu", "Jianming", "" ], [ "Yu", "Dong", "" ] ]

Fri, 8 May 2020 07:53:37 GMT

2020-05-08T07:53:37Z

[ "cs", "eess" ]

arxiv

null

The new Ghent criteria for Marfan syndrome: what do they change?

null

The diagnosis of Marfan syndrome (MFS) is challenging and international criteria have been proposed. The 1996 Ghent criteria were adopted worldwide, but new diagnostic criteria for MFS were released in 2010, giving more weight to aortic root aneurysm and ectopia lentis. We aimed to compare the diagnosis reached by applying this new nosology vs the Ghent nosology in a well-known series of 1009 probands defined by the presence of an FBN1 mutation. A total of 842 patients could be classified as MFS according to the new nosology (83%) as compared to 894 (89%) according to the 1996 Ghent criteria. The remaining 17% would be classified as ectopia lentis syndrome (ELS), mitral valve prolapse syndrome or mitral valve, aorta, skeleton and skin (MASS) syndrome, or potential MFS in patients aged less than 20 years. Taking into account the median age at last follow-up (29 years), the possibility has to be considered that these patients would go on to develop classic MFS with time. Although the number of patients for a given diagnosis differed only slightly, the new nosology led to a different diagnosis in 15% of cases. Indeed, 10% of MFS patients were reclassified as ELS or MASS in the absence of aortic dilatation; conversely, 5% were reclassified as MFS in the presence of aortic dilatation. The nosology is easier to apply because the systemic score is helpful to reach the diagnosis of MFS only in a minority of patients. Diagnostic criteria should be a flexible and dynamic tool so that reclassification of patients with alternative diagnosis is possible, requiring regular clinical and aortic follow-up.

null

pubmed

null

2,012

null

21,564,093

Clinical genetics

1812.10890

Keiho Matsumoto

Gysin triangles in the category of motifs with modulus

null

10.1017/S1474748021000554

null

math.AG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we study a Gysin triangle in the category of motives with modulus. We can understand this Gysin triangle as a motivic lift of the Gysin triangle of log-crystalline cohomology due to Nakkajima and Shiho. After that we compare motives with modulus and Voevodsky motives. The corollary implies that an object in the category of motives with modulus decomposes into a $p$-torsion part and a Voevodsky motive part. We can understand the corollary as a motivic analogue of the relationship between rigid cohomology and log-crystalline cohomology.

[ { "created": "Fri, 28 Dec 2018 04:33:28 GMT", "version": "v1" }, { "created": "Sun, 22 Mar 2020 12:06:41 GMT", "version": "v2" }, { "created": "Wed, 7 Oct 2020 12:18:00 GMT", "version": "v3" }, { "created": "Sun, 24 Oct 2021 14:17:12 GMT", "version": "v4" } ]

2023-06-22

[ [ "Matsumoto", "Keiho", "" ] ]

Fri, 28 Dec 2018 04:33:28 GMT

2018-12-28T04:33:28Z

[ "math" ]

arxiv

null

1404.1607

Charilaos Tsagkarakis Ph.D

F.K. Diakonos, G.C. Katsimiga, X.N. Maintas and C.E. Tsagkarakis

Symmetric solitonic excitations of the (1+1)-dimensional Abelian-Higgs "classical vacuum"

8 pages, 5 figures

Phys. Rev. E 91, 023202 (2015)

10.1103/PhysRevE.91.023202

null

hep-th cond-mat.supr-con nlin.PS

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the classical dynamics of the Abelian-Higgs model in (1+1) space-time dimensions bf for the case of strongly broken gauge symmetry. In this limit the wells of the potential are almost harmonic and sufficiently deep, presenting a scenario far from the associated critical point. Using a multiscale perturbation expansion, the equations of motion for the fields are reduced to a system of coupled nonlinear Schrodinger equations (CNLS). Exact solutions of the latter are used to obtain approximate analytical solutions for the full dynamics of both the gauge and Higgs field in the form of oscillons and oscillating kinks. Numerical simulations of the exact dynamics verify the validity of these solutions. We explore their persistence for a wide range of the model's single parameter which is the ratio of the Higgs mass to the gauge field mass . We show that only oscillons oscillating symmetrically with respect to the "classical vacuum", for both the gauge and the Higgs field, are long lived. Furthermore plane waves and oscillating kinks are shown to decay into oscillon- like patterns, due to the modulation instability mechanism.

[ { "created": "Sun, 6 Apr 2014 19:01:38 GMT", "version": "v1" }, { "created": "Tue, 8 Apr 2014 11:11:17 GMT", "version": "v2" }, { "created": "Thu, 12 Jun 2014 09:15:24 GMT", "version": "v3" }, { "created": "Fri, 28 Nov 2014 19:58:53 GMT", "version": "v4" }, { "created": "Wed, 21 Jan 2015 12:12:54 GMT", "version": "v5" } ]

2015-03-05

[ [ "Diakonos", "F. K.", "" ], [ "Katsimiga", "G. C.", "" ], [ "Maintas", "X. N.", "" ], [ "Tsagkarakis", "C. E.", "" ] ]

Sun, 6 Apr 2014 19:01:38 GMT

2014-04-06T19:01:38Z

[ "physics", "nlin" ]

arxiv

null

1603.04324

Louis-Philippe Thibault

Preprojective algebra structure on skew-group algebras

33 pages. Several modifications and additions were made in order to add precision and clarity

null

math.RT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We give a class of finite subgroups $G<SL(n, k)$ for which the skew-group algebra $k[x_1,\ldots, x_n]\#G$ does not admit the grading structure of a higher preprojective algebra. Namely, we prove that if a finite group $G<SL(n, k)$ is conjugate to a finite subgroup of $SL(n_1, k)\times SL(n_2, k)$, for some $n_1, n_2\geq 1$, then the skew-group algebra $k[x_1,\ldots,x_n]\#G$ is not Morita equivalent to a higher preprojective algebra. This is related to the preprojective algebra structure on the tensor product of two Koszul bimodule Calabi-Yau algebras. We prove that such an algebra cannot be endowed with a grading structure as required for a higher preprojective algebra. Moreover, we construct explicitly the bound quiver of the higher preprojective algebra over a finite-dimensional Koszul algebra of finite global dimension. We show in addition that preprojective algebras over higher representation-infinite Koszul algebras are derivation-quotient algebras whose relations are given by a superpotential.

[ { "created": "Mon, 14 Mar 2016 16:21:14 GMT", "version": "v1" }, { "created": "Tue, 22 Mar 2016 17:10:07 GMT", "version": "v2" }, { "created": "Tue, 3 Dec 2019 12:56:01 GMT", "version": "v3" } ]

2019-12-04

[ [ "Thibault", "Louis-Philippe", "" ] ]

Mon, 14 Mar 2016 16:21:14 GMT

2016-03-14T16:21:14Z

[ "math" ]

arxiv

null

1806.00033

Livio Liechti

Livio Liechti and Bal\'azs Strenner

Minimal pseudo-Anosov stretch factors on nonoriented surfaces

26 pages, 6 figures

Algebr. Geom. Topol. 20 (2020) 451-485

10.2140/agt.2020.20.451

null

math.GT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We determine the smallest stretch factor among pseudo-Anosov maps with an orientable invariant foliation on the closed nonorientable surfaces of genus 4, 5, 6, 7, 8, 10, 12, 14, 16, 18 and 20. We also determine the smallest stretch factor of an orientation-reversing pseudo-Anosov map with orientable invariant foliations on the closed orientable surfaces of genus 1, 3, 5, 7, 9 and 11. As a byproduct, we obtain that the stretch factor of a pseudo-Anosov map on a nonorientable surface or an orientation-reversing pseudo-Anosov map on an orientable surface does not have Galois conjugates on the unit circle. This shows that the techniques that were used to disprove Penner's conjecture on orientable surfaces are ineffective in the nonorientable cases.

[ { "created": "Thu, 31 May 2018 18:12:28 GMT", "version": "v1" }, { "created": "Thu, 26 Mar 2020 17:22:54 GMT", "version": "v2" } ]

2020-04-01

[ [ "Liechti", "Livio", "" ], [ "Strenner", "Balázs", "" ] ]

Thu, 31 May 2018 18:12:28 GMT

2018-05-31T18:12:28Z

[ "math" ]

arxiv

null

1009.4798

Andrea Baronchelli

Role of feedback and broadcasting in the naming game

7 pages, 6 figures

Phys. Rev. E 83, 046103 (2011)

10.1103/PhysRevE.83.046103

null

physics.soc-ph cond-mat.stat-mech cs.GT cs.MA cs.NI q-bio.PE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The naming game (NG) describes the agreement dynamics of a population of agents that interact locally in a pairwise fashion, and in recent years statistical physics tools and techniques have greatly contributed to shed light on its rich phenomenology. Here we investigate in details the role played by the way in which the two agents update their states after an interaction. We show that slightly modifying the NG rules in terms of which agent performs the update in given circumstances (i.e. after a success) can either alter dramatically the overall dynamics or leave it qualitatively unchanged. We understand analytically the first case by casting the model in the broader framework of a generalized NG. As for the second case, on the other hand, we note that the modified rule reproducing the main features of the usual NG corresponds in fact to a simplification of it consisting in the elimination of feedback between the agents. This allows us to introduce and study a very natural broadcasting scheme on networks that can be potentially relevant for different applications, such as the design and implementation of autonomous sensor networks, as pointed out in the recent literature.

[ { "created": "Fri, 24 Sep 2010 10:23:26 GMT", "version": "v1" }, { "created": "Tue, 12 Apr 2011 10:05:34 GMT", "version": "v2" } ]

2011-04-13

[ [ "Baronchelli", "Andrea", "" ] ]

Fri, 24 Sep 2010 10:23:26 GMT

2010-09-24T10:23:26Z

[ "q-bio", "physics", "cs" ]

arxiv

null

1903.03229

John Feser

John K. Feser, Samuel Madden, Nan Tang, Armando Solar-Lezama

Deductive Optimization of Relational Data Storage

null

cs.PL cs.DB

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Optimizing the physical data storage and retrieval of data are two key database management problems. In this paper, we propose a language that can express a wide range of physical database layouts, going well beyond the row- and column-based methods that are widely used in database management systems. We use deductive synthesis to turn a high-level relational representation of a database query into a highly optimized low-level implementation which operates on a specialized layout of the dataset. We build a compiler for this language and conduct experiments using a popular database benchmark, which shows that the performance of these specialized queries is competitive with a state-of-the-art in memory compiled database system.

[ { "created": "Fri, 8 Mar 2019 00:40:21 GMT", "version": "v1" }, { "created": "Sun, 14 Jul 2019 10:30:29 GMT", "version": "v2" }, { "created": "Wed, 5 Feb 2020 22:07:28 GMT", "version": "v3" } ]

2020-02-07

[ [ "Feser", "John K.", "" ], [ "Madden", "Samuel", "" ], [ "Tang", "Nan", "" ], [ "Solar-Lezama", "Armando", "" ] ]

Fri, 8 Mar 2019 00:40:21 GMT

2019-03-08T00:40:21Z

[ "cs" ]

arxiv

null

On the Role of Pre-trained Language Models in Word Ordering: A Case Study with {BART}

null

Word ordering is a constrained language generation task taking unordered words as input. Existing work uses linear models and neural networks for the task, yet pre-trained language models have not been studied in word ordering, let alone why they help. We use BART as an instance and show its effectiveness in the task. To explain why BART helps word ordering, we extend analysis with probing and empirically identify that syntactic dependency knowledge in BART is a reliable explanation. We also report performance gains with BART in the related partial tree linearization task, which readily extends our analysis.

null

acl

inproceedings

ou-etal-2022-role

Calzolari, Nicoletta and Huang, Chu-Ren and Kim, Hansaem and Pustejovsky, James and Wanner, Leo and Choi, Key-Sun and Ryu, Pum-Mo and Chen, Hsin-Hsi and Donatelli, Lucia and Ji, Heng and Kurohashi, Sadao and Paggio, Patrizia and Xue, Nianwen and Kim, Seokhwan and Hahm, Younggyun and He, Zhong and Lee, Tony Kyungil and Santus, Enrico and Bond, Francis and Na, Seung-Hoon

oct

2,022

Gyeongju, Republic of Korea

International Committee on Computational Linguistics

https://aclanthology.org/2022.coling-1.567/

Ou, Zebin and Zhang, Meishan and Zhang, Yue

Proceedings of the 29th International Conference on Computational Linguistics

6516--6529

null

1209.2064

Tyler J. Jarvis

Dan Edidin, Tyler J. Jarvis, and Takashi Kimura

Chern Classes and Compatible Power Operations in Inertial K-theory

Many improvements. Special thanks to the referee for helpful suggestions. To appear in Annals of K-Theory. arXiv admin note: text overlap with arXiv:1202.0603

AKT 2 (2017) 73-130

10.2140/akt.2017.2.73

null

math.AG math.AT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Let [X/G] be a smooth Deligne-Mumford quotient stack. In a previous paper the authors constructed a class of exotic products called inertial products on K(I[X/G]), the Grothendieck group of vector bundles on the inertia stack I[X/G]. In this paper we develop a theory of Chern classes and compatible power operations for inertial products. When G is diagonalizable these give rise to an augmented $\lambda$-ring structure on inertial K-theory. One well-known inertial product is the virtual product. Our results show that for toric Deligne-Mumford stacks there is a $\lambda$-ring structure on inertial K-theory. As an example, we compute the $\lambda$-ring structure on the virtual K-theory of the weighted projective lines P(1,2) and P(1,3). We prove that after tensoring with C, the augmentation completion of this $\lambda$-ring is isomorphic as a $\lambda$-ring to the classical K-theory of the crepant resolutions of singularities of the coarse moduli spaces of the cotangent bundles $T^*P(1,2)$ and $T^*P(1,3)$, respectively. We interpret this as a manifestation of mirror symmetry in the spirit of the Hyper-Kaehler Resolution Conjecture.

[ { "created": "Mon, 10 Sep 2012 17:09:24 GMT", "version": "v1" }, { "created": "Sat, 6 Jun 2015 14:59:09 GMT", "version": "v2" }, { "created": "Mon, 26 Oct 2015 17:21:56 GMT", "version": "v3" } ]

2016-11-23

[ [ "Edidin", "Dan", "" ], [ "Jarvis", "Tyler J.", "" ], [ "Kimura", "Takashi", "" ] ]

Mon, 10 Sep 2012 17:09:24 GMT

2012-09-10T17:09:24Z

[ "math" ]

arxiv

null

A pilot validation of CFD model results against PIV observations of haemodynamics in intracranial aneurysms treated with flow-diverting stents.

null

Flow-diverting (FD) stents are one of three common modes of treating intracranial aneurysms, yet knowledge of their effect on haemodynamics is incomplete. We used particle image velocimetry (PIV) to measure spatially-varying velocity of blood-analogue fluid within a patient-specific aneurysm model, and compared the observed flow behaviour to predictions from a computational fluid dynamics (CFD) model. In PIV experiments we characterised the flow on multiple cross-sections for three different arterial flowrates (150, 250, 400 mL/min) after deployment of a commercially-available FD stent. Our flow-diverting (FD) stent model for CFD simulation was constructed using a permeability adapted from the literature. Aneurysmal haemodynamics without the FD stent treatment provided good similarities between CFD and PIV results, and the results with a Silk stent treatment also provided acceptable concordances, thereby validating the use of CFD as a convenient and flexible tool for investigating intra-aneurysmal flow dynamics after FD stent treatment. Furthermore, for the first time, the porous-medium FD model stent was validated to be both efficient and effective to predict the flow-diversion effects of a FD stent treatment with a patient-specific intracranial aneurysm. Through the qualitative and quantitative comparison of CFD predictions against the experimental outcomes, this study gives confidence for future studies on aneurysmal haemodynamics and FD stent treatment effects to use CFD simulation.

null

pubmed

null

2,020

null

31,902,608

Journal of biomechanics

1611.01553

Vedad Hadzic

Roderick Bloem, Nicolas Braud-Santoni and Vedad Hadzic

QBF Solving by Counterexample-guided Expansion

This is a **very** old version of the paper arXiv:1807.08964 and should be taken down. I did not know you could just replace papers, and did not know whether I could change authors and similar, so that is why I made a different submission. Please take it down

null

cs.LO cs.AI

http://creativecommons.org/publicdomain/zero/1.0/

We introduce a novel generalization of Counterexample-Guided Inductive Synthesis (CEGIS) and instantiate it to yield a novel, competitive algorithm for solving Quantified Boolean Formulas (QBF). Current QBF solvers based on counterexample-guided expansion use a recursive approach which scales poorly with the number of quantifier alternations. Our generalization of CEGIS removes the need for this recursive approach, and we instantiate it to yield a simple and efficient algorithm for QBF solving. Lastly, this research is supported by a competitive, though straightforward, implementation of the algorithm, making it possible to study the practical impact of our algorithm design decisions, along with various optimizations.

