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

1109.0425

Chikashi Arita

Chikashi Arita, Andreas Schadschneider

Exact dynamical state of the exclusive queueing process with deterministic hopping

null

Phys. Rev. E 84, 051127 (2011)

10.1103/PhysRevE.84.051127

null

cond-mat.stat-mech nlin.CG

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

The exclusive queueing process (EQP) has recently been introduced as a model for the dynamics of queues which takes into account the spatial structure of the queue. It can be interpreted as a totally asymmetric exclusion process of varying length. Here we investigate the case of deterministic bulk hopping p=1 which turns out to be one of the rare cases where exact nontrivial results for the dynamical properties can be obtained. Using a time-dependent matrix product form we calculate several dynamical properties, e.g. the density profile of the system.

[ { "created": "Fri, 2 Sep 2011 12:27:22 GMT", "version": "v1" }, { "created": "Mon, 2 Jan 2012 21:41:21 GMT", "version": "v2" } ]

2015-05-30

[ [ "Arita", "Chikashi", "" ], [ "Schadschneider", "Andreas", "" ] ]

Fri, 2 Sep 2011 12:27:22 GMT

2011-09-02T12:27:22Z

[ "physics", "nlin" ]

arxiv

null

2109.04480

Harrison LaBollita

Harrison LaBollita and Antia S. Botana

Electronic structure and magnetic properties of higher-order layered nickelates: La$_{n+1}$Ni$_{n}$O$_{2n+2}$ ($n=4-6$)

11 pages, 6 figures

Phys. Rev. B 104, 035148 (2021)

10.1103/PhysRevB.104.035148

null

cond-mat.str-el cond-mat.supr-con

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

The recent discovery of superconductivity in Sr-doped NdNiO$_2$, with a critical temperature of $10-15$ K suggests the possibility of a new family of nickel-based high-temperature superconductors (HTS). NdNiO$_{2}$ is the $n=\infty$ member of a larger series of layered nickelates with chemical formula R$_{n+1}$Ni$_{n}$O$_{2n+2}$ (R $=$ La, Nd, Pr; $n = 2, 3, \dots, \infty$). The $n=3$ member has been experimentally and theoretically shown to be cuprate-like and a promising HTS candidate if electron doping could be achieved. The higher-order $n=4,5,$ and $6$ members of the series fall directly into the cuprate dome area of filling without the need of doping, thus making them promising materials to study, but have not been synthesized yet. Here, we perform first-principles calculations on hypothetical $n=4,5,$ and $6$ structures to study their electronic and magnetic properties and compare them with the known $n=\infty$ and $n=3$ materials. From our calculations, we find that the cuprate-like character of layered nickelates increases from the $n=\infty$ to the $n=3$ members as the charge transfer energy and the self-doping effect due to R-$d$ bands around the Fermi level gradually decrease.

[ { "created": "Thu, 9 Sep 2021 18:00:01 GMT", "version": "v1" }, { "created": "Mon, 13 Sep 2021 21:26:59 GMT", "version": "v2" } ]

2021-09-15

[ [ "LaBollita", "Harrison", "" ], [ "Botana", "Antia S.", "" ] ]

Thu, 9 Sep 2021 18:00:01 GMT

2021-09-09T18:00:01Z

[ "physics" ]

arxiv

null

Exploiting Partially Annotated Data in Temporal Relation Extraction

null

Annotating temporal relations (TempRel) between events described in natural language is known to be labor intensive, partly because the total number of TempRels is quadratic in the number of events. As a result, only a small number of documents are typically annotated, limiting the coverage of various lexical/semantic phenomena. In order to improve existing approaches, one possibility is to make use of the readily available, partially annotated data (P as in partial) that cover more documents. However, missing annotations in P are known to hurt, rather than help, existing systems. This work is a case study in exploring various usages of P for TempRel extraction. Results show that despite missing annotations, P is still a useful supervision signal for this task within a constrained bootstrapping learning framework. The system described in this system is publicly available.

null

acl

inproceedings

ning-etal-2018-exploiting

Nissim, Malvina and Berant, Jonathan and Lenci, Alessandro

jun

2,018

New Orleans, Louisiana

Association for Computational Linguistics

https://aclanthology.org/S18-2018/

Ning, Qiang and Yu, Zhongzhi and Fan, Chuchu and Roth, Dan

Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics

148--153

null

SirT1 regulation of antioxidant genes is dependent on the formation of a FoxO3a/PGC-1α complex.

null

SirT1 is a class III histone deacetylase that has been implicated in metabolic and reactive oxygen species control. In the vasculature it has been shown to decrease endothelial superoxide production, prevent endothelial dysfunction and atherosclerosis. However, the mechanisms that mediate SirT1 antioxidant functions remain to be characterized. The transcription factor FoxO3a and the transcriptional coactivator peroxisome proliferator activated receptor γ-coactivator 1α (PGC-1α) have been shown to induce the expression of antioxidant genes and to be deacetylated by SirT1.