[ { "created": "Fri, 4 Nov 2016 22:08:40 GMT", "version": "v1" }, { "created": "Tue, 8 Nov 2016 15:26:12 GMT", "version": "v2" }, { "created": "Fri, 25 Nov 2016 17:48:36 GMT", "version": "v3" }, { "created": "Fri, 27 Jul 2018 10:16:43 GMT", "version": "v4" } ]

2018-07-30

[ [ "Bloem", "Roderick", "" ], [ "Braud-Santoni", "Nicolas", "" ], [ "Hadzic", "Vedad", "" ] ]

Fri, 4 Nov 2016 22:08:40 GMT

2016-11-04T22:08:40Z

[ "cs" ]

arxiv

null

2206.01945

Di Zhao

Lanlan Su, Di Zhao, and Sei Zhen Khong

On the exponential convergence of input-output signals of nonlinear feedback systems

This paper has been submitted to IEEE Transactions on Automatic Control

null

eess.SY cs.SY math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This note studies the exponential convergence of input-output signals of discrete-time nonlinear systems composed of a feedback interconnection of a linear time-invariant system and a nonlinear uncertainty. Both the open-loop subsystems are allowed to be unbounded. Integral-quadratic-constraint-based conditions are proposed for these uncertain feedback systems, including the Lurye type, to exhibit the property that the endogenous input-output signals enjoy an exponential convergence rate for all initial conditions of the linear time-invariant subsystem. The conditions are established via a combination of tools, including integral quadratic constraints, directed gap, and exponential weightings.

[ { "created": "Sat, 4 Jun 2022 08:53:23 GMT", "version": "v1" }, { "created": "Tue, 4 Oct 2022 15:30:42 GMT", "version": "v2" }, { "created": "Wed, 12 Jun 2024 14:38:15 GMT", "version": "v3" } ]

2024-06-13

[ [ "Su", "Lanlan", "" ], [ "Zhao", "Di", "" ], [ "Khong", "Sei Zhen", "" ] ]

Sat, 4 Jun 2022 08:53:23 GMT

2022-06-04T08:53:23Z

[ "cs", "eess", "math" ]

arxiv

null

{H}unter{S}peech{L}ab at {G}erm{E}val 2021: Does Your Comment Claim A Fact? Contextualized Embeddings for {G}erman Fact-Claiming Comment Classification

null

In this paper we investigate the efficacy of using contextual embeddings from multilingual BERT and German BERT in identifying fact-claiming comments in German on social media. Additionally, we examine the impact of formulating the classification problem as a multi-task learning problem, where the model identifies toxicity and engagement of the comment in addition to identifying whether it is fact-claiming. We provide a thorough comparison of the two BERT based models compared with a logistic regression baseline and show that German BERT features trained using a multi-task objective achieves the best F1 score on the test set. This work was done as part of a submission to GermEval 2021 shared task on the identification of fact-claiming comments.

null

acl

inproceedings

panda-levitan-2021-hunterspeechlab

Risch, Julian and Stoll, Anke and Wilms, Lena and Wiegand, Michael

sep

2,021

Duesseldorf, Germany

Association for Computational Linguistics

https://aclanthology.org/2021.germeval-1.15/

Panda, Subhadarshi and Levitan, Sarah Ita

Proceedings of the GermEval 2021 Shared Task on the Identification of Toxic, Engaging, and Fact-Claiming Comments

100--104

null

1911.01202

Michael Plank

Michael J Plank

Asymptotic expansion approximation for spatial structure arising from directionally biased movement

null

Physica A: Statistical Mechanics and its Applications (2019)

10.1016/j.physa.2019.123290

null

q-bio.CB

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Spatial structure can arise in spatial point process models via a range of mechanisms, including neighbour-dependent directionally biased movement. This spatial structure is neglected by mean-field models, but can have important effects on population dynamics. Spatial moment dynamics are one way to obtain a deterministic approximation of a dynamic spatial point process that retains some information about spatial structure. However, the applicability of this approach is limited by the computational cost of numerically solving spatial moment dynamic equations at a sufficient resolution. We present an asymptotic expansion for the equilibrium solution to the spatial moment dynamics equations in the presence of neighbour-dependent directional bias. We show that the asymptotic expansion provides a highly efficient scheme for obtaining approximate equilibrium solutions to the spatial moment dynamics equations when bias is weak. This scheme will be particularly useful for performing parameter inference on spatial moment models.

[ { "created": "Tue, 22 Oct 2019 03:07:58 GMT", "version": "v1" }, { "created": "Tue, 5 Nov 2019 01:54:51 GMT", "version": "v2" } ]

2019-11-06

[ [ "Plank", "Michael J", "" ] ]

Tue, 22 Oct 2019 03:07:58 GMT

2019-10-22T03:07:58Z

[ "q-bio" ]

arxiv

null

1506.06696

Philip Leifeld

Philip Leifeld and Skyler J. Cranmer

A Theoretical and Empirical Comparison of the Temporal Exponential Random Graph Model and the Stochastic Actor-Oriented Model

null

stat.ME

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The temporal exponential random graph model (TERGM) and the stochastic actor-oriented model (SAOM, e.g., SIENA) are popular models for longitudinal network analysis. We compare these models theoretically, via simulation, and through a real-data example in order to assess their relative strengths and weaknesses. Though we do not aim to make a general claim about either being superior to the other across all specifications, we highlight several theoretical differences the analyst might consider and find that with some specifications, the two models behave very similarly, while each model out-predicts the other one the more the specific assumptions of the respective model are met.

[ { "created": "Mon, 22 Jun 2015 17:34:53 GMT", "version": "v1" }, { "created": "Mon, 7 Sep 2015 14:43:04 GMT", "version": "v2" }, { "created": "Tue, 3 May 2016 21:37:58 GMT", "version": "v3" }, { "created": "Fri, 8 Jun 2018 04:59:13 GMT", "version": "v4" } ]

2018-06-11

[ [ "Leifeld", "Philip", "" ], [ "Cranmer", "Skyler J.", "" ] ]

Mon, 22 Jun 2015 17:34:53 GMT

2015-06-22T17:34:53Z

[ "stat" ]

arxiv

null

1206.1416

Julian Lee

Julian Lee, Steve Press\'e

A derivation of the master equation from path entropy maximization

16 pages, no figure. The problem of Eq(26) running off the page margin is corrected

Journal of Chemical Physics 137, 074103 (2012)

10.1063/1.4743955

null

cond-mat.stat-mech math-ph math.MP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The master equation and, more generally, Markov processes are routinely used as models for stochastic processes. They are often justified on the basis of randomization and coarse-graining assumptions. Here instead, we derive n-th order Markov processes and the master equation as unique solutions to an inverse problem. In particular, we find that when the constraints are not enough to uniquely determine the stochastic model, the n-th order Markov process emerges as the unique maximum entropy solution to this otherwise under-determined problem. This gives a rigorous alternative for justifying such models while providing a systematic recipe for generalizing widely accepted stochastic models usually assumed to follow from first principles.

[ { "created": "Thu, 7 Jun 2012 07:55:17 GMT", "version": "v1" }, { "created": "Fri, 8 Jun 2012 05:15:44 GMT", "version": "v2" } ]

2012-09-27

[ [ "Lee", "Julian", "" ], [ "Pressé", "Steve", "" ] ]

Thu, 7 Jun 2012 07:55:17 GMT

2012-06-07T07:55:17Z

[ "physics", "math" ]

arxiv

null

Incorporating In-Source Fragment Information Improves Metabolite Identification Accuracy in Untargeted LC-MS Data Sets.

null

In-source fragmentation occurs as a byproduct of electrospray ionization. We find that ions produced as a result of in-source fragmentation often match fragment ions produced during MS/MS fragmentation, and we take advantage of this phenomenon in a novel algorithm to analyze LC-MS metabolomics data sets. Our approach organizes coeluting MS1 features into a single peak group and then identifies in-source fragments among coeluting features using MS/MS spectral libraries. We tested our approach using previously published data of verified metabolites and compared the results to features detected by other mainstream metabolomics tools. Our results indicate that considering in-source fragment information as a part of the identification process increases the annotation quality, allowing us to leverage MS/MS data in spectrum libraries even if MS/MS scans were not collected.

null

pubmed

null

2,019

null

30,295,490

Journal of proteome research

2201.04550

Merijn Floren

Merijn Floren, Koen Classens, Tom Oomen, Jean-Philippe No\"el

Feedback linearisation of mechanical systems using data-driven models

null

Journal of Sound and Vibration, vol. 577, p. 118335, 2024

10.1016/j.jsv.2024.118335

null

eess.SY cs.SY

http://creativecommons.org/licenses/by/4.0/

Linearising the dynamics of nonlinear mechanical systems is an important and open research area. A common approach is feedback linearisation, which is a nonlinear control method that transforms the input-output response of a nonlinear system into an equivalent linear one. The main problem with feedback linearisation is that it requires an accurate first-principles model of the system, which are typically hard to obtain. In this paper, we design an alternative control approach that exploits data-driven models to linearise the input-output response of nonlinear mechanical systems. Specifically, a model-based reference tracking architecture is developed for nonlinear feedback systems with output nonlinearities. The overall methodology shows a high degree of performance combined with significant robustness against imperfect modelling and extrapolation. These findings are demonstrated using large set of synthetic experiments conducted on a asymmetric Duffing oscillator and using an experimental prototype of a high-precision motion system.

[ { "created": "Wed, 12 Jan 2022 16:44:26 GMT", "version": "v1" }, { "created": "Tue, 4 Feb 2025 08:42:43 GMT", "version": "v2" } ]

2025-02-05

[ [ "Floren", "Merijn", "" ], [ "Classens", "Koen", "" ], [ "Oomen", "Tom", "" ], [ "Noël", "Jean-Philippe", "" ] ]

Wed, 12 Jan 2022 16:44:26 GMT

2022-01-12T16:44:26Z

[ "cs", "eess" ]

arxiv

null

2103.00502

Shijun Zhang

Zuowei Shen, Haizhao Yang, Shijun Zhang

Optimal Approximation Rate of ReLU Networks in terms of Width and Depth

null

Journal de Math\'ematiques Pures et Appliqu\'ees, Volume 157, January 2022, Pages 101-135

10.1016/j.matpur.2021.07.009

null

cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper concentrates on the approximation power of deep feed-forward neural networks in terms of width and depth. It is proved by construction that ReLU networks with width $\mathcal{O}\big(\max\{d\lfloor N^{1/d}\rfloor,\, N+2\}\big)$ and depth $\mathcal{O}(L)$ can approximate a H\"older continuous function on $[0,1]^d$ with an approximation rate $\mathcal{O}\big(\lambda\sqrt{d} (N^2L^2\ln N)^{-\alpha/d}\big)$, where $\alpha\in (0,1]$ and $\lambda>0$ are H\"older order and constant, respectively. Such a rate is optimal up to a constant in terms of width and depth separately, while existing results are only nearly optimal without the logarithmic factor in the approximation rate. More generally, for an arbitrary continuous function $f$ on $[0,1]^d$, the approximation rate becomes $\mathcal{O}\big(\,\sqrt{d}\,\omega_f\big( (N^2L^2\ln N)^{-1/d}\big)\,\big)$, where $\omega_f(\cdot)$ is the modulus of continuity. We also extend our analysis to any continuous function $f$ on a bounded set. Particularly, if ReLU networks with depth $31$ and width $\mathcal{O}(N)$ are used to approximate one-dimensional Lipschitz continuous functions on $[0,1]$ with a Lipschitz constant $\lambda>0$, the approximation rate in terms of the total number of parameters, $W=\mathcal{O}(N^2)$, becomes $\mathcal{O}(\tfrac{\lambda}{W\ln W})$, which has not been discovered in the literature for fixed-depth ReLU networks.

[ { "created": "Sun, 28 Feb 2021 13:15:55 GMT", "version": "v1" }, { "created": "Wed, 24 Mar 2021 08:06:23 GMT", "version": "v2" }, { "created": "Fri, 9 Apr 2021 16:33:29 GMT", "version": "v3" }, { "created": "Sat, 17 Jul 2021 04:05:16 GMT", "version": "v4" }, { "created": "Sat, 24 Jul 2021 16:48:34 GMT", "version": "v5" } ]

2021-12-15

[ [ "Shen", "Zuowei", "" ], [ "Yang", "Haizhao", "" ], [ "Zhang", "Shijun", "" ] ]

Sun, 28 Feb 2021 13:15:55 GMT

2021-02-28T13:15:55Z

[ "cs", "stat" ]

arxiv

null

1606.04722

Xi Wu

Xi Wu, Fengan Li, Arun Kumar, Kamalika Chaudhuri, Somesh Jha, Jeffrey F. Naughton

Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics

null

cs.LG cs.CR cs.DB stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

While significant progress has been made separately on analytics systems for scalable stochastic gradient descent (SGD) and private SGD, none of the major scalable analytics frameworks have incorporated differentially private SGD. There are two inter-related issues for this disconnect between research and practice: (1) low model accuracy due to added noise to guarantee privacy, and (2) high development and runtime overhead of the private algorithms. This paper takes a first step to remedy this disconnect and proposes a private SGD algorithm to address \emph{both} issues in an integrated manner. In contrast to the white-box approach adopted by previous work, we revisit and use the classical technique of {\em output perturbation} to devise a novel "bolt-on" approach to private SGD. While our approach trivially addresses (2), it makes (1) even more challenging. We address this challenge by providing a novel analysis of the $L_2$-sensitivity of SGD, which allows, under the same privacy guarantees, better convergence of SGD when only a constant number of passes can be made over the data. We integrate our algorithm, as well as other state-of-the-art differentially private SGD, into Bismarck, a popular scalable SGD-based analytics system on top of an RDBMS. Extensive experiments show that our algorithm can be easily integrated, incurs virtually no overhead, scales well, and most importantly, yields substantially better (up to 4X) test accuracy than the state-of-the-art algorithms on many real datasets.

[ { "created": "Wed, 15 Jun 2016 11:14:29 GMT", "version": "v1" }, { "created": "Sun, 26 Feb 2017 16:26:59 GMT", "version": "v2" }, { "created": "Thu, 23 Mar 2017 17:35:09 GMT", "version": "v3" } ]

2017-03-24

[ [ "Wu", "Xi", "" ], [ "Li", "Fengan", "" ], [ "Kumar", "Arun", "" ], [ "Chaudhuri", "Kamalika", "" ], [ "Jha", "Somesh", "" ], [ "Naughton", "Jeffrey F.", "" ] ]

Wed, 15 Jun 2016 11:14:29 GMT

2016-06-15T11:14:29Z

[ "cs", "stat" ]

arxiv

null

cond-mat/9812300

Dirk Helbing

Dirk Helbing and Michael Schreckenberg

Cellular Automata Simulating Experimental Properties of Traffic Flows

For related work see http://www.theo2.physik.uni-stuttgart.de/helbing.html and http://traffic.comphys.uni-duisburg.de/member/home_schreck.html

Physical Review E 59, R2505-R2508 (1999)

10.1103/PhysRevE.59.R2505

null

cond-mat.stat-mech comp-gas nlin.CG

null

A model for 1D traffic flow is developed, which is discrete in space and time. Like the cellular automaton model by Nagel and Schreckenberg [J. Phys. I France 2, 2221 (1992)], it is simple, fast, and can describe stop-and-go traffic. Due to its relation to the optimal velocity model by Bando et al. [Phys. Rev. E 51, 1035 (1995)], its instability mechanism is of deterministic nature. The model can be easily calibrated to empirical data and displays the experimental features of traffic data recently reported by Kerner and Rehborn [Phys. Rev. E 53, R1297 (1996)].

[ { "created": "Thu, 17 Dec 1998 11:47:46 GMT", "version": "v1" }, { "created": "Sun, 21 Mar 1999 07:13:53 GMT", "version": "v2" } ]

2009-10-31

[ [ "Helbing", "Dirk", "" ], [ "Schreckenberg", "Michael", "" ] ]

Thu, 17 Dec 1998 11:47:46 GMT

1998-12-17T11:47:46Z

[ "physics", "nlin" ]

arxiv

null

1702.05583

Quan Nguyen

Avner Peleg, Quan M. Nguyen, Toan T. Huynh

Soliton-like behavior in fast two-pulse collisions in weakly perturbed linear physical systems

21 pages, 4 figures

Eur. Phys. J. D (2017) 71: 315

10.1140/epjd/e2017-80358-4

null

nlin.PS physics.optics

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We demonstrate that pulses of linear physical systems, weakly perturbed by nonlinear dissipation, exhibit soliton-like behavior in fast collisions. The behavior is demonstrated for linear waveguides with weak cubic loss and for systems described by linear diffusion-advection models with weak quadratic loss. We show that in both systems, the expressions for the collision-induced amplitude shifts due to the nonlinear loss have the same form as the expression for the amplitude shift in a fast collision between two optical solitons in the presence of weak cubic loss. Our analytic predictions are confirmed by numerical simulations with the corresponding coupled linear evolution models with weak nonlinear loss. These results open the way for studying dynamics of fast collisions between pulses of weakly perturbed linear physical systems in an arbitrary spatial dimension.

[ { "created": "Sat, 18 Feb 2017 08:00:20 GMT", "version": "v1" }, { "created": "Mon, 29 May 2017 14:47:19 GMT", "version": "v2" } ]

2017-12-06

[ [ "Peleg", "Avner", "" ], [ "Nguyen", "Quan M.", "" ], [ "Huynh", "Toan T.", "" ] ]

Sat, 18 Feb 2017 08:00:20 GMT

2017-02-18T08:00:20Z

[ "physics", "nlin" ]

arxiv

null

2203.15162

Yu Chen

Yongjian Xu and Huabin Cheng and Ning Xu and Yu Chen and Chengwang Xie

A Distribution Evolutionary Algorithm for the Graph Coloring Problem

null

cs.NE

http://creativecommons.org/licenses/by/4.0/

Graph coloring is a challenging combinatorial optimization problem with a wide range of applications. In this paper, a distribution evolutionary algorithm based on a population of probability model (DEA-PPM) is developed to address it efficiently. Unlike existing estimation of distribution algorithms where a probability model is updated by generated solutions, DEA-PPM employs a distribution population based on a novel probability model, and an orthogonal exploration strategy is introduced to search the distribution space with the assistance of an refinement strategy. By sampling the distribution population, efficient search in the solution space is realized based on a tabu search process. Meanwhile, DEA-PPM introduces an iterative vertex removal strategy to improve the efficiency of $k$-coloring, and an inherited initialization strategy is implemented to address the chromatic problem well. The cooperative evolution of the distribution population and the solution population leads to a good balance between exploration and exploitation. Numerical results demonstrate that the DEA-PPM of small population size is competitive to the state-of-the-art metaheuristics.utes to its competitiveness to the state-of-the-art metaheuristics.

[ { "created": "Tue, 29 Mar 2022 00:50:30 GMT", "version": "v1" }, { "created": "Wed, 21 Sep 2022 14:38:50 GMT", "version": "v2" }, { "created": "Tue, 2 May 2023 13:50:15 GMT", "version": "v3" } ]

2023-05-03

[ [ "Xu", "Yongjian", "" ], [ "Cheng", "Huabin", "" ], [ "Xu", "Ning", "" ], [ "Chen", "Yu", "" ], [ "Xie", "Chengwang", "" ] ]

Tue, 29 Mar 2022 00:50:30 GMT

2022-03-29T00:50:30Z

[ "cs" ]

arxiv

null

2110.11156

Parley Ruogu Yang

Parley Ruogu Yang, Ryan Lucas

DMS, AE, DAA: methods and applications of adaptive time series model selection, ensemble, and financial evaluation

Key words: Time series, model selection, model evaluation, cross-asset strategy, market crash, VIX

null

stat.AP econ.EM q-fin.ST stat.ML stat.OT

http://creativecommons.org/licenses/by/4.0/

We introduce three adaptive time series learning methods, called Dynamic Model Selection (DMS), Adaptive Ensemble (AE), and Dynamic Asset Allocation (DAA). The methods respectively handle model selection, ensembling, and contextual evaluation in financial time series. Empirically, we use the methods to forecast the returns of four key indices in the US market, incorporating information from the VIX and Yield curves. We present financial applications of the learning results, including fully-automated portfolios and dynamic hedging strategies. The strategies strongly outperform long-only benchmarks over our testing period, spanning from Q4 2015 to the end of 2021. The key outputs of the learning methods are interpreted during the 2020 market crash.

[ { "created": "Thu, 21 Oct 2021 13:59:54 GMT", "version": "v1" }, { "created": "Wed, 17 Nov 2021 15:51:42 GMT", "version": "v2" }, { "created": "Tue, 5 Jul 2022 11:26:19 GMT", "version": "v3" } ]

2022-07-06

[ [ "Yang", "Parley Ruogu", "" ], [ "Lucas", "Ryan", "" ] ]

Thu, 21 Oct 2021 13:59:54 GMT

2021-10-21T13:59:54Z

[ "stat", "econ", "q-fin" ]

arxiv

null

1308.0695

Liane Gabora

Reply to Commentaries on 'An Evolutionary Framework for Cultural Change: Selectionism versus Communal Exchange'

8 pages; this is a reply to commentaries on q-bio.PE/0773336

Physics of Life Reviews, 10(2), 162-167 (2013)

10.1016/j.plrev.2013.05.007

null

q-bio.PE nlin.AO q-bio.NC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The commentators have brought a wealth of new perspectives to the question of how culture evolves. Each of their diverse disciplines--ranging from psychology to biology to anthropology to economics to engineering--has a valuable contribution to make to our understanding of this complex, multifaceted topic. Though the vast majority of their comments were supportive of my approach, it is natural that a reply such as this focus on points where my views differ from that of the commentators. ... I conclude by saying that I am grateful to the commentators for their diverse perspectives and insights, their overall support for the project, and provocative ideas for where to go from here. Clearly there are many fascinating avenues to explore as we move forward on our quest to understand how culture evolves.

[ { "created": "Sat, 3 Aug 2013 13:44:11 GMT", "version": "v1" }, { "created": "Sun, 30 Jun 2019 01:56:10 GMT", "version": "v2" } ]

2019-07-02

[ [ "Gabora", "Liane", "" ] ]

Sat, 3 Aug 2013 13:44:11 GMT

2013-08-03T13:44:11Z

[ "q-bio", "nlin" ]

arxiv

null

2105.10087

Zhehua Mao

Zhehua Mao, Liang Zhao, Shoudong Huang, Yiting Fan, and Alex Pui-Wai Lee

DSR: Direct Simultaneous Registration for Multiple 3D Images

10 pages, 3 figures, The 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022

Medical Image Computing and Computer Assisted Intervention (2022)

10.1007/978-3-031-16446-0_10

null

cs.CV cs.RO eess.IV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper presents a novel algorithm named Direct Simultaneous Registration (DSR) that registers a collection of 3D images in a simultaneous fashion without specifying any reference image, feature extraction and matching, or information loss or reuse. The algorithm optimizes the global poses of local image frames by maximizing the similarity between a predefined panoramic image and local images. Although we formulate the problem as a Direct Bundle Adjustment (DBA) that jointly optimizes the poses of local frames and the intensities of the panoramic image, by investigating the independence of pose estimation from the panoramic image in the solving process, DSR is proposed to solve the poses only and proved to be able to obtain the same optimal poses as DBA. The proposed method is particularly suitable for the scenarios where distinct features are not available, such as Transesophageal Echocardiography (TEE) images. DSR is evaluated by comparing it with four widely used methods via simulated and in-vivo 3D TEE images. It is shown that the proposed method outperforms these four methods in terms of accuracy and requires much fewer computational resources than the state-of-the-art accumulated pairwise estimates (APE).

[ { "created": "Fri, 21 May 2021 01:42:11 GMT", "version": "v1" }, { "created": "Mon, 15 Aug 2022 07:01:56 GMT", "version": "v2" } ]

2023-02-14

[ [ "Mao", "Zhehua", "" ], [ "Zhao", "Liang", "" ], [ "Huang", "Shoudong", "" ], [ "Fan", "Yiting", "" ], [ "Lee", "Alex Pui-Wai", "" ] ]

Fri, 21 May 2021 01:42:11 GMT

2021-05-21T01:42:11Z

[ "cs", "eess" ]

arxiv

null

Content Modeling for Automated Oral Proficiency Scoring System

null

We developed an automated oral proficiency scoring system for non-native English speakers' spontaneous speech. Automated systems that score holistic proficiency are expected to assess a wide range of performance categories, and the content is one of the core performance categories. In order to assess the quality of the content, we trained a Siamese convolutional neural network (CNN) to model the semantic relationship between key points generated by experts and a test response. The correlation between human scores and Siamese CNN scores was comparable to human-human agreement (r=0.63), and it was higher than the baseline content features. The inclusion of Siamese CNN-based feature to the existing state-of-the-art automated scoring model achieved a small but statistically significant improvement. However, the new model suffered from score inflation for long atypical responses with serious content issues. We investigated the reasons of this score inflation by analyzing the associations with linguistic features and identifying areas strongly associated with the score errors.

null

acl

inproceedings

yoon-lee-2019-content

Yannakoudakis, Helen and Kochmar, Ekaterina and Leacock, Claudia and Madnani, Nitin and Pil{\'a}n, Ildik{\'o} and Zesch, Torsten

aug

2,019

Florence, Italy

Association for Computational Linguistics

https://aclanthology.org/W19-4441/

Yoon, Su-Youn and Lee, Chong Min

Proceedings of the Fourteenth Workshop on Innovative Use of NLP for Building Educational Applications

394--401

null

2104.04473

Deepak Narayanan

Deepak Narayanan, Mohammad Shoeybi, Jared Casper, Patrick LeGresley, Mostofa Patwary, Vijay Anand Korthikanti, Dmitri Vainbrand, Prethvi Kashinkunti, Julie Bernauer, Bryan Catanzaro, Amar Phanishayee, Matei Zaharia

Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM

Accepted to SC 2021

null

cs.CL cs.DC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Large language models have led to state-of-the-art accuracies across a range of tasks. However, training these models efficiently is challenging for two reasons: a) GPU memory capacity is limited, making it impossible to fit large models on even a multi-GPU server, and b) the number of compute operations required to train these models can result in unrealistically long training times. Consequently, new methods of model parallelism such as tensor and pipeline parallelism have been proposed. Unfortunately, naive usage of these methods leads to fundamental scaling issues at thousands of GPUs, e.g., due to expensive cross-node communication or devices spending significant time waiting on other devices to make progress. In this paper, we show how different types of parallelism methods (tensor, pipeline, and data parallelism) can be composed to scale to thousands of GPUs and models with trillions of parameters. We survey techniques for pipeline parallelism and propose a novel interleaved pipeline parallelism schedule that can improve throughput by 10+% with memory footprint comparable to existing approaches. We quantitatively study the trade-offs between tensor, pipeline, and data parallelism, and provide intuition as to how to configure distributed training of a large model. Our approach allows us to perform training iterations on a model with 1 trillion parameters at 502 petaFLOP/s on 3072 GPUs with achieved per-GPU throughput of 52% of theoretical peak. Our code is open sourced at https://github.com/nvidia/megatron-lm.

[ { "created": "Fri, 9 Apr 2021 16:43:11 GMT", "version": "v1" }, { "created": "Fri, 14 May 2021 17:44:52 GMT", "version": "v2" }, { "created": "Fri, 30 Jul 2021 07:18:32 GMT", "version": "v3" }, { "created": "Sun, 15 Aug 2021 07:11:58 GMT", "version": "v4" }, { "created": "Mon, 23 Aug 2021 19:41:13 GMT", "version": "v5" } ]

2021-08-25

[ [ "Narayanan", "Deepak", "" ], [ "Shoeybi", "Mohammad", "" ], [ "Casper", "Jared", "" ], [ "LeGresley", "Patrick", "" ], [ "Patwary", "Mostofa", "" ], [ "Korthikanti", "Vijay Anand", "" ], [ "Vainbrand", "Dmitri", "" ], [ "Kashinkunti", "Prethvi", "" ], [ "Bernauer", "Julie", "" ], [ "Catanzaro", "Bryan", "" ], [ "Phanishayee", "Amar", "" ], [ "Zaharia", "Matei", "" ] ]

Fri, 9 Apr 2021 16:43:11 GMT

2021-04-09T16:43:11Z

[ "cs" ]

arxiv

null

Annotating dropped pronouns in {C}hinese newswire text

null

We propose an annotation framework to explicitly identify dropped subject pronouns in Chinese. We acknowledge and specify 10 concrete pronouns that exist as words in Chinese and 4 abstract pronouns that do not correspond to Chinese words, but that are recognized conceptually, to native Chinese speakers. These abstract pronouns are identified as ''''''``unspecified'''''''', ''''''``pleonastic'''''''', ''''''``event'''''''', and ''''''``existential'''''''' and are argued to exist cross-linguistically. We trained two annotators, fluent in Chinese, and adjudicated their annotations to form a gold standard. We achieved an inter-annotator agreement kappa of .6 and an observed agreement of .7. We found that annotators had the most difficulty with the abstract pronouns, such as ''''''``unspecified'''''''' and ''''''``event'''''''', but we posit that further specification and training has the potential to significantly improve these results. We believe that this annotated data will serve to help improve Machine Translation models that translate from Chinese to a non pro-drop language, like English, that requires all subject pronouns to be explicit.

null

acl

inproceedings

baran-etal-2012-annotating

Calzolari, Nicoletta and Choukri, Khalid and Declerck, Thierry and Do{\u{g}}an, Mehmet U{\u{g}}ur and Maegaard, Bente and Mariani, Joseph and Moreno, Asuncion and Odijk, Jan and Piperidis, Stelios

may

2,012

Istanbul, Turkey

European Language Resources Association (ELRA)

https://aclanthology.org/L12-1177/

Baran, Elizabeth and Yang, Yaqin and Xue, Nianwen

Proceedings of the Eighth International Conference on Language Resources and Evaluation ({LREC}`12)

2795--2799

null

2207.00752

Hirofumi Notsu

Md Mamunur Rasid, Masato Kimura, Md Masum Murshed, Erny Rahayu Wijayanti, and Hirofumi Notsu

A two-step Lagrange-Galerkin scheme for the shallow water equations with a transmission boundary condition and its application to the Bay of Bengal region. Part I: Flat bottom topography

The contents, as well as the title and the abstract, are partially modified

Mathematics, 11(7), 1633, 2023

10.3390/math11071633

null

math.NA cs.NA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This study presents a two-step Lagrange-Galerkin scheme for the shallow water equations with a transmission boundary condition (TBC). Firstly, the experimental order of convergence of the scheme is shown to see the second-order accuracy in time. Secondly, the effect of the TBC on a simple domain is discussed; the artificial reflections are kept from the Dirichlet boundaries and removed significantly from the transmission boundaries. Thirdly, the scheme is applied to a complex practical domain, i.e., the Bay of Bengal region, which is non-convex and includes islands. The effect of the TBC is discussed again for the complex domain; the artificial reflections are removed significantly from transmission boundaries, which are set on open sea boundaries. Based on the numerical results, it is revealed that the scheme has the following properties; (i) the same advantages of Lagrange-Galerkin methods (the CFL-free robustness for convection-dominated problems and the symmetry of the matrices for the system of linear equations); (ii) second-order accuracy in time; (iii) mass preservation of the function for the water level from the reference height (until the contact with the transmission boundaries of the wave); and (iv) no significant artificial reflection from the transmission boundaries. The numerical results by the scheme are presented in this paper for the flat bottom topography of the domain. In the next part of this work, Part II, the scheme will be applied to rapidly varying bottom surfaces and a real bottom topography of the Bay of Bengal region.

[ { "created": "Sat, 2 Jul 2022 06:35:08 GMT", "version": "v1" }, { "created": "Tue, 5 Jul 2022 15:42:13 GMT", "version": "v2" }, { "created": "Wed, 27 Jul 2022 20:59:59 GMT", "version": "v3" }, { "created": "Sat, 11 Feb 2023 10:30:11 GMT", "version": "v4" } ]

2023-03-31

[ [ "Rasid", "Md Mamunur", "" ], [ "Kimura", "Masato", "" ], [ "Murshed", "Md Masum", "" ], [ "Wijayanti", "Erny Rahayu", "" ], [ "Notsu", "Hirofumi", "" ] ]

Sat, 2 Jul 2022 06:35:08 GMT

2022-07-02T06:35:08Z

[ "cs", "math" ]

arxiv

null

Haemophagocytic lymphohistiocytosis: a case series from Mumbai.

null

A retrospective review of ten patients (8 girls, 2 boys) admitted over a 9-month period with haemophagocytic lymphohistiocytosis (HLH) is presented. Presenting features included fever and hepatosplenomegaly (10), bleeding manifestations (7), lymphadenopathy (4), skin rash (4), shock (4), jaundice (3), CNS disorder (3), renal failure (2) and arthritis (2). Three infants had familial HLH (FHL) while the other seven patients had acquired (secondary) HLH. Two patients with FHL had very low perforin levels (0 and 0.05%). There was secondary HLH owing to systemic onset juvenile idiopathic arthritis in two patients, and one each had anaplastic large cell lymphoma, measles with pneumonia, disseminated tuberculosis, dengue hemorrhagic fever and lymphoproliferative disorder. Cytopenia affecting two or three lineages in peripheral blood was present in all while haemophagocytosis in bone marrow was documented in nine patients .Other important laboratory parameters were raised ferritin (9), raised LDH (9), hypertriglyceridaemia (7) and hypofibrinogenaemia (5). The patients were treated according to the HLH2004 protocol. Diagnosis of HLH should be considered early in patients presenting with unremitting fever, hepatosplenomegaly and cytopenias as without appropriate treatment HLH is usually fatal.

null

pubmed

null

2,011

null

21,575,318

Annals of tropical paediatrics

1811.09194

Garth Wells

Sander Rhebergen and Garth N. Wells

An embedded--hybridized discontinuous Galerkin finite element method for the Stokes equations

null

Computer Methods in Applied Mechanics and Engineering, 358 (2020), 112619

10.1016/j.cma.2019.112619

null

math.NA cs.CE cs.NA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present and analyze a new embedded--hybridized discontinuous Galerkin finite element method for the Stokes problem. The method has the attractive properties of full hybridized methods, namely an $H({\rm div})$-conforming velocity field, pointwise satisfaction of the continuity equation and \emph{a priori} error estimates for the velocity that are independent of the pressure. The embedded--hybridized formulation has advantages over a full hybridized formulation in that it has fewer global degrees-of-freedom for a given mesh and the algebraic structure of the resulting linear system is better suited to fast iterative solvers. The analysis results are supported by a range of numerical examples that demonstrate rates of convergence, and which show computational efficiency gains over a full hybridized formulation.