null

pubmed

null

2,013

null

23,461,683

Antioxidants & redox signaling

null

Proteomics under pressure: development of essential sample preparation techniques in proteomics using ultrahigh hydrostatic pressure.

null

The minimization of preanalytical variables in sample preparation is imperative for successful discovery-driven and translational research involving large-scale biomolecular profiling. Here, we demonstrate a novel technique using high hydrostatic pressure in addition to several chaotropes and solvents to maximize efficiency of both cell lysis and enzymatic digestion while minimizing the time, manual involvement in sample processing, and preanalytical variability introduced prior to mass spectrometry-based proteomic analysis. The digestion techniques were evaluated and optimized for in-solution, in-gel, and on-membrane applications using protein standards and cell lysates. The lysis techniques were evaluated using human HepG2 cells. Our results demonstrate that the use of elevated pressure and organic solvents can achieve superior protein recovery of organelle-, complex-, and especially membrane-associated proteins, meanwhile obtaining more than a 20-fold increase in throughput with improved reproducibility. This study introduces the concept of ultrahigh-performance sample preparation platforms for targeted characterization of proteome subsets in biological systems.

null

pubmed

null

2,011

null

22,029,901

Journal of proteome research

1405.3982

Benjamin Fuks

Eric Conte, B\'eranger Dumont, Benjamin Fuks, Chris Wymant

Designing and recasting LHC analyses with MadAnalysis 5

17 pages, 3 figures, 10 tables; version accepted by EPJC

Eur.Phys.J. C74 (2014) 3103

10.1140/epjc/s10052-014-3103-0

CERN-PH-TH/2014-088, LPSC 14-079, LAPTH-033/14, MCNET-14-11

hep-ph hep-ex

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

We present an extension of the expert mode of the MadAnalysis 5 program dedicated to the design or reinterpretation of high-energy physics collider analyses. We detail the predefined classes, functions and methods available to the user and emphasize the most recent developments. The latter include the possible definition of multiple sub-analyses and a novel user-friendly treatment for the selection criteria. We illustrate this approach by two concrete examples: a CMS search for supersymmetric partners of the top quark and a phenomenological analysis targeting hadronically decaying monotop systems.

[ { "created": "Thu, 15 May 2014 20:00:04 GMT", "version": "v1" }, { "created": "Fri, 10 Oct 2014 13:57:33 GMT", "version": "v2" } ]

2014-10-21

[ [ "Conte", "Eric", "" ], [ "Dumont", "Béranger", "" ], [ "Fuks", "Benjamin", "" ], [ "Wymant", "Chris", "" ] ]

Thu, 15 May 2014 20:00:04 GMT

2014-05-15T20:00:04Z

[ "physics" ]

arxiv

null

1105.5332

Andrej Cvetkovski

Andrej Cvetkovski and Mark Crovella

Multidimensional Scaling in the Poincare Disk

null

Applied Mathematics & Information Sciences, 10(1):125, 2016

null

stat.ML cs.SI

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

Multidimensional scaling (MDS) is a class of projective algorithms traditionally used in Euclidean space to produce two- or three-dimensional visualizations of datasets of multidimensional points or point distances. More recently however, several authors have pointed out that for certain datasets, hyperbolic target space may provide a better fit than Euclidean space. In this paper we develop PD-MDS, a metric MDS algorithm designed specifically for the Poincare disk (PD) model of the hyperbolic plane. Emphasizing the importance of proceeding from first principles in spite of the availability of various black box optimizers, our construction is based on an elementary hyperbolic line search and reveals numerous particulars that need to be carefully addressed when implementing this as well as more sophisticated iterative optimization methods in a hyperbolic space model.

[ { "created": "Thu, 26 May 2011 16:05:23 GMT", "version": "v1" }, { "created": "Sun, 29 May 2011 06:06:30 GMT", "version": "v2" }, { "created": "Fri, 12 Feb 2016 09:39:02 GMT", "version": "v3" } ]

2016-02-15

[ [ "Cvetkovski", "Andrej", "" ], [ "Crovella", "Mark", "" ] ]

Thu, 26 May 2011 16:05:23 GMT

2011-05-26T16:05:23Z

[ "cs", "stat" ]

arxiv

null

1804.04579

Hsian-Hua Tseng

David Anderson, Linda Chen, Hsian-Hua Tseng, Hiroshi Iritani

On the finiteness of quantum K-theory of a homogeneous space

Appendix B by Hiroshi Iritani. 32 pages, final version to appear in IMRN

International Mathematics Research Notices, Volume 2022, Issue 2, January 2022, Pages 1313--1349

null

math.AG math.CO

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

We show that the product in the quantum K-ring of a generalized flag manifold $G/P$ involves only finitely many powers of the Novikov variables. In contrast to previous approaches to this finiteness question, we exploit the finite difference module structure of quantum K-theory. At the core of the proof is a bound on the asymptotic growth of the $J$-function, which in turn comes from an analysis of the singularities of the zastava spaces studied in geometric representation theory. An appendix by H. Iritani establishes the equivalence between finiteness and a quadratic growth condition on certain shift operators.