[ { "created": "Thu, 22 Nov 2018 14:36:33 GMT", "version": "v1" }, { "created": "Wed, 18 Sep 2019 13:47:31 GMT", "version": "v2" } ]

2023-07-06

[ [ "Rhebergen", "Sander", "" ], [ "Wells", "Garth N.", "" ] ]

Thu, 22 Nov 2018 14:36:33 GMT

2018-11-22T14:36:33Z

[ "cs", "math" ]

arxiv

null

1901.02123

Jinha Park

Jinha Park, B. Kahng

Competing synchronization on random networks

null

J. Stat. Mech. (2020) 073407

10.1088/1742-5468/ab97b0

null

cond-mat.stat-mech cond-mat.dis-nn nlin.CD physics.bio-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The synchronization pattern of a fully connected competing Kuramoto model with a uniform intrinsic frequency distribution $g(\omega)$ was recently considered. This competing Kuramoto model assigns two coupling constants with opposite signs, $K_1 < 0$ and $K_2 > 0$, to the $1-p$ and $p$ fractions of nodes, respectively. This model has a rich phase diagram that includes incoherent, $\pi$, and traveling wave (TW) phases and a hybrid phase transition with abnormal properties that occurs through an intermediate metastable $\pi$ state. Here, we consider the competing Kuramoto model on Erd\H{o}s--R\'enyi (ER) random networks. Numerical simulations and the mean-field solution based on the annealed network approximation reveal that in this case, when the mean degree of the random networks is large, the features of the phase diagram and transition types are consistent overall with those on completely connected networks. However, when the mean degree is small, the mean-field solution is not consistent with the numerical simulation results; specifically, the TW state does not occur, and thus the phase diagram is changed, owing to the strong heterogeneity of the local environment. By contrast, for the original Kuramoto oscillators, the annealed mean-field solution is consistent with the numerical simulation result for ER networks.

[ { "created": "Tue, 8 Jan 2019 01:53:07 GMT", "version": "v1" }, { "created": "Tue, 31 Mar 2020 02:40:31 GMT", "version": "v2" } ]

2020-09-04

[ [ "Park", "Jinha", "" ], [ "Kahng", "B.", "" ] ]

Tue, 8 Jan 2019 01:53:07 GMT

2019-01-08T01:53:07Z

[ "physics", "nlin" ]

arxiv

null

1311.2240

Sergey Ya. Startsev

Non-Point Invertible Transformations and Integrability of Partial Difference Equations

null

SIGMA 10 (2014), 066, 13 pages

10.3842/SIGMA.2014.066

null

nlin.SI math-ph math.MP

http://creativecommons.org/licenses/by-nc-sa/3.0/

This article is devoted to the partial difference quad-graph equations that can be represented in the form $\varphi (u(i+1,j),u(i+1,j+1))=\psi (u(i,j),u(i,j+1))$, where the map $(w,z) \rightarrow (\varphi(w,z),\psi(w,z))$ is injective. The transformation $v(i,j)=\varphi (u(i,j),u(i,j+1))$ relates any of such equations to a quad-graph equation. It is proved that this transformation maps Darboux integrable equations of the above form into Darboux integrable equations again and decreases the orders of the transformed integrals by one in the $j$-direction. As an application of this fact, the Darboux integrable equations possessing integrals of the second order in the $j$-direction are described under an additional assumption. The transformation also maps symmetries of the original equations into symmetries of the transformed equations (i.e. preserves the integrability in the sense of the symmetry approach) and acts as a difference substitution for symmetries of a special form. The latter fact allows us to derive necessary conditions of Darboux integrability for the equations defined in the first sentence of the abstract.

[ { "created": "Sun, 10 Nov 2013 02:23:41 GMT", "version": "v1" }, { "created": "Mon, 2 Dec 2013 23:20:39 GMT", "version": "v2" }, { "created": "Mon, 28 Apr 2014 14:50:10 GMT", "version": "v3" }, { "created": "Tue, 17 Jun 2014 03:54:09 GMT", "version": "v4" } ]

2014-06-18

[ [ "Startsev", "Sergey Ya.", "" ] ]

Sun, 10 Nov 2013 02:23:41 GMT

2013-11-10T02:23:41Z

[ "nlin", "math" ]

arxiv

null

astro-ph/0009447

Juri Poutanen

Boris E. Stern, Yana Tikhomirova, Dmitrii Kompaneets, Roland Svensson, Juri Poutanen

An Off-line Scan of the BATSE Daily Records and a Large Uniform Sample of Gamma-Ray Bursts

14 pages; 23 figures; revised version accepted to ApJ; electronic version of Table 2 is available at ftp://ftp.astro.su.se/pub/head/grb/catalogs/etable2.txt The GRB archive is available at http://www.astro.su.se/groups/head/grb_archive.html

Astrophys.J. 563 (2001) 80

10.1086/322295

Stockholm Observatory High Energy Astrophysics Preprint # 37

astro-ph

null

During a scan of the archival BATSE daily records covering the entire 9.1 years (TJD 8369-11690) of the BATSE operation, 3906 gamma-ray bursts (GRBs) have been detected. 2068 of these GRBs are previously known BATSE triggers while 1838 of them are new non-triggered bursts. It is important that all events were detected in the same type of data and were processed with the same procedure. Therefore these 3906 GRBs constitute a uniform sample. We have created a publically available electronic data base containing this sample. We describe the procedures of the data reduction, the selection of the GRB candidates, and the statistical tests for possible non-GRB contaminations. We also describe a novel test burst method used to measure the scan efficiency and the information obtained using the test bursts. Our scan decreases the BATSE detection threshold to ~0.1 photons/sec/cm2. As a first result, we show that the differential log N - log P distribution corrected for the detection efficiency extends to low brightnesses without any indication of a turn-over. Any reasonable extrapolation of the new log N - log P to lower brightnesses imply a rate of several thousands of GRBs in the Universe per year.

[ { "created": "Wed, 27 Sep 2000 19:35:41 GMT", "version": "v1" }, { "created": "Fri, 11 May 2001 09:19:58 GMT", "version": "v2" } ]

2009-10-31

[ [ "Stern", "Boris E.", "" ], [ "Tikhomirova", "Yana", "" ], [ "Kompaneets", "Dmitrii", "" ], [ "Svensson", "Roland", "" ], [ "Poutanen", "Juri", "" ] ]

Wed, 27 Sep 2000 19:35:41 GMT

2000-09-27T19:35:41Z

[ "physics" ]

arxiv

null

1006.1937

EPTCS

S. Barry Cooper, Prakash Panangaden, Elham Kashefi

Proceedings Sixth Workshop on Developments in Computational Models: Causality, Computation, and Physics

null

EPTCS 26, 2010

10.4204/EPTCS.26

null

quant-ph cs.LO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

DCM 2010 provides a forum for ideas about new computing means and models, with a particular emphasis in 2010 on computational and causal models related to physics and biology. We believe that bringing together different approaches - in a community with the strong foundational background characteristic of FLoC - results in inspirational cross-boundary exchanges, and innovative further research. Day two of this pre-FLoC 2010 workshop is given over to physics and quantum related computation. The content of day one is more typical of previous DCM workshops - covering a full spectrum of topics related to the development of new computational models or new features for traditional computational models. DCM 2010 was designed to foster interactions, and provide a forum for presenting new ideas and work in progress. It is also intended to enable newcomers to learn about current research in this area.

[ { "created": "Thu, 10 Jun 2010 01:57:45 GMT", "version": "v1" }, { "created": "Fri, 23 Jul 2010 04:36:08 GMT", "version": "v2" } ]

2010-07-26

[ [ "Cooper", "S. Barry", "" ], [ "Panangaden", "Prakash", "" ], [ "Kashefi", "Elham", "" ] ]

Thu, 10 Jun 2010 01:57:45 GMT

2010-06-10T01:57:45Z

[ "physics", "cs" ]

arxiv

null

1911.12825

Jhelum Chakravorty

Jhelum Chakravorty, Nadeem Ward, Julien Roy, Maxime Chevalier-Boisvert, Sumana Basu, Andrei Lupu, Doina Precup

Option-Critic in Cooperative Multi-agent Systems

null

cs.AI cs.MA cs.SY eess.SY math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we investigate learning temporal abstractions in cooperative multi-agent systems, using the options framework (Sutton et al, 1999). First, we address the planning problem for the decentralized POMDP represented by the multi-agent system, by introducing a \emph{common information approach}. We use the notion of \emph{common beliefs} and broadcasting to solve an equivalent centralized POMDP problem. Then, we propose the Distributed Option Critic (DOC) algorithm, which uses centralized option evaluation and decentralized intra-option improvement. We theoretically analyze the asymptotic convergence of DOC and build a new multi-agent environment to demonstrate its validity. Our experiments empirically show that DOC performs competitively against baselines and scales with the number of agents.

[ { "created": "Thu, 28 Nov 2019 18:38:19 GMT", "version": "v1" }, { "created": "Mon, 6 Jan 2020 05:50:51 GMT", "version": "v2" }, { "created": "Thu, 19 Mar 2020 23:11:08 GMT", "version": "v3" } ]

2020-03-23

[ [ "Chakravorty", "Jhelum", "" ], [ "Ward", "Nadeem", "" ], [ "Roy", "Julien", "" ], [ "Chevalier-Boisvert", "Maxime", "" ], [ "Basu", "Sumana", "" ], [ "Lupu", "Andrei", "" ], [ "Precup", "Doina", "" ] ]

Thu, 28 Nov 2019 18:38:19 GMT

2019-11-28T18:38:19Z

[ "cs", "eess", "math" ]

arxiv

null

1105.0864

Gerald Teschl

Julian King, Karl Unterkofler, Gerald Teschl, Susanne Teschl, Pawel Mochalski, Helin Koc, Hartmann Hinterhuber, and Anton Amann

A modeling-based evaluation of isothermal rebreathing for breath gas analyses of highly soluble volatile organic compounds

21 pages

J. Breath Res. 6, 016005, 11pp (2012)

10.1088/1752-7155/6/1/016005

null

q-bio.QM physics.med-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Isothermal rebreathing has been proposed as an experimental technique for estimating the alveolar levels of hydrophilic volatile organic compounds (VOCs) in exhaled breath. Using the prototypic test compound acetone we demonstrate that the end-tidal breath profiles of such substances during isothermal rebreathing show characteristics that contradict the conventional pulmonary inert gas elimination theory due to Farhi. On the other hand, these profiles can reliably be captured by virtue of a previously developed mathematical model for the general exhalation kinetics of highly soluble, blood-borne VOCs, which explicitly takes into account airway gas exchange as major determinant of the observable breath output. This model allows for a mechanistic analysis of various rebreathing protocols suggested in the literature. In particular, it clarifies the discrepancies between in vitro and in vivo blood-breath ratios of hydrophilic VOCs and yields further quantitative insights into the physiological components of isothermal rebreathing.

[ { "created": "Wed, 4 May 2011 16:20:02 GMT", "version": "v1" }, { "created": "Fri, 16 Dec 2011 15:30:38 GMT", "version": "v2" } ]

2012-12-11

[ [ "King", "Julian", "" ], [ "Unterkofler", "Karl", "" ], [ "Teschl", "Gerald", "" ], [ "Teschl", "Susanne", "" ], [ "Mochalski", "Pawel", "" ], [ "Koc", "Helin", "" ], [ "Hinterhuber", "Hartmann", "" ], [ "Amann", "Anton", "" ] ]

Wed, 4 May 2011 16:20:02 GMT

2011-05-04T16:20:02Z

[ "q-bio", "physics" ]

arxiv

null

0809.1135

Gersende Fort

Pierre Etor\'e (CMAP), Gersende Fort (LTCI), Benjamin Jourdain (CERMICS), Eric Moulines (LTCI)

On adaptive stratification

null

math.PR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper investigates the use of stratified sampling as a variance reduction technique for approximating integrals over large dimensional spaces. The accuracy of this method critically depends on the choice of the space partition, the strata, which should be ideally fitted to thesubsets where the functions to integrate is nearly constant, and on the allocation of the number of samples within each strata. When the dimension is large and the function to integrate is complex, finding such partitions and allocating the sample is a highly non-trivial problem. In this work, we investigate a novel method to improve the efficiency of the estimator "on the fly", by jointly sampling and adapting the strata and the allocation within the strata. The accuracy of estimators when this method is used is examined in detail, in the so-called asymptotic regime (i.e. when both the number of samples and the number of strata are large). We illustrate the use of the method for the computation of the price of path-dependent options in models with both constant and stochastic volatility. The use of this adaptive technique yields variance reduction by factors sometimes larger than 1000 compared to classical Monte Carlo estimators.

[ { "created": "Sat, 6 Sep 2008 05:22:21 GMT", "version": "v1" }, { "created": "Fri, 4 Sep 2009 06:34:44 GMT", "version": "v2" }, { "created": "Tue, 15 Sep 2009 07:18:12 GMT", "version": "v3" } ]

2009-09-15

[ [ "Etoré", "Pierre", "", "CMAP" ], [ "Fort", "Gersende", "", "LTCI" ], [ "Jourdain", "Benjamin", "", "CERMICS" ], [ "Moulines", "Eric", "", "LTCI" ] ]

Sat, 6 Sep 2008 05:22:21 GMT

2008-09-06T05:22:21Z

[ "math" ]

arxiv

null

1709.01808

Hamid Reza Moradi

H.R. Moradi, S. Furuichi and M. Sababheh

On the Operator Jensen-Mercer Inequality

null

math.FA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Mercer inequality for convex functions is a variant of Jensen's inequality, with an operator version that is still valid without operator convexity. This paper is two folded. First, we present a Mercer-type inequality for operators without assuming convexity nor operator convexity. Yet, this form refines the known inequalities in the literature. Second, we present a log-convex version for operators. We then use these results to refine some inequalities related to quasi-arithmetic means of Mercer's type for operators.

[ { "created": "Wed, 6 Sep 2017 12:49:41 GMT", "version": "v1" }, { "created": "Tue, 8 May 2018 05:36:48 GMT", "version": "v2" }, { "created": "Tue, 11 Feb 2020 16:16:44 GMT", "version": "v3" }, { "created": "Thu, 5 Mar 2020 06:46:25 GMT", "version": "v4" } ]

2020-03-06

[ [ "Moradi", "H. R.", "" ], [ "Furuichi", "S.", "" ], [ "Sababheh", "M.", "" ] ]

Wed, 6 Sep 2017 12:49:41 GMT

2017-09-06T12:49:41Z

[ "math" ]

arxiv

null

1905.12155

Matteo Dell'Amico Ph.D.

Michael Mitzenmacher and Matteo Dell'Amico

The Supermarket Model with Known and Predicted Service Times

Revision, 16 pages, 22 figures. Published at IEEE TPDS

null

10.1109/TPDS.2022.3146195

null

cs.PF cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The supermarket model refers to a system with a large number of queues, where new customers choose d queues at random and join the one with the fewest customers. This model demonstrates the power of even small amounts of choice, as compared to simply joining a queue chosen uniformly at random, for load balancing systems. In this work we perform simulation-based studies to consider variations where service times for a customer are predicted, as might be done in modern settings using machine learning techniques or related mechanisms. Our primary takeaway is that using even seemingly weak predictions of service times can yield significant benefits over blind First In First Out queueing in this context. However, some care must be taken when using predicted service time information to both choose a queue and order elements for service within a queue; while in many cases using the information for both choosing and ordering is beneficial, in many of our simulation settings we find that simply using the number of jobs to choose a queue is better when using predicted service times to order jobs in a queue. In our simulations, we evaluate both synthetic and real-world workloads--in the latter, service times are predicted by machine learning. Our results provide practical guidance for the design of real-world systems; moreover, we leave many natural theoretical open questions for future work, validating their relevance to real-world situations.

[ { "created": "Thu, 23 May 2019 15:33:45 GMT", "version": "v1" }, { "created": "Thu, 8 Oct 2020 14:44:09 GMT", "version": "v2" }, { "created": "Sat, 22 Jan 2022 14:26:00 GMT", "version": "v3" }, { "created": "Thu, 17 Feb 2022 08:30:37 GMT", "version": "v4" } ]

2022-02-18

[ [ "Mitzenmacher", "Michael", "" ], [ "Dell'Amico", "Matteo", "" ] ]

Thu, 23 May 2019 15:33:45 GMT

2019-05-23T15:33:45Z

[ "cs" ]

arxiv

null

1805.08608

Kyle Baldwin

K. A. Baldwin, J.-B. de Fouchier, P. Atkinson, R. J. A. Hill, M. R. Swift, and D. J. Fairhurst

Magnetic Levitation Stabilized by Streaming Fluid Flows

5 pages, 3 figures, SI material included as pdf

Phys. Rev. Lett. 121, 064502 (2018)

10.1103/PhysRevLett.121.064502

null

cond-mat.soft

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We demonstrate that the ubiquitous laboratory magnetic stirrer provides a simple passive method of magnetic levitation, in which the so-called `flea' levitates indefinitely. We study the onset of levitation and quantify the flea's motion (a combination of vertical oscillation, spinning and "waggling"), finding excellent agreement with a mechanical analytical model. The waggling motion drives recirculating flow, producing a centripetal reaction force that stabilises the flea. Our findings have implications for the locomotion of artificial swimmers, for the development of bidirectional microfluidic pumps and provide an alternative to sophisticated commercial levitators.

[ { "created": "Tue, 22 May 2018 14:26:07 GMT", "version": "v1" }, { "created": "Thu, 24 May 2018 14:58:45 GMT", "version": "v2" } ]

2018-08-15

[ [ "Baldwin", "K. A.", "" ], [ "de Fouchier", "J. -B.", "" ], [ "Atkinson", "P.", "" ], [ "Hill", "R. J. A.", "" ], [ "Swift", "M. R.", "" ], [ "Fairhurst", "D. J.", "" ] ]

Tue, 22 May 2018 14:26:07 GMT

2018-05-22T14:26:07Z

[ "physics" ]

arxiv

null

0707.1341

Liang Jiang

Liang Jiang, M. V. Gurudev Dutt, Emre Togan, Lily Childress, Paola Cappellaro, Jacob M. Taylor, Mikhail D. Lukin

Coherence of an optically illuminated single nuclear spin qubit

5 pages, 2 figures

Phys. Rev. Lett. 100, 073001 (2008)

10.1103/PhysRevLett.100.073001

null

quant-ph

null

We investigate the coherence properties of individual nuclear spin quantum bits in diamond [Dutt et al., Science, 316, 1312 (2007)] when a proximal electronic spin associated with a nitrogen-vacancy (NV) center is being interrogated by optical radiation. The resulting nuclear spin dynamics are governed by time-dependent hyperfine interaction associated with rapid electronic transitions, which can be described by a spin-fluctuator model. We show that due to a process analogous to motional averaging in nuclear magnetic resonance, the nuclear spin coherence can be preserved after a large number of optical excitation cycles. Our theoretical analysis is in good agreement with experimental results. It indicates a novel approach that could potentially isolate the nuclear spin system completely from the electronic environment.

[ { "created": "Mon, 9 Jul 2007 21:57:39 GMT", "version": "v1" }, { "created": "Wed, 20 Feb 2008 01:42:18 GMT", "version": "v2" } ]

2008-02-20

[ [ "Jiang", "Liang", "" ], [ "Dutt", "M. V. Gurudev", "" ], [ "Togan", "Emre", "" ], [ "Childress", "Lily", "" ], [ "Cappellaro", "Paola", "" ], [ "Taylor", "Jacob M.", "" ], [ "Lukin", "Mikhail D.", "" ] ]

Mon, 9 Jul 2007 21:57:39 GMT

2007-07-09T21:57:39Z

[ "physics" ]

arxiv

null

1302.5447

Laura Tolos

Cristina Manuel, Jaume Tarrus and Laura Tolos

Bulk viscosity coefficients due to phonons in superfluid neutron stars

25 pages, 10 figures, published version

JCAP07(2013)003

10.1088/1475-7516/2013/07/003

null

astro-ph.HE hep-ph nucl-th

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We calculate the three bulk viscosity coefficients as arising from the collisions among phonons in superfluid neutron stars. We use effective field theory techniques to extract the allowed phonon collisional processes, written as a function of the equation of state of the system. The solution of the dynamical evolution of the phonon number density allows us to calculate the bulk viscosity coefficients as function of the phonon collisional rate and the phonon dispersion law, which depends on the neutron pairing gap. Our method of computation is rather general, and could be used for different superfluid systems, provided they share the same underlying symmetries. We find that the behavior with temperature of the bulk viscosity coefficients is dominated by the contributions coming from the collinear regime of the $2\leftrightarrow3$ phonon processes. For typical star radial pulsation frequencies of $\omega \sim 10^{4} s^{-1}$, we obtain that the bulk viscosity coefficients at densities $n \gtrsim 4n_0$ are within 10% from its static value for $T \lesssim 10^9 K$ and for the case of strong neutron superfluidity in the core with a maximum value of the $^3P_2$ gap above 1 MeV, while, otherwise, the static solution is not a valid approximation to the bulk viscosity coefficients. Compared to previous results from Urca and modified Urca reactions, we conclude that at $T\sim 10^9$K phonon collisions give the leading contribution to the bulk viscosities in the core of the neutron stars, except for $n \sim 2n_0$ when the opening of the Urca processes takes place.

[ { "created": "Thu, 21 Feb 2013 22:53:09 GMT", "version": "v1" }, { "created": "Sun, 9 Jun 2013 13:17:48 GMT", "version": "v2" }, { "created": "Tue, 2 Jul 2013 14:07:25 GMT", "version": "v3" } ]

2013-07-03

[ [ "Manuel", "Cristina", "" ], [ "Tarrus", "Jaume", "" ], [ "Tolos", "Laura", "" ] ]

Thu, 21 Feb 2013 22:53:09 GMT

2013-02-21T22:53:09Z

[ "physics" ]

arxiv

null

2006.15712

Sonja Otten

Sonja Otten, Ruslan Krenzler, Hans Daduna, Karsten Kruse

Exponential single server queues in an interactive random environment

null

Stochastic Systems 13 (2023), 271-319

10.1287/stsy.2023.0106

null

math.PR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We consider exponential single server queues with state-dependent arrival and service rates which evolve under influences of external environments. The transitions of the queues are influenced by the environment's state and the movements of the environment depend on the status of the queues (bi-directional interaction). The environment is constructed in a way to encompass various models from the recent Operations Research literature, where a queue is coupled with an inventory or with reliability issues. With a Markovian joint queueing-environment process we prove separability for a large class of such interactive systems, i.e. the steady state distribution is of product form and explicitly given. The queue and the environment processes decouple asymptotically and in steady state. For non-separable systems we develop ergodicity and exponential ergodicity criteria via Lyapunov functions. By examples we explain principles for bounding departure rates of served customers (throughputs) of non-separable systems by throughputs of related separable systems as upper and lower bound.

[ { "created": "Sun, 28 Jun 2020 21:05:00 GMT", "version": "v1" }, { "created": "Fri, 17 Mar 2023 20:20:32 GMT", "version": "v2" } ]

2023-07-19

[ [ "Otten", "Sonja", "" ], [ "Krenzler", "Ruslan", "" ], [ "Daduna", "Hans", "" ], [ "Kruse", "Karsten", "" ] ]

Sun, 28 Jun 2020 21:05:00 GMT

2020-06-28T21:05:00Z

[ "math" ]

arxiv

null

0904.0939

Michael Strickland

Michael Strickland and David Yager-Elorriaga

A Parallel Algorithm for Solving the 3d Schrodinger Equation

18 pages, 7 figures; published version

J.Comput.Phys.229:6015-6026,2010

10.1016/j.jcp.2010.04.032

null

quant-ph hep-ph nucl-th physics.atom-ph physics.comp-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We describe a parallel algorithm for solving the time-independent 3d Schrodinger equation using the finite difference time domain (FDTD) method. We introduce an optimized parallelization scheme that reduces communication overhead between computational nodes. We demonstrate that the compute time, t, scales inversely with the number of computational nodes as t ~ N_nodes^(-0.95 +/- 0.04). This makes it possible to solve the 3d Schrodinger equation on extremely large spatial lattices using a small computing cluster. In addition, we present a new method for precisely determining the energy eigenvalues and wavefunctions of quantum states based on a symmetry constraint on the FDTD initial condition. Finally, we discuss the usage of multi-resolution techniques in order to speed up convergence on extremely large lattices.

[ { "created": "Mon, 6 Apr 2009 15:48:21 GMT", "version": "v1" }, { "created": "Mon, 28 Sep 2009 19:30:30 GMT", "version": "v2" }, { "created": "Thu, 21 Jan 2010 18:35:32 GMT", "version": "v3" }, { "created": "Fri, 4 Jun 2010 23:25:29 GMT", "version": "v4" } ]

2014-11-18

[ [ "Strickland", "Michael", "" ], [ "Yager-Elorriaga", "David", "" ] ]

Mon, 6 Apr 2009 15:48:21 GMT

2009-04-06T15:48:21Z

[ "physics" ]

arxiv

null

2109.06633

Tommaso Lorenzi

Tommaso Lorenzi and Kevin J Painter

Trade-offs between chemotaxis and proliferation shape the phenotypic structuring of invading waves

35

null

10.1016/j.ijnonlinmec.2021.103885

null

q-bio.PE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Chemotaxis-driven invasions have been proposed across a broad spectrum of biological processes, from cancer to ecology. The influential system of equations introduced by Keller and Segel has proven a popular choice in the modelling of such phenomena, but in its original form restricts to a homogeneous population. To account for the possibility of phenotypic heterogeneity, we extend to the case of a population continuously structured across space, time and phenotype, where the latter determines variation in chemotactic responsiveness, proliferation rate, and the level of chemical environment modulation. The extended model considered here comprises a non-local partial differential equation for the local phenotype distribution of cells which is coupled, through an integral term, with a differential equation for the concentration of an attractant, which is sensed and degraded by the cells. In the framework of this model, we concentrate on a chemotaxis/proliferation trade-off scenario, where the cell phenotypes span a spectrum of states from highly-chemotactic but minimally-proliferative to minimally-chemotactic but highly-proliferative. Using a combination of numerical simulation and formal asymptotic analysis, we explore the properties of travelling-wave solutions. The results of our study demonstrate how incorporating phenotypic heterogeneity may lead to a highly-structured wave profile, where cells in different phenotypic states dominate different spatial positions across the invading wave, and clarify how the phenotypic structuring of the wave can be shaped by trade-offs between chemotaxis and proliferation.

[ { "created": "Wed, 8 Sep 2021 20:52:57 GMT", "version": "v1" }, { "created": "Mon, 20 Sep 2021 14:44:12 GMT", "version": "v2" }, { "created": "Thu, 25 Nov 2021 07:30:40 GMT", "version": "v3" } ]

2022-02-02

[ [ "Lorenzi", "Tommaso", "" ], [ "Painter", "Kevin J", "" ] ]

Wed, 8 Sep 2021 20:52:57 GMT

2021-09-08T20:52:57Z

[ "q-bio" ]

arxiv

null

nlin/0408022

Oleg Yevtushenko

S.M. Soskin, O.M. Yevtushenko, and R. Mannella

Divergence of the Chaotic Layer Width and Strong Acceleration of the Spatial Chaotic Transport in Periodic Systems Driven by an Adiabatic ac Force

4 pages including 3 EPS figures, this is an improved version of the paper (accepted to PRL, 2005)

Phys. Rev. Lett. 95, 224101 (2005)

10.1103/PhysRevLett.95.224101

null

nlin.CD cond-mat.mes-hall cond-mat.stat-mech

null

We show for the first time that a {\it weak} perturbation in a Hamiltonian system may lead to an arbitrarily {\it wide} chaotic layer and {\it fast} chaotic transport. This {\it generic} effect occurs in any spatially periodic Hamiltonian system subject to a sufficiently slow ac force. We explain it and develop an explicit theory for the layer width, verified in simulations. Chaotic spatial transport as well as applications to the diffusion of particles on surfaces, threshold devices and others are discussed.

[ { "created": "Wed, 11 Aug 2004 16:42:27 GMT", "version": "v1" }, { "created": "Wed, 12 Oct 2005 10:48:04 GMT", "version": "v2" } ]

2007-05-23

[ [ "Soskin", "S. M.", "" ], [ "Yevtushenko", "O. M.", "" ], [ "Mannella", "R.", "" ] ]

Wed, 11 Aug 2004 16:42:27 GMT

2004-08-11T16:42:27Z

[ "physics", "nlin" ]

arxiv

null

math/0612235

Antongiulio Fornasiero

Antongiulio Fornasiero and Marcello Mamino

Arithmetic of Dedekind cuts of ordered Abelian groups

null

Annals of Pure and Applied Logic 156 (2008) 210-244

10.1016/j.apal.2008.05.001

null

math.LO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the set of Dedekind cuts over a linearly ordered Abelian group as a structure over the language (0,<,+,-). Moreover, we obtain a simple set of axioms for the universal part of the theory of such structures. Finally, we prove that every structure satisfying the given axioms is a sub-structure of the set of cuts over a suitable group.

[ { "created": "Sun, 10 Dec 2006 16:22:28 GMT", "version": "v1" }, { "created": "Wed, 13 Jun 2007 10:46:36 GMT", "version": "v2" }, { "created": "Tue, 16 Dec 2008 13:49:09 GMT", "version": "v3" } ]

2008-12-16

[ [ "Fornasiero", "Antongiulio", "" ], [ "Mamino", "Marcello", "" ] ]

Sun, 10 Dec 2006 16:22:28 GMT

2006-12-10T16:22:28Z

[ "math" ]

arxiv

null

A technical note concerning emergency bed demand.

null

A common method for modelling emergency referrals to a hospital in-patient unit is to assume that they follow a Poisson distribution. If lengths of stay are independent and identically distributed, then the distribution of bed demand can be derived by analysis of a queueing system with an infinite number of servers. Undemanding analysis can be used to express the mean and variance of bed demand in terms of simple formulae involving mean referral rates and the distribution of lengths of stay.

null

pubmed

null

2,011

null

21,559,850

Health care management science

0910.2203

Dariush Heidarian

Dariush Heidarian and Arun Paramekanti

Supersolidity in the triangular lattice spin-1/2 XXZ model: A variational perspective

4 pages, 4 figures, 1 table, published version

Phys. Rev. Lett. 104, 015301 (2010)

10.1103/PhysRevLett.104.015301

null

cond-mat.str-el

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the spin-1/2 XXZ model on the triangular lattice with a nearest neighbor antiferromagnetic Ising coupling $J_z>0$ and unfrustrated ($J_\perp<0$) or frustrated ($J_\perp>0$) kinetic terms in zero magnetic field. Incorporating long-range Jastrow correlations over a mean field spin state, we obtain the variational phase diagram of this model on large lattices for arbitrary $J_z$ and either sign of $J_\perp$. For $J_\perp<0$, we find a $\sqrt{3}\times\sqrt{3}$ supersolid for $J_z/|J_\perp| \gtrsim 4.7$, in excellent agreement with quantum Monte Carlo data. For $J_\perp >0$, a distinct $\sqrt{3}\times\sqrt{3}$ supersolid is found to emerge for $J_z/J_\perp \geq 1$. Both supersolids exhibit a spontaneous density deviation from half-filling. At $J_z/J_\perp=\infty$, the crystalline order parameters of these two supersolids are nearly identical, consistent with exact results.

[ { "created": "Mon, 12 Oct 2009 17:17:49 GMT", "version": "v1" }, { "created": "Tue, 5 Jan 2010 17:38:46 GMT", "version": "v2" } ]

2015-05-14

[ [ "Heidarian", "Dariush", "" ], [ "Paramekanti", "Arun", "" ] ]

Mon, 12 Oct 2009 17:17:49 GMT

2009-10-12T17:17:49Z

[ "physics" ]

arxiv

null

1612.00168

Adam Martin-Schwarze

Adam Martin-Schwarze, Jarad Niemi, and Philip Dixon

Assessing the impacts of time to detection distribution assumptions on detection probability estimation

Expands previous simulation study and adds simulations across a range of `true' detection values

JABES (2017) 22: 465-480

10.1007/s13253-017-0300-y

null

q-bio.PE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Abundance estimates from animal point-count surveys require accurate estimates of detection probabilities. The standard model for estimating detection from removal-sampled point-count surveys assumes that organisms at a survey site are detected at a constant rate; however, this assumption is often not justified. We consider a class of N-mixture models that allows for detection heterogeneity over time through a flexibly defined time-to-detection distribution (TTDD) and allows for fixed and random effects for both abundance and detection. Our model is thus a combination of survival time-to-event analysis with unknown-N, unknown-p abundance estimation. We specifically explore two-parameter families of TTDDs, e.g. gamma, that can additionally include a mixture component to model increased probability of detection in the initial observation period. We find that modeling a TTDD by using a two-parameter family is necessary when data have a chance of arising from a distribution of this nature. In addition, models with a mixture component can outperform non-mixture models even when the truth is non-mixture. Finally, we analyze an Overbird data set from the Chippewa National Forest using mixed effect models for both abundance and detection. We demonstrate that the effects of explanatory variables on abundance and detection are consistent across mixture TTDDs but that flexible TTDDs result in lower estimated probabilities of detection and therefore higher estimates of abundance.