[ { "created": "Thu, 12 Apr 2018 15:49:29 GMT", "version": "v1" }, { "created": "Wed, 23 Jan 2019 20:27:18 GMT", "version": "v2" }, { "created": "Mon, 24 Aug 2020 16:12:53 GMT", "version": "v3" } ]

2022-01-25

[ [ "Anderson", "David", "" ], [ "Chen", "Linda", "" ], [ "Tseng", "Hsian-Hua", "" ], [ "Iritani", "Hiroshi", "" ] ]

Thu, 12 Apr 2018 15:49:29 GMT

2018-04-12T15:49:29Z

[ "math" ]

arxiv

null

1504.04520

Manuel Gonz\'alez Navarrete

Manuel Gonz\'alez-Navarrete

Type-dependent stochastic Ising model describing the dynamics of a non-symmetric feedback module

21 pages, 5 figures. To appears in Mathematical Biosciences and Engineering

null

math.PR math-ph math.MP q-bio.MN

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

We study an alternative approach to model the dynamical behaviors of biological feedback loop, that is, a type-dependent spin system, this class of stochastic models was introduced by Fern\'andez et. al (2009), and are useful since take account to inherent variability of gene expression. We analyze a non-symmetric feedback module being an extension for the repressilator, the first synthetic biological oscillator, invented by Elowitz and Leibler (2000). We consider a mean-field dynamics for a type-dependent Ising model, and then study the empirical-magnetization vector representing concentration of molecules. We apply a convergence result from stochastic jump processes to deterministic trajectories and present a bifurcation analysis for the associated dynamical system. We show that non-symmetric module under study can exhibit very rich behaviours, including the empirical oscillations described by repressilator.

[ { "created": "Fri, 17 Apr 2015 14:54:03 GMT", "version": "v1" }, { "created": "Thu, 22 Oct 2015 15:38:21 GMT", "version": "v2" }, { "created": "Sun, 6 Mar 2016 02:27:18 GMT", "version": "v3" } ]

2016-03-08

[ [ "González-Navarrete", "Manuel", "" ] ]

Fri, 17 Apr 2015 14:54:03 GMT

2015-04-17T14:54:03Z

[ "q-bio", "math" ]

arxiv

null

0902.4177

Krishnan Prasad

K. Prasad, B. Sundar Rajan

Convolutional Codes for Network-Error Correction

10 pages, 5 figures; Revised version; A shortened version of this paper is to appear in the proceedings of GlobeComm 2009

null

cs.IT math.IT

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

In this work, we introduce convolutional codes for network-error correction in the context of coherent network coding. We give a construction of convolutional codes that correct a given set of error patterns, as long as consecutive errors are separated by a certain interval. We also give some bounds on the field size and the number of errors that can get corrected in a certain interval. Compared to previous network error correction schemes, using convolutional codes is seen to have advantages in field size and decoding technique. Some examples are discussed which illustrate the several possible situations that arise in this context.

[ { "created": "Tue, 24 Feb 2009 15:57:38 GMT", "version": "v1" }, { "created": "Thu, 26 Feb 2009 13:31:05 GMT", "version": "v2" }, { "created": "Thu, 6 Aug 2009 13:18:12 GMT", "version": "v3" } ]

2009-08-06

[ [ "Prasad", "K.", "" ], [ "Rajan", "B. Sundar", "" ] ]

Tue, 24 Feb 2009 15:57:38 GMT

2009-02-24T15:57:38Z

[ "cs", "math" ]

arxiv

null

1210.5753

David Damanik

David Damanik (Rice University), Mark Embree (Rice University), Anton Gorodetski (UC Irvine)

Spectral Properties of Schr\"odinger Operators Arising in the Study of Quasicrystals

56 pages

Mathematics of aperiodic order, 307-370, Prog. Math. Phys., 309, 2015

null

math-ph math.DS math.MP math.NA math.SP

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

We survey results that have been obtained for self-adjoint operators, and especially Schr\"odinger operators, associated with mathematical models of quasicrystals. After presenting general results that hold in arbitrary dimensions, we focus our attention on the one-dimensional case, and in particular on several key examples. The most prominent of these is the Fibonacci Hamiltonian, for which much is known by now and to which an entire section is devoted here. Other examples that are discussed in detail are given by the more general class of Schr\"odinger operators with Sturmian potentials. We put some emphasis on the methods that have been introduced quite recently in the study of these operators, many of them coming from hyperbolic dynamics. We conclude with a multitude of numerical calculations that illustrate the validity of the known rigorous results and suggest conjectures for further exploration.

[ { "created": "Sun, 21 Oct 2012 19:20:40 GMT", "version": "v1" }, { "created": "Tue, 6 May 2014 15:56:03 GMT", "version": "v2" } ]

2016-04-22

[ [ "Damanik", "David", "", "Rice University" ], [ "Embree", "Mark", "", "Rice University" ], [ "Gorodetski", "Anton", "", "UC Irvine" ] ]

Sun, 21 Oct 2012 19:20:40 GMT

2012-10-21T19:20:40Z

[ "math" ]

arxiv

null

0705.0862

Quesne Christiane

C. Quesne

Spectrum generating algebras for position-dependent mass oscillator Schrodinger equations

21 pages, no figure, 2 misprints corrected; published version

J. Phys. A: Math. Theor. 40 (2007) 13107-13119

10.1088/1751-8113/40/43/018

ULB/229/CQ/07/3

math-ph math.MP math.QA quant-ph

null

The interest of quadratic algebras for position-dependent mass Schr\"odinger equations is highlighted by constructing spectrum generating algebras for a class of d-dimensional radial harmonic oscillators with $d \ge 2$ and a specific mass choice depending on some positive parameter $\alpha$. Via some minor changes, the one-dimensional oscillator on the line with the same kind of mass is included in this class. The existence of a single unitary irreducible representation belonging to the positive-discrete series type for $d \ge 2$ and of two of them for d=1 is proved. The transition to the constant-mass limit $\alpha \to 0$ is studied and deformed su(1,1) generators are constructed. These operators are finally used to generate all the bound-state wavefunctions by an algebraic procedure.

[ { "created": "Mon, 7 May 2007 07:48:08 GMT", "version": "v1" }, { "created": "Tue, 7 Aug 2007 14:34:29 GMT", "version": "v2" }, { "created": "Tue, 9 Oct 2007 14:45:44 GMT", "version": "v3" } ]

2009-11-13

[ [ "Quesne", "C.", "" ] ]

Mon, 7 May 2007 07:48:08 GMT

2007-05-07T07:48:08Z

[ "physics", "math" ]

arxiv

null

2003.13815

Mohammed Abdelsamea

Asmaa Abbas, Mohammed M. Abdelsamea, Mohamed Medhat Gaber

Classification of COVID-19 in chest X-ray images using DeTraC deep convolutional neural network

null

eess.IV cs.CV cs.LG stat.ML

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

Chest X-ray is the first imaging technique that plays an important role in the diagnosis of COVID-19 disease. Due to the high availability of large-scale annotated image datasets, great success has been achieved using convolutional neural networks (CNNs) for image recognition and classification. However, due to the limited availability of annotated medical images, the classification of medical images remains the biggest challenge in medical diagnosis. Thanks to transfer learning, an effective mechanism that can provide a promising solution by transferring knowledge from generic object recognition tasks to domain-specific tasks. In this paper, we validate and adapt our previously developed CNN, called Decompose, Transfer, and Compose (DeTraC), for the classification of COVID-19 chest X-ray images. DeTraC can deal with any irregularities in the image dataset by investigating its class boundaries using a class decomposition mechanism. The experimental results showed the capability of DeTraC in the detection of COVID-19 cases from a comprehensive image dataset collected from several hospitals around the world. High accuracy of 95.12% (with a sensitivity of 97.91%, a specificity of 91.87%, and a precision of 93.36%) was achieved by DeTraC in the detection of COVID-19 X-ray images from normal, and severe acute respiratory syndrome cases.

[ { "created": "Thu, 26 Mar 2020 15:18:45 GMT", "version": "v1" }, { "created": "Sat, 18 Apr 2020 08:54:31 GMT", "version": "v2" }, { "created": "Sun, 17 May 2020 12:02:33 GMT", "version": "v3" } ]

2020-05-19

[ [ "Abbas", "Asmaa", "" ], [ "Abdelsamea", "Mohammed M.", "" ], [ "Gaber", "Mohamed Medhat", "" ] ]

Thu, 26 Mar 2020 15:18:45 GMT

2020-03-26T15:18:45Z

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

arxiv

null

Back and forth between cancer treatment and cancer control programs: Insights from the Cuban experience.

null

Cancer control is a wider concept than oncology, and includes comprehensive actions for prevention, early diagnosis, treatment, services organization, and education, aiming to modify hard indicators such as incidence, mortality rates, and survival at a population scale. Based on these concepts, organized national cancer programs appeared in several countries in the second half of the 20th century. But at the same time, scientific efforts began to modify the landscape of cancer control. Evidence of mortality reductions began to appear, cancer-driving mutations became measurable, many novel drugs were registered, the methodology of clinical trials spread through health systems, targeted drugs and immunotherapy entered into the mainstream of therapeutics, and treatment goals started to shift from cure to chronic control. The implementation and impact of organized interventions for cancer control show variations according to the context of diverse countries, and scientists and health decision makers can learn from studying these diverse experiences. Among the salient features of cancer control in Cuba are the simultaneous development of a primary care network with abundant human resources and a national biotechnology industry with capacity to provide both generic and innovating drugs and diagnostic systems. The program intentionally assumes the goal of accelerating the transformation of advanced cancer into a chronic disease susceptible of long-term control. The implications of this strategy for population interventions and for scientific research are discussed.