[ { "created": "Thu, 1 Dec 2016 08:04:07 GMT", "version": "v1" }, { "created": "Fri, 19 May 2017 19:05:13 GMT", "version": "v2" } ]

2019-04-09

[ [ "Martin-Schwarze", "Adam", "" ], [ "Niemi", "Jarad", "" ], [ "Dixon", "Philip", "" ] ]

Thu, 1 Dec 2016 08:04:07 GMT

2016-12-01T08:04:07Z

[ "q-bio" ]

arxiv

null

1905.05075

Haji M. Furqan Madni

Haji M. Furqan, Muhammad Sohaib J. Solaija, Jehad M. Hamamreh, and Huseyin Arslan

Intelligent Physical Layer Security Approach for V2X Communication

null

COMMAG-19-00169

eess.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Intelligent transportation systems (ITS) with advanced sensing and computing technologies are expected to support a whole new set of services including pedestrian and vehicular safety, internet access for vehicles, and eventually, driverless cars. Wireless communication is a major driving factor behind ITS, enabling reliable communication between vehicles, infrastructure, pedestrians and network, generally referred to as vehicle to everything (V2X) communication. However, the broadcast nature of wireless communication renders it prone to jamming, eavesdropping and spoofing attacks which can adversely affect ITS. Keeping in view this issue, we suggest the use of an intelligent security framework for V2X communication security, referred to as intelligent V2X security (IV2XS), to provide a reliable and robust solution capable of adapting to different conditions, scenarios and user requirements. We also identify the conditions that impact the security and describe the open challenges in achieving a realistic IV2XS system.

[ { "created": "Mon, 13 May 2019 15:08:00 GMT", "version": "v1" }, { "created": "Sun, 12 Jul 2020 12:55:54 GMT", "version": "v2" }, { "created": "Sat, 8 Aug 2020 12:04:58 GMT", "version": "v3" } ]

2020-08-11

[ [ "Furqan", "Haji M.", "" ], [ "Solaija", "Muhammad Sohaib J.", "" ], [ "Hamamreh", "Jehad M.", "" ], [ "Arslan", "Huseyin", "" ] ]

Mon, 13 May 2019 15:08:00 GMT

2019-05-13T15:08:00Z

[ "eess" ]

arxiv

null

2001.08431

Thomas Yee

Thomas William Yee

On the Hauck-Donner Effect in Wald Tests: Detection, Tipping Points, and Parameter Space Characterization

6 figures

JASA 2022

10.1080/01621459.2021.1886936

null

stat.ME math.ST stat.CO stat.TH

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The Wald test remains ubiquitous in statistical practice despite shortcomings such as its inaccuracy in small samples and lack of invariance under reparameterization. This paper develops on another but lesser-known shortcoming called the Hauck--Donner effect (HDE) whereby a Wald test statistic is not monotonely increasing as a function of increasing distance between the parameter estimate and the null value. Resulting in an upward biased $p$-value and loss of power, the aberration can lead to very damaging consequences such as in variable selection. The HDE afflicts many types of regression models and corresponds to estimates near the boundary of the parameter space. This article presents several new results, and its main contributions are to (i) propose a very general test for detecting the HDE, regardless of its underlying cause; (ii) fundamentally characterize the HDE by pairwise ratios of Wald and Rao score and likelihood ratio test statistics for 1-parameter distributions; (iii) show that the parameter space may be partitioned into an interior encased by 5 HDE severity measures (faint, weak, moderate, strong, extreme); (iv) prove that a necessary condition for the HDE in a 2 by 2 table is a log odds ratio of at least 2; (v) give some practical guidelines about HDE-free hypothesis testing. Overall, practical post-fit tests can now be conducted potentially to any model estimated by iteratively reweighted least squares, such as the generalized linear model (GLM) and Vector GLM (VGLM) classes, the latter which encompasses many popular regression models.

[ { "created": "Thu, 23 Jan 2020 10:15:50 GMT", "version": "v1" }, { "created": "Wed, 14 Sep 2022 01:03:56 GMT", "version": "v2" } ]

2022-09-15

[ [ "Yee", "Thomas William", "" ] ]

Thu, 23 Jan 2020 10:15:50 GMT

2020-01-23T10:15:50Z

[ "math", "stat" ]

arxiv

null

2112.09227

Uzi Pereg

Uzi Pereg, Christian Deppe, and Holger Boche

Communication with Unreliable Entanglement Assistance

null

quant-ph cs.IT math.IT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Entanglement resources can increase transmission rates substantially. Unfortunately, entanglement is a fragile resource that is quickly degraded by decoherence effects. In order to generate entanglement for optical communication, the transmitter and the receiver first prepare entangled spin-photon pairs locally, and then the photon at the transmitter is sent to the receiver through an optical fiber or free space. Without feedback, the transmitter does not know whether the entangled photon has reached the receiver. The present work introduces a new model of unreliable entanglement assistance, whereby the communication system operates whether entanglement assistance is present or not. While the sender is ignorant, the receiver knows whether the entanglement generation was successful. In the case of a failure, the receiver decodes less information. In this manner, the effective transmission rate is adapted according to the assistance status. Regularized formulas are derived for the classical and quantum capacity regions with unreliable entanglement assistance, characterizing the tradeoff between the unassisted rate and the excess rate that can be obtained from entanglement assistance. It is further established that time division between entanglement-assisted and unassisted coding strategies is optimal for the noiseless qubit channel, but can be strictly suboptimal for a noisy channel.

[ { "created": "Thu, 16 Dec 2021 22:24:56 GMT", "version": "v1" }, { "created": "Fri, 18 Feb 2022 22:45:53 GMT", "version": "v2" }, { "created": "Mon, 9 May 2022 10:41:12 GMT", "version": "v3" }, { "created": "Sat, 4 Mar 2023 12:20:05 GMT", "version": "v4" } ]

2023-03-07

[ [ "Pereg", "Uzi", "" ], [ "Deppe", "Christian", "" ], [ "Boche", "Holger", "" ] ]

Thu, 16 Dec 2021 22:24:56 GMT

2021-12-16T22:24:56Z

[ "physics", "math", "cs" ]

arxiv

null

The physician as teacher. Epistemic function, cognitive function and the incommensurability of errors.

null

In many Hippocratic writings, the writers' attention is often focused on registering and discussing medical errors. Far from being sporadic and fortuitous, these discussions represent a privileged rhetorical resource in order to produce different effects. The aims of my paper will be: 1) to determine some of the most important contexts in which errors become the object of medical discourse; 2) to distinguish, per exempla, the typologies of errors made object of discourse; 3) to give an epistemological outline which may clarify which functions these discourses have and whether these functions respond coherently to a conscious plan of medical knowledge.

null

pubmed

null

2,010

null

21,560,574

Studies in ancient medicine

2008.03008

Bayu Adhi Nugroho

Bayu A. Nugroho

An Aggregate Method for Thorax Diseases Classification

code available: https://github.com/bayu-ladom-ipok/weOpen

null

eess.IV cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A common problem found in real-word medical image classification is the inherent imbalance of the positive and negative patterns in the dataset where positive patterns are usually rare. Moreover, in the classification of multiple classes with neural network, a training pattern is treated as a positive pattern in one output node and negative in all the remaining output nodes. In this paper, the weights of a training pattern in the loss function are designed based not only on the number of the training patterns in the class but also on the different nodes where one of them treats this training pattern as positive and the others treat it as negative. We propose a combined approach of weights calculation algorithm for deep network training and the training optimization from the state-of-the-art deep network architecture for thorax diseases classification problem. Experimental results on the Chest X-Ray image dataset demonstrate that this new weighting scheme improves classification performances, also the training optimization from the EfficientNet improves the performance furthermore. We compare the aggregate method with several performances from the previous study of thorax diseases classifications to provide the fair comparisons against the proposed method.

[ { "created": "Fri, 7 Aug 2020 06:36:07 GMT", "version": "v1" }, { "created": "Mon, 10 Aug 2020 03:12:57 GMT", "version": "v2" }, { "created": "Wed, 12 Aug 2020 03:15:53 GMT", "version": "v3" }, { "created": "Thu, 12 Nov 2020 07:15:33 GMT", "version": "v4" }, { "created": "Thu, 24 Dec 2020 12:35:14 GMT", "version": "v5" } ]

2020-12-25

[ [ "Nugroho", "Bayu A.", "" ] ]

Fri, 7 Aug 2020 06:36:07 GMT

2020-08-07T06:36:07Z

[ "cs", "eess" ]

arxiv

null

2112.14249

Rahul Singh

Isaac Meza and Rahul Singh

Nested Nonparametric Instrumental Variable Regression: Long Term, Mediated, and Time Varying Treatment Effects

null

stat.ML cs.LG econ.EM math.ST stat.TH

http://creativecommons.org/licenses/by/4.0/

Several causal parameters in short panel data models are scalar summaries of a function called a nested nonparametric instrumental variable regression (nested NPIV). Examples include long term, mediated, and time varying treatment effects identified using proxy variables. However, it appears that no prior estimators or guarantees for nested NPIV exist, preventing flexible estimation and inference for these causal parameters. A major challenge is compounding ill posedness due to the nested inverse problems. We analyze adversarial estimators of nested NPIV, and provide sufficient conditions for efficient inference on the causal parameter. Our nonasymptotic analysis has three salient features: (i) introducing techniques that limit how ill posedness compounds; (ii) accommodating neural networks, random forests, and reproducing kernel Hilbert spaces; and (iii) extending to causal functions, e.g. long term heterogeneous treatment effects. We measure long term heterogeneous treatment effects of Project STAR and mediated proximal treatment effects of the Job Corps.

[ { "created": "Tue, 28 Dec 2021 18:29:56 GMT", "version": "v1" }, { "created": "Mon, 3 Oct 2022 23:38:58 GMT", "version": "v2" }, { "created": "Mon, 11 Mar 2024 02:12:52 GMT", "version": "v3" } ]

2024-03-12

[ [ "Meza", "Isaac", "" ], [ "Singh", "Rahul", "" ] ]

Tue, 28 Dec 2021 18:29:56 GMT

2021-12-28T18:29:56Z

[ "cs", "math", "stat", "econ" ]

arxiv

null

2009.05303

Weijian Chen

Weijian Chen, Fuli Feng, Qifan Wang, Xiangnan He, Chonggang Song, Guohui Ling, Yongdong Zhang

CatGCN: Graph Convolutional Networks with Categorical Node Features

Accepted by TKDE as a regular paper

null

10.1109/TKDE.2021.3133013

null

cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Recent studies on Graph Convolutional Networks (GCNs) reveal that the initial node representations (i.e., the node representations before the first-time graph convolution) largely affect the final model performance. However, when learning the initial representation for a node, most existing work linearly combines the embeddings of node features, without considering the interactions among the features (or feature embeddings). We argue that when the node features are categorical, e.g., in many real-world applications like user profiling and recommender system, feature interactions usually carry important signals for predictive analytics. Ignoring them will result in suboptimal initial node representation and thus weaken the effectiveness of the follow-up graph convolution. In this paper, we propose a new GCN model named CatGCN, which is tailored for graph learning when the node features are categorical. Specifically, we integrate two ways of explicit interaction modeling into the learning of initial node representation, i.e., local interaction modeling on each pair of node features and global interaction modeling on an artificial feature graph. We then refine the enhanced initial node representations with the neighborhood aggregation-based graph convolution. We train CatGCN in an end-to-end fashion and demonstrate it on semi-supervised node classification. Extensive experiments on three tasks of user profiling (the prediction of user age, city, and purchase level) from Tencent and Alibaba datasets validate the effectiveness of CatGCN, especially the positive effect of performing feature interaction modeling before graph convolution.

[ { "created": "Fri, 11 Sep 2020 09:25:17 GMT", "version": "v1" }, { "created": "Thu, 17 Sep 2020 07:41:55 GMT", "version": "v2" }, { "created": "Fri, 3 Dec 2021 11:06:46 GMT", "version": "v3" }, { "created": "Wed, 9 Mar 2022 10:46:22 GMT", "version": "v4" } ]

2022-03-10

[ [ "Chen", "Weijian", "" ], [ "Feng", "Fuli", "" ], [ "Wang", "Qifan", "" ], [ "He", "Xiangnan", "" ], [ "Song", "Chonggang", "" ], [ "Ling", "Guohui", "" ], [ "Zhang", "Yongdong", "" ] ]

Fri, 11 Sep 2020 09:25:17 GMT

2020-09-11T09:25:17Z

[ "cs", "stat" ]

arxiv

null

Proteomic profiling of yeast heterochromatin connects direct physical and genetic interactions.

null

Heterochromatin domains are stably repressed chromatin structures composed of a core assembly of silencing proteins that condense adjacent nucleosomes. The minimal heterochromatin structure can serve as a platform for recruitment of complementary regulatory factors. We find that a reconstituted budding yeast heterochromatin domain can act as a platform to recruit multiple factors that play a role in regulating heterochromatin function. We uncover the direct interaction between the SIR heterochromatin complex and a chromosomal boundary protein that restricts the spread of heterochromatin. We find that the SIR complex relieves a mechanism of auto-inhibition within the boundary protein Yta7, allowing the Yta7 bromodomain to engage chromatin. Our results suggest that budding yeast shares with other eukaryotes the ability to establish complex heterochromatin domains that coordinate multiple mechanisms of silencing regulation through physical interactions.

null

pubmed

null

2,019

null

30,310,994

Current genetics

2103.09287

Swati Gupta

Jad Salem, Swati Gupta, Vijay Kamble

Algorithmic Challenges in Ensuring Fairness at the Time of Decision

null

cs.LG cs.AI cs.CY math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Algorithmic decision-making in societal contexts, such as retail pricing, loan administration, recommendations on online platforms, etc., can be framed as stochastic optimization under bandit feedback, which typically requires experimentation with different decisions for the sake of learning. Such experimentation often results in perceptions of unfairness among people impacted by these decisions; for instance, there have been several recent lawsuits accusing companies that deploy algorithmic pricing practices of price gouging. Motivated by the changing legal landscape surrounding algorithmic decision-making, we introduce the well-studied fairness notion of envy-freeness within the context of stochastic convex optimization. Our notion requires that upon receiving decisions in the present time, groups do not envy the decisions received by any of the other groups, both in the present as well as the past. This results in a novel trajectory-constrained stochastic optimization problem that renders existing techniques inapplicable. The main technical contribution of this work is to show problem settings where there is no gap in achievable regret (up to logarithmic factors) when envy-freeness is imposed. In particular, in our main result, we develop a near-optimal envy-free algorithm that achieves $\tilde{O}(\sqrt{T})$ regret for smooth convex functions that satisfy the PL inequality. This algorithm has a coordinate-descent structure, in which we carefully leverage gradient information to ensure monotonic sampling along each dimension, while avoiding overshooting the constrained optimum with high probability. This latter aspect critically uses smoothness and the structure of the envy-freeness constraints, while the PL inequality allows for sufficient progress towards the optimal solution. We discuss several open questions that arise from this analysis, which may be of independent interest.

[ { "created": "Tue, 16 Mar 2021 19:06:28 GMT", "version": "v1" }, { "created": "Sun, 18 Sep 2022 16:06:33 GMT", "version": "v2" }, { "created": "Fri, 18 Oct 2024 21:29:47 GMT", "version": "v3" } ]

2024-10-22

[ [ "Salem", "Jad", "" ], [ "Gupta", "Swati", "" ], [ "Kamble", "Vijay", "" ] ]

Tue, 16 Mar 2021 19:06:28 GMT

2021-03-16T19:06:28Z

[ "cs", "math" ]

arxiv

null

1905.12962

Mike Gartrell

Mike Gartrell, Victor-Emmanuel Brunel, Elvis Dohmatob, Syrine Krichene

Learning Nonsymmetric Determinantal Point Processes

NeurIPS 2019

null

cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Determinantal point processes (DPPs) have attracted substantial attention as an elegant probabilistic model that captures the balance between quality and diversity within sets. DPPs are conventionally parameterized by a positive semi-definite kernel matrix, and this symmetric kernel encodes only repulsive interactions between items. These so-called symmetric DPPs have significant expressive power, and have been successfully applied to a variety of machine learning tasks, including recommendation systems, information retrieval, and automatic summarization, among many others. Efficient algorithms for learning symmetric DPPs and sampling from these models have been reasonably well studied. However, relatively little attention has been given to nonsymmetric DPPs, which relax the symmetric constraint on the kernel. Nonsymmetric DPPs allow for both repulsive and attractive item interactions, which can significantly improve modeling power, resulting in a model that may better fit for some applications. We present a method that enables a tractable algorithm, based on maximum likelihood estimation, for learning nonsymmetric DPPs from data composed of observed subsets. Our method imposes a particular decomposition of the nonsymmetric kernel that enables such tractable learning algorithms, which we analyze both theoretically and experimentally. We evaluate our model on synthetic and real-world datasets, demonstrating improved predictive performance compared to symmetric DPPs, which have previously shown strong performance on modeling tasks associated with these datasets.