null

pubmed

null

2,018

null

30,318,079

Seminars in oncology

1702.03464

Nicolas Garcia Trillos

Gromov-Hausdorff limit of Wasserstein spaces on point clouds

null

math.MG math.AP math.PR math.ST stat.ML stat.TH

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

We consider a point cloud $X_n := \{ x_1, \dots, x_n \}$ uniformly distributed on the flat torus $\mathbb{T}^d : = \mathbb{R}^d / \mathbb{Z}^d $, and construct a geometric graph on the cloud by connecting points that are within distance $\varepsilon$ of each other. We let $\mathcal{P}(X_n)$ be the space of probability measures on $X_n$ and endow it with a discrete Wasserstein distance $W_n$ as introduced independently by Chow et al, Maas, and Mielke for general finite Markov chains. We show that as long as $\varepsilon= \varepsilon_n$ decays towards zero slower than an explicit rate depending on the level of uniformity of $X_n$, then the space $(\mathcal{P}(X_n), W_n)$ converges in the Gromov-Hausdorff sense towards the space of probability measures on $\mathbb{T}^d$ endowed with the Wasserstein distance. The analysis presented in this paper is a first step in the study of stability of evolution equations defined over random point clouds as the number of points grows to infinity.

[ { "created": "Sat, 11 Feb 2017 22:29:06 GMT", "version": "v1" }, { "created": "Thu, 13 Apr 2017 13:04:51 GMT", "version": "v2" }, { "created": "Fri, 24 May 2019 09:03:07 GMT", "version": "v3" } ]

2019-05-27

[ [ "Trillos", "Nicolas Garcia", "" ] ]

Sat, 11 Feb 2017 22:29:06 GMT

2017-02-11T22:29:06Z

[ "math", "stat" ]

arxiv

null

2111.03811

Haozhe Zhang

Haozhe Zhang, Zexin Cai, Xiaoyi Qin, Ming Li

SIG-VC: A Speaker Information Guided Zero-shot Voice Conversion System for Both Human Beings and Machines

null

cs.SD cs.AI eess.AS

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

Nowadays, as more and more systems achieve good performance in traditional voice conversion (VC) tasks, people's attention gradually turns to VC tasks under extreme conditions. In this paper, we propose a novel method for zero-shot voice conversion. We aim to obtain intermediate representations for speaker-content disentanglement of speech to better remove speaker information and get pure content information. Accordingly, our proposed framework contains a module that removes the speaker information from the acoustic feature of the source speaker. Moreover, speaker information control is added to our system to maintain the voice cloning performance. The proposed system is evaluated by subjective and objective metrics. Results show that our proposed system significantly reduces the trade-off problem in zero-shot voice conversion, while it also manages to have high spoofing power to the speaker verification system.

[ { "created": "Sat, 6 Nov 2021 06:22:45 GMT", "version": "v1" }, { "created": "Fri, 4 Mar 2022 12:30:53 GMT", "version": "v2" }, { "created": "Sun, 2 Apr 2023 04:21:40 GMT", "version": "v3" } ]

2023-04-04

[ [ "Zhang", "Haozhe", "" ], [ "Cai", "Zexin", "" ], [ "Qin", "Xiaoyi", "" ], [ "Li", "Ming", "" ] ]

Sat, 6 Nov 2021 06:22:45 GMT

2021-11-06T06:22:45Z

[ "cs", "eess" ]

arxiv

null

2010.15530

A. Daniel Carnerero

A. Daniel Carnerero, Daniel R. Ramirez and Teodoro Alamo

Probabilistic interval predictor based on dissimilarity functions

10 pages, 3 figures

IEEE Transactions on Automatic Control 2021

10.1109/TAC.2021.3136137

null

eess.SY cs.SY math.ST stat.TH

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

This work presents a new methodology to obtain probabilistic interval predictions of a dynamical system. The proposed strategy uses stored past system measurements to estimate the future evolution of the system. The method relies on the use of dissimilarity functions to estimate the conditional probability density function of the outputs. A family of empirical probability density functions, parameterized by means of two scalars, is introduced. It is shown that the proposed family encompasses the multivariable normal probability density function as a particular case. We show that the presented approach constitutes a generalization of classical estimation methods. A validation scheme is used to tune the two parameters on which the methodology relies. In order to prove the effectiveness of the presented methodology, some numerical examples and comparisons are provided.

[ { "created": "Thu, 29 Oct 2020 12:54:33 GMT", "version": "v1" }, { "created": "Sun, 19 Dec 2021 12:14:39 GMT", "version": "v2" } ]

2021-12-21

[ [ "Carnerero", "A. Daniel", "" ], [ "Ramirez", "Daniel R.", "" ], [ "Alamo", "Teodoro", "" ] ]

Thu, 29 Oct 2020 12:54:33 GMT

2020-10-29T12:54:33Z

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

arxiv

null

1906.02931

Yan Li

Yan Li, Ethan X.Fang, Huan Xu, Tuo Zhao

Inductive Bias of Gradient Descent based Adversarial Training on Separable Data

null

cs.LG stat.ML

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

Adversarial training is a principled approach for training robust neural networks. Despite of tremendous successes in practice, its theoretical properties still remain largely unexplored. In this paper, we provide new theoretical insights of gradient descent based adversarial training by studying its computational properties, specifically on its inductive bias. We take the binary classification task on linearly separable data as an illustrative example, where the loss asymptotically attains its infimum as the parameter diverges to infinity along certain directions. Specifically, we show that when the adversarial perturbation during training has bounded $\ell_2$-norm, the classifier learned by gradient descent based adversarial training converges in direction to the maximum $\ell_2$-norm margin classifier at the rate of $\tilde{\mathcal{O}}(1/\sqrt{T})$, significantly faster than the rate $\mathcal{O}(1/\log T)$ of training with clean data. In addition, when the adversarial perturbation during training has bounded $\ell_q$-norm for some $q\ge 1$, the resulting classifier converges in direction to a maximum mixed-norm margin classifier, which has a natural interpretation of robustness, as being the maximum $\ell_2$-norm margin classifier under worst-case $\ell_q$-norm perturbation to the data. Our findings provide theoretical backups for adversarial training that it indeed promotes robustness against adversarial perturbation.

[ { "created": "Fri, 7 Jun 2019 07:22:56 GMT", "version": "v1" }, { "created": "Mon, 10 Jun 2019 19:24:27 GMT", "version": "v2" }, { "created": "Fri, 26 Jul 2019 14:58:47 GMT", "version": "v3" } ]

2019-07-29

[ [ "Li", "Yan", "" ], [ "Fang", "Ethan X.", "" ], [ "Xu", "Huan", "" ], [ "Zhao", "Tuo", "" ] ]

Fri, 7 Jun 2019 07:22:56 GMT

2019-06-07T07:22:56Z

[ "cs", "stat" ]

arxiv

null

Environmental neuroscience.

null

Environmental neuroscience is an emerging field devoted to the scientific study of brain-mediated, bidirectional relationships between organisms and their social and physical environments. A key feature of environmental neuroscience is the rigorous quantification of environmental features that affect the brain and subsequent behavior. In addition, environmental neuroscience considers factors that vary across multiple temporal and spatial scales that interact to produce behavior (e.g., synapses, neural circuits, cognition, local social interactions, citywide social interactions, citywide physical structures). Environmental neuroscientists then measure the spatial and temporal dynamics of the interactions between different levels of analysis. For example, we demonstrate through hierarchical systems theory and mathematical modeling how interacting with urban greenspace may reduce psychopathology via improvements in neurocognitive functioning, which, in turn, may increase social interactions. This example illustrates how different levels of analysis (e.g., neurocognitive factors, the physical environment, and the social environment) may be combined to understand behavior in novel ways. In addition, we advocate for the collection of data across these scales and measuring their interactions, which will generate rich data sets that will continue to yield insights as new ways to model these complex multilevel systems are developed. We believe that examining all of these levels of analysis at different temporal and spatial scales in addition to modeling their relationships will lead to advances in understanding behavior. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

null

pubmed

null

2,019

null

31,829,683

The American psychologist

cond-mat/0407140

Michael Levin

Michael Levin, Xiao-Gang Wen

Photons and electrons as emergent phenomena

9 pages, 12 figures, RevTex4. Homepage http://dao.mit.edu/~wen/

Rev. Mod. Phys. 77, 871-879 (2005)

10.1103/RevModPhys.77.871

null

cond-mat.str-el hep-th

null

Recent advances in condensed matter theory have revealed that new and exotic phases of matter can exist in spin models (or more precisely, local bosonic models) via a simple physical mechanism, known as "string-net condensation." These new phases of matter have the unusual property that their collective excitations are gauge bosons and fermions. In some cases, the collective excitations can behave just like the photons, electrons, gluons, and quarks in our vacuum. This suggests that photons, electrons, and other elementary particles may have a unified origin -- string-net condensation in our vacuum. In addition, the string-net picture indicates how to make artificial photons, artificial electrons, and artificial quarks and gluons in condensed matter systems.