[ { "created": "Thu, 30 May 2019 11:09:16 GMT", "version": "v1" }, { "created": "Thu, 5 Dec 2019 15:16:33 GMT", "version": "v2" }, { "created": "Fri, 13 Nov 2020 02:42:02 GMT", "version": "v3" } ]

2020-11-16

[ [ "Gartrell", "Mike", "" ], [ "Brunel", "Victor-Emmanuel", "" ], [ "Dohmatob", "Elvis", "" ], [ "Krichene", "Syrine", "" ] ]

Thu, 30 May 2019 11:09:16 GMT

2019-05-30T11:09:16Z

[ "cs", "stat" ]

arxiv

null

nlin/0401008

Andrei Svinin Kirillovich

A.K. Svinin

Invariant Submanifolds of Darboux-KP Chain and Extension of the Discete KP Hierarchy

Latex, 22 p., accepted for publication in Theor. Math. Phys

null

nlin.SI

null

Invariant submanifolds of the so-called Darboux-KP chain are investigated. It is shown that restriction of dynamics on some class of invariant submanifolds yields the extension of the discrete KP hierarchy, while the intersections leads to Lax pairs for a broad class of differential-difference systems with finite number of fields. Some attention is given to investigation of self-similar reductions. It is shown that self-similar ansatzes lead to purely discrete equations with dependence on some number of parameters together with equations governing deformations with respect to these parameters. Some examples are provided. In particular it is shown that well known first discrete Painleve equation (dPI) corresponds to Volterra lattice hierarchy. It is written down equations which naturally generalize dPI in the sense that they have first Painleve transcedent in continuous limit.

[ { "created": "Thu, 8 Jan 2004 03:44:20 GMT", "version": "v1" }, { "created": "Thu, 22 Apr 2004 12:33:33 GMT", "version": "v2" }, { "created": "Wed, 15 Sep 2004 00:18:44 GMT", "version": "v3" } ]

2007-05-23

[ [ "Svinin", "A. K.", "" ] ]

Thu, 8 Jan 2004 03:44:20 GMT

2004-01-08T03:44:20Z

[ "nlin" ]

arxiv

null

2105.07451

Debesh Jha

Abhishek Srivastava, Debesh Jha, Sukalpa Chanda, Umapada Pal, H{\aa}vard D. Johansen, Dag Johansen, Michael A. Riegler, Sharib Ali, P{\aa}l Halvorsen

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

null

IEEE Journal of Biomedical and Health Informatics, 2022

null

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

Methods based on convolutional neural networks have improved the performance of biomedical image segmentation. However, most of these methods cannot efficiently segment objects of variable sizes and train on small and biased datasets, which are common for biomedical use cases. While methods exist that incorporate multi-scale fusion approaches to address the challenges arising with variable sizes, they usually use complex models that are more suitable for general semantic segmentation problems. In this paper, we propose a novel architecture called Multi-Scale Residual Fusion Network (MSRF-Net), which is specially designed for medical image segmentation. The proposed MSRF-Net is able to exchange multi-scale features of varying receptive fields using a Dual-Scale Dense Fusion (DSDF) block. Our DSDF block can exchange information rigorously across two different resolution scales, and our MSRF sub-network uses multiple DSDF blocks in sequence to perform multi-scale fusion. This allows the preservation of resolution, improved information flow and propagation of both high- and low-level features to obtain accurate segmentation maps. The proposed MSRF-Net allows to capture object variabilities and provides improved results on different biomedical datasets. Extensive experiments on MSRF-Net demonstrate that the proposed method outperforms the cutting-edge medical image segmentation methods on four publicly available datasets. We achieve the dice coefficient of 0.9217, 0.9420, and 0.9224, 0.8824 on Kvasir-SEG, CVC-ClinicDB, 2018 Data Science Bowl dataset, and ISIC-2018 skin lesion segmentation challenge dataset respectively. We further conducted generalizability tests and achieved a dice coefficient of 0.7921 and 0.7575 on CVC-ClinicDB and Kvasir-SEG, respectively.

[ { "created": "Sun, 16 May 2021 15:19:56 GMT", "version": "v1" }, { "created": "Mon, 31 Jan 2022 03:56:04 GMT", "version": "v2" } ]

2022-02-01

[ [ "Srivastava", "Abhishek", "" ], [ "Jha", "Debesh", "" ], [ "Chanda", "Sukalpa", "" ], [ "Pal", "Umapada", "" ], [ "Johansen", "Håvard D.", "" ], [ "Johansen", "Dag", "" ], [ "Riegler", "Michael A.", "" ], [ "Ali", "Sharib", "" ], [ "Halvorsen", "Pål", "" ] ]

Sun, 16 May 2021 15:19:56 GMT

2021-05-16T15:19:56Z

[ "cs", "eess" ]

arxiv

null

1205.1712

Naresh Sharma

Naresh Sharma and Naqueeb Ahmad Warsi

On the strong converses for the quantum channel capacity theorems

Added the strong converse for the erasure channel for maximally entangled inputs and corrected minor typos

null

10.1103/PhysRevLett.110.080501

null

quant-ph cs.IT math.IT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A unified approach to prove the converses for the quantum channel capacity theorems is presented. These converses include the strong converse theorems for classical or quantum information transfer with error exponents and novel explicit upper bounds on the fidelity measures reminiscent of the Wolfowitz strong converse for the classical channel capacity theorems. We provide a new proof for the error exponents for the classical information transfer. A long standing problem in quantum information theory has been to find out the strong converse for the channel capacity theorem when quantum information is sent across the channel. We give the quantum error exponent thereby giving a one-shot exponential upper bound on the fidelity. We then apply our results to show that the strong converse holds for the quantum information transfer across an erasure channel for maximally entangled channel inputs.

[ { "created": "Tue, 8 May 2012 14:42:01 GMT", "version": "v1" }, { "created": "Wed, 9 May 2012 19:53:08 GMT", "version": "v2" }, { "created": "Wed, 20 Jun 2012 17:16:55 GMT", "version": "v3" } ]

2013-03-14

[ [ "Sharma", "Naresh", "" ], [ "Warsi", "Naqueeb Ahmad", "" ] ]

Tue, 8 May 2012 14:42:01 GMT

2012-05-08T14:42:01Z

[ "physics", "math", "cs" ]

arxiv

null

1603.06050

Philip Ernst

Philip Ernst, James Thompson, and Yinsen Miao

Tukey's transformational ladder for portfolio management

32 pages

Financial Markets and Portfolio Management (2017) 31(3): 317-355

null

q-fin.PM

http://creativecommons.org/licenses/by-nc-sa/4.0/

Over the past half-century, the empirical finance community has produced vast literature on the advantages of the equally weighted S\&P 500 portfolio as well as the often overlooked disadvantages of the market capitalization weighted Standard and Poor's (S\&P 500) portfolio (see \cite{Bloom}, \cite{Uppal}, \cite{Jacobs}, \cite{Treynor}). However, portfolio allocation based on Tukey's transformational ladde have, rather surprisingly, remained absent from the literature. In this work, we consider the S\&P 500 portfolio over the 1958-2015 time horizon weighted by Tukey's transformational ladder (\cite{Tukey2}): $1/x^2,\,\, 1/x,\,\, 1/\sqrt{x},\,\, \text{log}(x),\,\, \sqrt{x},\,\, x,\,\, \text{and} \,\, x^2$, where $x$ is defined as the market capitalization weighted S\&P 500 portfolio. Accounting for dividends and transaction fees, we find that the 1/$x^2$ weighting strategy produces cumulative returns that significantly dominates all other portfolios, achieving a compound annual growth rate of 18\% over the 1958-2015 horizon. Our story is furthered by a startling phenomenon: both the cumulative and annual returns of the $1/x^2$ weighting strategy are superior to those of the $1/x$ weighting strategy, which are in turn superior to those of the 1/$\sqrt{x}$ weighted portfolio, and so forth, ending with the $x^2$ transformation, whose cumulative returns are the lowest of the seven transformations of Tukey's transformational ladder. The order of cumulative returns precisely follows that of Tukey's transformational ladder. To the best of our knowledge, we are the first to discover this phenomenon.

[ { "created": "Sat, 19 Mar 2016 06:08:53 GMT", "version": "v1" }, { "created": "Tue, 3 May 2016 08:57:31 GMT", "version": "v2" }, { "created": "Mon, 23 May 2016 20:20:24 GMT", "version": "v3" }, { "created": "Sat, 22 Apr 2017 04:37:01 GMT", "version": "v4" }, { "created": "Sun, 6 Aug 2017 16:19:30 GMT", "version": "v5" } ]

2017-08-08

[ [ "Ernst", "Philip", "" ], [ "Thompson", "James", "" ], [ "Miao", "Yinsen", "" ] ]

Sat, 19 Mar 2016 06:08:53 GMT

2016-03-19T06:08:53Z

[ "q-fin" ]

arxiv

null

1105.1612

Kenji Kajiwara

Hisashi Ando, Mike Hay, Kenji Kajiwara and Tetsu Masuda

An explicit formula for the discrete power function associated with circle patterns of Schramm type

33 pages. Change from the previous version: Section 5 has been added where the immersion property of the discrete power function is established when the real part of the exponent is equal to one

null

nlin.SI math.DG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present an explicit formula for the discrete power function introduced by Bobenko, which is expressed in terms of the hypergeometric \tau functions for the sixth Painlev\'e equation. The original definition of the discrete power function imposes strict conditions on the domain and the value of the exponent. However, we show that one can extend the value of the exponent to arbitrary complex numbers except even integers and the domain to a discrete analogue of the Riemann surface. Moreover, we show that the discrete power function is an immersion when the real part of the exponent is equal to one.

[ { "created": "Mon, 9 May 2011 10:02:41 GMT", "version": "v1" }, { "created": "Tue, 10 May 2011 01:17:31 GMT", "version": "v2" }, { "created": "Thu, 26 Jan 2012 12:24:07 GMT", "version": "v3" } ]

2012-01-27

[ [ "Ando", "Hisashi", "" ], [ "Hay", "Mike", "" ], [ "Kajiwara", "Kenji", "" ], [ "Masuda", "Tetsu", "" ] ]

Mon, 9 May 2011 10:02:41 GMT

2011-05-09T10:02:41Z

[ "nlin", "math" ]

arxiv

null

Molecular detection and phylogenetic analysis of canine tick-borne pathogens from Korea.

null

Ticks transmit more pathogens than any other arthropod vector and are of paramount veterinary and zoonotic significance. Domestic dogs are an important reservoir of zoonotic agents, particularly because of their close contact with humans. Previous studies in Korea have examined canine tick-borne pathogens (CTBP) by ELISA and real-time PCR methods. However, phylogenetic information on CTBP in Korea is lacking. This study assessed the prevalence, risk factors, and co-infectivity of CTBP such as piroplasms, rickettsiae, Coxiella burnetii, hepatozoa, hemotropic mycoplasmas, and Borrelia spp., using PCR and phylogenetic analyses. Of the 510 dogs tested, three CTBP, including Anaplasma phagocytophilum 16S rRNA (13; 2.6 %), Hepatozoon canis 18S rRNA (2; 0.4 %), and Mycoplasma haemocanis 16S rRNA (1; 0.2 %) were detected. PCR adapted to amplify A. phagocytophilum msp2 and groEL genes generated amplicons thereof in two out of 13 positive dogs. One shelter dog was co-infected with A. phagocytophilum and M. haemocanis. However, Ehrlichia spp., Rickettsia spp., C. burnetii, Borrelia spp., Babesia spp., and Theileria spp. pathogens were not detected. To the best of our knowledge, this is the first study to undertake a phylogenetic analysis of H. canis and M. haemocanis in dogs reared in Korea. Although previous studies have improved our understanding of evolutionary behaviors and host-pathogen relations of CTBP, additional investigations are required to pin down vectors and reservoirs of CTBP in Korea. A surveillance system for arthropod vectors and CTBP in dogs should be established to monitor pathogen distribution and mitigate pathogen spread proactively, such as with ectoparasite medications in dogs.

null

pubmed

null

2,020

null

31,839,473

Ticks and tick-borne diseases

1812.06175

Yaodong Yang Mr.

Yaodong Yang, Alisa Kolesnikova, Stefan Lessmann, Tiejun Ma, Ming-Chien Sung, Johnnie E.V. Johnson

Can Deep Learning Predict Risky Retail Investors? A Case Study in Financial Risk Behavior Forecasting

Within the "equal" contribution, Yaodong Yang contributed the core deep learning algorithm along with its experimental results, and the first draft of the manuscript (including Figure 1,2,3,4,7,8,9,11, and Table 3)

null

q-fin.RM cs.LG stat.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The paper examines the potential of deep learning to support decisions in financial risk management. We develop a deep learning model for predicting whether individual spread traders secure profits from future trades. This task embodies typical modeling challenges faced in risk and behavior forecasting. Conventional machine learning requires data that is representative of the feature-target relationship and relies on the often costly development, maintenance, and revision of handcrafted features. Consequently, modeling highly variable, heterogeneous patterns such as trader behavior is challenging. Deep learning promises a remedy. Learning hierarchical distributed representations of the data in an automatic manner (e.g. risk taking behavior), it uncovers generative features that determine the target (e.g., trader's profitability), avoids manual feature engineering, and is more robust toward change (e.g. dynamic market conditions). The results of employing a deep network for operational risk forecasting confirm the feature learning capability of deep learning, provide guidance on designing a suitable network architecture and demonstrate the superiority of deep learning over machine learning and rule-based benchmarks.

[ { "created": "Fri, 14 Dec 2018 21:31:41 GMT", "version": "v1" }, { "created": "Thu, 17 Jan 2019 16:31:04 GMT", "version": "v2" }, { "created": "Sun, 17 Nov 2019 22:59:23 GMT", "version": "v3" } ]

2019-11-19

[ [ "Yang", "Yaodong", "" ], [ "Kolesnikova", "Alisa", "" ], [ "Lessmann", "Stefan", "" ], [ "Ma", "Tiejun", "" ], [ "Sung", "Ming-Chien", "" ], [ "Johnson", "Johnnie E. V.", "" ] ]

Fri, 14 Dec 2018 21:31:41 GMT

2018-12-14T21:31:41Z

[ "cs", "stat", "q-fin" ]

arxiv

null

1709.02892

Qun He

Qun He, Yali Chen, Songting Yin and Tingting Ren

Isoparametric hypersurfaces in Finsler space forms

22 pages, comments are welcome!

null

math.DG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we study isoparametric hypersurfaces in Finsler space forms by investigating focal points, tubes and parallel hypersurfaces of submanifolds. We prove that the focal submanifolds of isoparametric hypersurfaces are anisotropic-minimal and obtain the Cartan-type formula in a Finsler space form with vanishing reversible torsion, from which we give some classifications on the number of distinct principal curvatures or their multiplicities.

[ { "created": "Sat, 9 Sep 2017 01:43:12 GMT", "version": "v1" }, { "created": "Tue, 22 Jan 2019 05:20:50 GMT", "version": "v2" }, { "created": "Fri, 27 Mar 2020 11:57:36 GMT", "version": "v3" } ]

2020-03-30

[ [ "He", "Qun", "" ], [ "Chen", "Yali", "" ], [ "Yin", "Songting", "" ], [ "Ren", "Tingting", "" ] ]

Sat, 9 Sep 2017 01:43:12 GMT

2017-09-09T01:43:12Z

[ "math" ]

arxiv

null

1701.06482

Lena Maier-Hein

Lena Maier-Hein, Swaroop Vedula, Stefanie Speidel, Nassir Navab, Ron Kikinis, Adrian Park, Matthias Eisenmann, Hubertus Feussner, Germain Forestier, Stamatia Giannarou, Makoto Hashizume, Darko Katic, Hannes Kenngott, Michael Kranzfelder, Anand Malpani, Keno M\"arz, Thomas Neumuth, Nicolas Padoy, Carla Pugh, Nicolai Schoch, Danail Stoyanov, Russell Taylor, Martin Wagner, Gregory D. Hager, Pierre Jannin

Surgical Data Science: Enabling Next-Generation Surgery

10 pages, 2 figures, White paper corresponding to http://www.surgical-data-science.org/workshop2016

Nature Biomedical Engineering 2017

10.1038/s41551-017-0132-7

null

cs.CY

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper introduces Surgical Data Science as an emerging scientific discipline. Key perspectives are based on discussions during an intensive two-day international interactive workshop that brought together leading researchers working in the related field of computer and robot assisted interventions. Our consensus opinion is that increasing access to large amounts of complex data, at scale, throughout the patient care process, complemented by advances in data science and machine learning techniques, has set the stage for a new generation of analytics that will support decision-making and quality improvement in interventional medicine. In this article, we provide a consensus definition for Surgical Data Science, identify associated challenges and opportunities and provide a roadmap for advancing the field.