[ { "created": "Tue, 6 Jul 2004 18:05:33 GMT", "version": "v1" }, { "created": "Fri, 23 Sep 2005 14:20:49 GMT", "version": "v2" } ]

2009-11-10

[ [ "Levin", "Michael", "" ], [ "Wen", "Xiao-Gang", "" ] ]

Tue, 6 Jul 2004 18:05:33 GMT

2004-07-06T18:05:33Z

[ "physics" ]

arxiv

null

{CTAL}: Pre-training Cross-modal Transformer for Audio-and-Language Representations

null

Existing audio-language task-specific predictive approaches focus on building complicated late-fusion mechanisms. However, these models are facing challenges of overfitting with limited labels and low model generalization abilities. In this paper, we present a Cross-modal Transformer for Audio-and-Language, i.e., CTAL, which aims to learn the intra-modality and inter-modality connections between audio and language through two proxy tasks on a large amount of audio-and-language pairs: masked language modeling and masked cross-modal acoustic modeling. After fine-tuning our pre-trained model on multiple downstream audio-and-language tasks, we observe significant improvements across various tasks, such as, emotion classification, sentiment analysis, and speaker verification. On this basis, we further propose a specially-designed fusion mechanism that can be used in fine-tuning phase, which allows our pre-trained model to achieve better performance. Lastly, we demonstrate detailed ablation studies to prove that both our novel cross-modality fusion component and audio-language pre-training methods significantly contribute to the promising results. The code and pre-trained models are available at \url{https://github.com/tal-ai/CTAL_EMNLP2021}.

null

acl

inproceedings

li-etal-2021-ctal

Moens, Marie-Francine and Huang, Xuanjing and Specia, Lucia and Yih, Scott Wen-tau

nov

2,021

Online and Punta Cana, Dominican Republic

Association for Computational Linguistics

https://aclanthology.org/2021.emnlp-main.323/

Li, Hang and Ding, Wenbiao and Kang, Yu and Liu, Tianqiao and Wu, Zhongqin and Liu, Zitao

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

3966--3977

null

Free/Open Source Shallow-Transfer Based Machine Translation for {S}panish and {A}ragonese

null

This article describes the development of a bidirectional shallow-transfer based machine translation system for Spanish and Aragonese, based on the Apertium platform, reusing the resources provided by other translators built for the platform. The system, and the morphological analyser built for it, are both the first resources of their kind for Aragonese. The morphological analyser has coverage of over 80{\textbackslash}{\%}, and is being reused to create a spelling checker for Aragonese. The translator is bidirectional: the Word Error Rate for Spanish to Aragonese is 16.83{\%}, while Aragonese to Spanish is 11.61{\%}.

null

acl

inproceedings

cortes-etal-2012-free

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/L12-1153/

Cort{\'e}s, Juan Pablo Mart{\'i}nez and O{'}Regan, Jim and Tyers, Francis

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

2153--2157

null

Perspectives of future pharmacists on the potential for development and implementation of pharmacist prescribing in Qatar.

null

Background Pharmacists in many developed countries have been granted prescribing authorities under what is known as "non-medical prescribing" or "pharmacist prescribing". However, such prescribing privileges are not available in many developing countries. Objective The objective of this study was to determine the perspectives of future pharmacists (recent pharmacy graduates and pharmacy students) on pharmacist prescribing and its potential implementation in Qatar. Methods A convergent parallel mixed-methods design was used: (1) a cross-sectional survey using a pre-tested questionnaire and; (2) focus group discussions to allow for an in-depth understanding of the issue, with a focus on pharmacists prescribing competencies as well as barriers for its implementation. Main outcome measures Future pharmacists' perspectives and attitudes towards pharmacist prescribing in Qatar. Results The majority of the respondents (94.4%) indicated awareness of the prescribing competency related to selecting treatment options. Furthermore, the majority (92.4%) believed that pharmacists should undergo prescribing training and accreditation before been legally allowed to prescribe, a point that was reiterated in the focus group discussions. Participants generally expressed support for collaborative and supplementary prescribing models when developing prescribing frameworks for Qatar. Four categories emerged under the theme barriers to implementation of pharmacist prescribing: lack of prescribing competency, pharmacist mindset, lack of accessibility to patient records and counseling rooms, and diversity of education and training background. Conclusion The majority of recent pharmacy graduates and students were in favor of pharmacist prescribing been implemented in Qatar. However, a special training program was deemed necessary to qualify pharmacists to prescribe safely and effectively.

null

pubmed

null

2,020

null

31,898,166

International journal of clinical pharmacy

2209.05247

Jannik Z\"urn

Jannik Z\"urn, Sebastian Weber, Wolfram Burgard

TrackletMapper: Ground Surface Segmentation and Mapping from Traffic Participant Trajectories

19 pages, 14 figures, CoRL 2022 v4 (updated acknowledgements)

null

cs.RO cs.CV

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

Robustly classifying ground infrastructure such as roads and street crossings is an essential task for mobile robots operating alongside pedestrians. While many semantic segmentation datasets are available for autonomous vehicles, models trained on such datasets exhibit a large domain gap when deployed on robots operating in pedestrian spaces. Manually annotating images recorded from pedestrian viewpoints is both expensive and time-consuming. To overcome this challenge, we propose TrackletMapper, a framework for annotating ground surface types such as sidewalks, roads, and street crossings from object tracklets without requiring human-annotated data. To this end, we project the robot ego-trajectory and the paths of other traffic participants into the ego-view camera images, creating sparse semantic annotations for multiple types of ground surfaces from which a ground segmentation model can be trained. We further show that the model can be self-distilled for additional performance benefits by aggregating a ground surface map and projecting it into the camera images, creating a denser set of training annotations compared to the sparse tracklet annotations. We qualitatively and quantitatively attest our findings on a novel large-scale dataset for mobile robots operating in pedestrian areas. Code and dataset will be made available at http://trackletmapper.cs.uni-freiburg.de.