[ { "created": "Mon, 23 Jan 2017 16:30:52 GMT", "version": "v1" }, { "created": "Tue, 31 Jan 2017 10:50:05 GMT", "version": "v2" } ]

2018-05-17

[ [ "Maier-Hein", "Lena", "" ], [ "Vedula", "Swaroop", "" ], [ "Speidel", "Stefanie", "" ], [ "Navab", "Nassir", "" ], [ "Kikinis", "Ron", "" ], [ "Park", "Adrian", "" ], [ "Eisenmann", "Matthias", "" ], [ "Feussner", "Hubertus", "" ], [ "Forestier", "Germain", "" ], [ "Giannarou", "Stamatia", "" ], [ "Hashizume", "Makoto", "" ], [ "Katic", "Darko", "" ], [ "Kenngott", "Hannes", "" ], [ "Kranzfelder", "Michael", "" ], [ "Malpani", "Anand", "" ], [ "März", "Keno", "" ], [ "Neumuth", "Thomas", "" ], [ "Padoy", "Nicolas", "" ], [ "Pugh", "Carla", "" ], [ "Schoch", "Nicolai", "" ], [ "Stoyanov", "Danail", "" ], [ "Taylor", "Russell", "" ], [ "Wagner", "Martin", "" ], [ "Hager", "Gregory D.", "" ], [ "Jannin", "Pierre", "" ] ]

Mon, 23 Jan 2017 16:30:52 GMT

2017-01-23T16:30:52Z

[ "cs" ]

arxiv

null

0812.1931

John March-Russell

John March-Russell, Christopher McCabe, Matthew McCullough

Inelastic Dark Matter, Non-Standard Halos and the DAMA/LIBRA Results

22 pages, 10 figures, PDFLaTex Additional analysis of Via Lactea simulation included

JHEP 0905:071,2009

10.1088/1126-6708/2009/05/071

OUTP-08 19 P

astro-ph hep-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The DAMA collaboration have claimed to detect particle dark matter (DM) via an annual modulation in their observed recoil event rate. This appears to be in strong disagreement with the null results of other experiments if interpreted in terms of elastic DM scattering, while agreement for a small region of parameter space is possible for inelastic DM (iDM) due to the altered kinematics of the collision. To date most analyses assume a simple galactic halo DM velocity distribution, the Standard Halo Model, but direct experimental support for the SHM is severely lacking and theoretical studies indicate possible significant differences. We investigate the dependence of DAMA and the other direct detection experiments on the local DM velocity distribution, utilizing the results of the Via Lactea and Dark Disc numerical simulations. We also investigate effects of varying the solar circular velocity, the DM escape velocity, and the DAMA quenching factor within experimental limits. Our data set includes the latest ZEPLIN-III results, as well as full publicly available data sets. Due to the more sensitive dependence of the inelastic cross section on the velocity distribution, we find that with Via Lactea the DAMA results can be consistent with all other experiments over an enlarged region of iDM parameter space, with higher mass particles being preferred, while Dark Disc does not lead to an improvement. A definitive test of DAMA for iDM requires heavy element detectors.

[ { "created": "Wed, 10 Dec 2008 15:11:55 GMT", "version": "v1" }, { "created": "Fri, 12 Dec 2008 17:29:29 GMT", "version": "v2" }, { "created": "Fri, 6 Mar 2009 16:18:17 GMT", "version": "v3" } ]

2014-11-18

[ [ "March-Russell", "John", "" ], [ "McCabe", "Christopher", "" ], [ "McCullough", "Matthew", "" ] ]

Wed, 10 Dec 2008 15:11:55 GMT

2008-12-10T15:11:55Z

[ "physics" ]

arxiv

null

2208.03895

Hanwen Du

Hanwen Du, Hui Shi, Pengpeng Zhao, Deqing Wang, Victor S.Sheng, Yanchi Liu, Guanfeng Liu and Lei Zhao

Contrastive Learning with Bidirectional Transformers for Sequential Recommendation

Accepted by CIKM 2022

null

cs.IR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Contrastive learning with Transformer-based sequence encoder has gained predominance for sequential recommendation. It maximizes the agreements between paired sequence augmentations that share similar semantics. However, existing contrastive learning approaches in sequential recommendation mainly center upon left-to-right unidirectional Transformers as base encoders, which are suboptimal for sequential recommendation because user behaviors may not be a rigid left-to-right sequence. To tackle that, we propose a novel framework named \textbf{C}ontrastive learning with \textbf{Bi}directional \textbf{T}ransformers for sequential recommendation (\textbf{CBiT}). Specifically, we first apply the slide window technique for long user sequences in bidirectional Transformers, which allows for a more fine-grained division of user sequences. Then we combine the cloze task mask and the dropout mask to generate high-quality positive samples and perform multi-pair contrastive learning, which demonstrates better performance and adaptability compared with the normal one-pair contrastive learning. Moreover, we introduce a novel dynamic loss reweighting strategy to balance between the cloze task loss and the contrastive loss. Experiment results on three public benchmark datasets show that our model outperforms state-of-the-art models for sequential recommendation.

[ { "created": "Mon, 8 Aug 2022 03:33:06 GMT", "version": "v1" }, { "created": "Sun, 14 Aug 2022 02:52:08 GMT", "version": "v2" }, { "created": "Wed, 17 Aug 2022 10:16:59 GMT", "version": "v3" } ]

2022-08-18

[ [ "Du", "Hanwen", "" ], [ "Shi", "Hui", "" ], [ "Zhao", "Pengpeng", "" ], [ "Wang", "Deqing", "" ], [ "Sheng", "Victor S.", "" ], [ "Liu", "Yanchi", "" ], [ "Liu", "Guanfeng", "" ], [ "Zhao", "Lei", "" ] ]

Mon, 8 Aug 2022 03:33:06 GMT

2022-08-08T03:33:06Z

[ "cs" ]

arxiv

null

Molecular Surveillance of Multidrug-Resistant Acinetobacter baumannii.

null

Acinetobacter baumannii, a bacterial strain which demonstrates an elevated wide range multidrug resistance to commonly prescribed antibiotics, has been linked to recent major global outbreaks, raising a major clinical concern. Its reduced antibiotic susceptibility is closely related to the acquisition of a potent carbapenemase and/or intrinsic gene "over expression" through insertion sequences. Hence, this study aimed at investigating the antimicrobial susceptibility and molecular mechanisms underlying β-lactam resistance in A. baumannii, isolated at an academic medical centre. To understand the basis of resistance, 103 multidrug-resistant (MDR) A. baumannii isolates were collected, their antibiotic susceptibility was tested phenotypically, and then molecular analyses were performed, by testing a range of commonly encountered carbapenemases-OXA-51, OXA-23, NDM, VIM, and KPC. All strains demonstrated pan-resistance to most of the advanced antibiotics tested, including piperacillin/tazobactam, ceftazidime, cefepime, and ciprofloxacin. Moreover, majority of isolates exhibited resistance to imipenem (98.1%) and trimethoprim (90.3%). Approximately 50% of the strains showed meropenem, amikacin, and gentamycin resistance; however, lower resistance rate to tigecycline (4.9%) was noted. Moreover, isolates contained potent carbapenemases such as the intrinsic OXA-51 (89.3%), as well as the acquired resistant genes OXA-23 (68.9%), NDM (84.5%), and VIM (88.3%). The insertion sequence element ISAba1 was only detected in 35.9% of the strains. Potent resistant genes known to be carried on mobile genetic elements that aid the spread of highly resistant phenotypes were observed in a majority of isolates. These findings enforce the need for vigilant infection control measures and continuous surveillance.

null

pubmed

null

2,020

null

31,832,843

Current microbiology

2004.01978

Darius Saif

Darius Saif and Chung-Horng Lung and Ashraf Matrawy

An Early Benchmark of Quality of Experience Between HTTP/2 and HTTP/3 using Lighthouse

null

10.1109/ICC42927.2021.9500258

null

cs.NI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Google's QUIC (GQUIC) is an emerging transport protocol designed to reduce HTTP latency. Deployed across its platforms and positioned as an alternative to TCP+TLS, GQUIC is feature rich: offering reliable data transmission and secure communication. It addresses TCP+TLS's (i) Head of Line Blocking (HoLB), (ii) excessive round-trip times on connection establishment, and (iii) entrenchment. Efforts by the IETF are in progress to standardize the next generation of HTTP's (HTTP/3, or H3) delivery, with their own variant of QUIC. While performance benchmarks have been conducted between GQUIC and HTTP/2-over-TCP (H2), no such analysis to our knowledge has taken place between H2 and H3. In addition, past studies rely on Page Load Time as their main, if not only, metric. The purpose of this work is to benchmark the latest draft specification of H3 and dig further into a user's Quality of Experience (QoE) using Lighthouse: an open source (and metric diverse) auditing tool. Our findings show that, for one of H3's early implementations, H3 is mostly worse but achieves a higher average throughput

[ { "created": "Sat, 4 Apr 2020 17:18:53 GMT", "version": "v1" }, { "created": "Sat, 13 Jun 2020 01:55:56 GMT", "version": "v2" }, { "created": "Sun, 11 Oct 2020 17:28:44 GMT", "version": "v3" } ]

2023-05-23

[ [ "Saif", "Darius", "" ], [ "Lung", "Chung-Horng", "" ], [ "Matrawy", "Ashraf", "" ] ]

Sat, 4 Apr 2020 17:18:53 GMT

2020-04-04T17:18:53Z

[ "cs" ]

arxiv

null

[Off-label drugs use in children--clinical perspective].

null

Administration of drugs off-label occurs commonly in pediatrics. The majority of pediatricians prescribe unregistered drugs deliberately. The most common reasons that prompt them to do so are the necessity to treat their patients while there is no alternative medication, and their conviction that the treatment with an unregistered drug is more efficient and/or safer than the treatment with a registered one. The pediatric practice departs from the registered use of drugs most commonly in the following points: the dose, indications, different age of a patient. It is an alarming fact that a substantial group of medications used to treat children has not been included into clinical trials which might result in their registration as drugs suitable for use in pediatrics. Current legal regulations, which have been introduced mainly with the view of protecting patients' safety, considerably inhibit the implementation of clinical trials with children. Moreover, parents, guardians, as well as health-care providers often perceive medical experiments in a very negative way, which causes further difficulties. It should be stressed that the administration of drugs off-label is each pediatrician's autonomous decision for which they take full responsibility.

null

pubmed

null

2,011

null

21,563,434

Przeglad lekarski

1603.02235

Agnes Lagnoux

Thierry Klein (IMT), A Lagnoux (IMT), P Petit (IMT)

Deviation results for sparse tables in hashing with linear probing

null

math.PR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We consider the model of hashing with linear probing and we establish the moderate and large deviations for the total displacement in sparse tables. In this context, Weibull-like-tailed random variables appear. Deviations for sums of such heavy-tailed random variables are studied in \cite{Nagaev69-1,Nagaev69-2}. Here we adapt the proofs therein to deal with conditioned sums of such variables and solve the open question in \cite{TFC12}. By the way, we establish the deviations of the total displacement in full tables, which can be derived from the deviations of empirical processes of i.i.d.\ random variables established in \cite{Wu94}..

[ { "created": "Mon, 7 Mar 2016 20:04:51 GMT", "version": "v1" }, { "created": "Tue, 19 Jan 2021 09:39:26 GMT", "version": "v2" }, { "created": "Wed, 15 Dec 2021 10:17:01 GMT", "version": "v3" } ]

2021-12-16

[ [ "Klein", "Thierry", "", "IMT" ], [ "Lagnoux", "A", "", "IMT" ], [ "Petit", "P", "", "IMT" ] ]

Mon, 7 Mar 2016 20:04:51 GMT

2016-03-07T20:04:51Z

[ "math" ]

arxiv

null

1907.01860

Patricio Cerda

Patricio Cerda (PARIETAL), Ga\"el Varoquaux (NEUROSPIN)

Encoding high-cardinality string categorical variables

null

IEEE Transactions on Knowledge and Data Engineering, Institute of Electrical and Electronics Engineers, pp.1-1

10.1109/TKDE.2020.2992529

null

cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Statistical models usually require vector representations of categorical variables, using for instance one-hot encoding. This strategy breaks down when the number of categories grows, as it creates high-dimensional feature vectors. Additionally, for string entries, one-hot encoding does not capture information in their representation.Here, we seek low-dimensional encoding of high-cardinality string categorical variables. Ideally, these should be: scalable to many categories; interpretable to end users; and facilitate statistical analysis. We introduce two encoding approaches for string categories: a Gamma-Poisson matrix factorization on substring counts, and the min-hash encoder, for fast approximation of string similarities. We show that min-hash turns set inclusions into inequality relations that are easier to learn. Both approaches are scalable and streamable. Experiments on real and simulated data show that these methods improve supervised learning with high-cardinality categorical variables. We recommend the following: if scalability is central, the min-hash encoder is the best option as it does not require any data fit; if interpretability is important, the Gamma-Poisson factorization is the best alternative, as it can be interpreted as one-hot encoding on inferred categories with informative feature names. Both models enable autoML on the original string entries as they remove the need for feature engineering or data cleaning.

[ { "created": "Wed, 3 Jul 2019 11:36:07 GMT", "version": "v1" }, { "created": "Thu, 11 Jul 2019 12:32:56 GMT", "version": "v2" }, { "created": "Wed, 17 Jul 2019 10:17:59 GMT", "version": "v3" }, { "created": "Thu, 5 Sep 2019 08:54:06 GMT", "version": "v4" }, { "created": "Mon, 18 May 2020 15:18:45 GMT", "version": "v5" } ]

2020-07-16

[ [ "Cerda", "Patricio", "", "PARIETAL" ], [ "Varoquaux", "Gaël", "", "NEUROSPIN" ] ]

Wed, 3 Jul 2019 11:36:07 GMT

2019-07-03T11:36:07Z

[ "cs", "stat" ]

arxiv

null

2002.00130

Tao Jiang

Tao Jiang, Stephanie J. London, Mi Kyeong Lee, Josyf C. Mychaleckyj, Alison A. Motsinger-Reif

Higher Criticism Tuned Regression For Weak And Sparse Signals

null

q-bio.QM stat.ME

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Here we propose a novel searching scheme for a tuning parameter in high-dimensional penalized regression methods to address variable selection and modeling when sample sizes are limited compared to the data dimensions. Our method is motivated by high-throughput biological data such as genome-wide association studies (GWAS) and epigenome-wide association studies (EWAS). We propose a new estimate of the regularization parameter $\lambda$ in penalized regression methods based on an estimated lower bound of the proportion of false null hypotheses with confidence $(1 - \alpha)$. The bound is estimated by applying the empirical null distribution of the higher criticism statistic, a second-level significance test constructed by dependent $p$-values using a multi-split regression and aggregation method. A tuning parameter estimate in penalized regression, $\lambda$, corresponds with the lower bound of the proportion of false null hypotheses. Different penalized regression methods with varied signal sparsity and strength are compared in the multi-split method setting. We demonstrate the performance of our method using both simulation experiments and the applications of real data on (1) lipid-trait genetics from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) clinical trial and (2) epigenetic analysis evaluating smoking's influence in differential methylation in the Agricultural Lung Health Study. The proposed algorithm is included in the HCTR package, available at https://cran.r-project.org/web/packages/HCTR/index.html.

[ { "created": "Sat, 1 Feb 2020 02:55:50 GMT", "version": "v1" }, { "created": "Fri, 7 Feb 2020 16:19:59 GMT", "version": "v2" }, { "created": "Mon, 10 Feb 2020 18:55:51 GMT", "version": "v3" } ]

2020-02-11

[ [ "Jiang", "Tao", "" ], [ "London", "Stephanie J.", "" ], [ "Lee", "Mi Kyeong", "" ], [ "Mychaleckyj", "Josyf C.", "" ], [ "Motsinger-Reif", "Alison A.", "" ] ]

Sat, 1 Feb 2020 02:55:50 GMT

2020-02-01T02:55:50Z

[ "q-bio", "stat" ]

arxiv

null

1912.07812

Guangyi Zhang

Guangyi Zhang and Ali Etemad

Capsule Attention for Multimodal EEG-EOG Representation Learning with Application to Driver Vigilance Estimation

Accepted by IEEE Transactions on Neural Systems and Rehabilitation Engineering

null

10.1109/TNSRE.2021.3089594

null

cs.LG cs.CV eess.SP stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Driver vigilance estimation is an important task for transportation safety. Wearable and portable brain-computer interface devices provide a powerful means for real-time monitoring of the vigilance level of drivers to help with avoiding distracted or impaired driving. In this paper, we propose a novel multimodal architecture for in-vehicle vigilance estimation from Electroencephalogram and Electrooculogram. To enable the system to focus on the most salient parts of the learned multimodal representations, we propose an architecture composed of a capsule attention mechanism following a deep Long Short-Term Memory (LSTM) network. Our model learns hierarchical dependencies in the data through the LSTM and capsule feature representation layers. To better explore the discriminative ability of the learned representations, we study the effect of the proposed capsule attention mechanism including the number of dynamic routing iterations as well as other parameters. Experiments show the robustness of our method by outperforming other solutions and baseline techniques, setting a new state-of-the-art. We then provide an analysis on different frequency bands and brain regions to evaluate their suitability for driver vigilance estimation. Lastly, an analysis on the role of capsule attention, multimodality, and robustness to noise is performed, highlighting the advantages of our approach.

[ { "created": "Tue, 17 Dec 2019 04:20:08 GMT", "version": "v1" }, { "created": "Mon, 13 Jan 2020 19:59:07 GMT", "version": "v2" }, { "created": "Tue, 15 Dec 2020 16:31:51 GMT", "version": "v3" }, { "created": "Sun, 13 Jun 2021 16:14:03 GMT", "version": "v4" } ]

2021-06-15

[ [ "Zhang", "Guangyi", "" ], [ "Etemad", "Ali", "" ] ]

Tue, 17 Dec 2019 04:20:08 GMT

2019-12-17T04:20:08Z

[ "cs", "eess", "stat" ]

arxiv

null