[ { "created": "Mon, 12 Sep 2022 13:43:10 GMT", "version": "v1" }, { "created": "Fri, 16 Sep 2022 07:54:09 GMT", "version": "v2" }, { "created": "Mon, 3 Oct 2022 07:37:14 GMT", "version": "v3" }, { "created": "Sun, 8 Jan 2023 16:18:11 GMT", "version": "v4" } ]

2023-01-10

[ [ "Zürn", "Jannik", "" ], [ "Weber", "Sebastian", "" ], [ "Burgard", "Wolfram", "" ] ]

Mon, 12 Sep 2022 13:43:10 GMT

2022-09-12T13:43:10Z

[ "cs" ]

arxiv

null

1701.06299

Vasily E. Tarasov

Valentina V. Tarasova, Vasily E. Tarasov

Economic Growth Model with Constant Pace and Dynamic Memory

7 pages, PDF

Problems of Modern Science and Education. 2017. No.2(84). P.40-45

10.20861/2304-2338-2017-84-001

null

q-fin.EC

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

The article discusses a generalization of model of economic growth with constant pace, which takes into account the effects of dynamic memory. Memory means that endogenous or exogenous variable at a given time depends not only on their value at that time, but also on their values at previous times. To describe the dynamic memory we use derivatives of non-integer orders. We obtain the solutions of fractional differential equations with derivatives of non-integral order, which describe the dynamics of the output caused by the changes of the net investments and effects of power-law fading memory.

[ { "created": "Mon, 23 Jan 2017 08:53:46 GMT", "version": "v1" }, { "created": "Wed, 3 Apr 2019 08:17:14 GMT", "version": "v2" } ]

2019-04-04

[ [ "Tarasova", "Valentina V.", "" ], [ "Tarasov", "Vasily E.", "" ] ]

Mon, 23 Jan 2017 08:53:46 GMT

2017-01-23T08:53:46Z

[ "q-fin" ]

arxiv

null

1902.05108

Eliahu Cohen

Eliahu Cohen, Marina Cort\^es, Avshalom C. Elitzur, Lee Smolin

Realism and causality I: Pilot wave and retrocausal models as possible facilitators

Minorly revised version. Accepted to Phys. Rev. D

Phys. Rev. D 102, 124027 (2020)

10.1103/PhysRevD.102.124027

null

quant-ph gr-qc

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

Of all basic principles of classical physics, realism should arguably be the last to be given up when seeking a better interpretation of quantum mechanics. We examine the de Broglie-Bohm pilot wave theory as a well developed example of a realistic theory. We present three challenges to a naive reading of pilot-wave theory, each based on a system of several entangled particles. With the help of a coarse graining of pilot wave theory into a discrete system, we show how these challenges can be answered. However this comes with a cost. In the description of individual systems, particles appear to scatter off empty branches of the wave function as if they were particles, and conversely travel through particles as if they were waves. More generally, the "particles" of pilot wave theory are led by the guidance equation to move in ways no classical particle would, involving apparent violations of the principles of inertia and momentum conservation.We next argue that the aforementioned cost can be avoided within a retrocausal model. In the proposed version of the pilot wave theory, the particle is guided by a combination of advanced and retarded waves. The resulting account for quantum physics seems to have greater heuristic power, it demands less damage to intuition, and moreover provides some general hints regarding spacetime and causality. This is the first of two papers. In the second [1] we show that, in the context of an explicit model, retrocausality, with respect to an effective, emergent spacetime metric, can coexist with a strict irreversibility of causal processes.

[ { "created": "Wed, 13 Feb 2019 20:10:21 GMT", "version": "v1" }, { "created": "Tue, 19 Mar 2019 19:03:34 GMT", "version": "v2" }, { "created": "Tue, 28 Jan 2020 20:28:05 GMT", "version": "v3" }, { "created": "Thu, 9 Apr 2020 09:09:37 GMT", "version": "v4" } ]

2020-12-22

[ [ "Cohen", "Eliahu", "" ], [ "Cortês", "Marina", "" ], [ "Elitzur", "Avshalom C.", "" ], [ "Smolin", "Lee", "" ] ]

Wed, 13 Feb 2019 20:10:21 GMT

2019-02-13T20:10:21Z

[ "physics" ]

arxiv

null

1910.03332

Xiaowei Wu

Monika Henzinger and Xiaowei Wu

Upper and Lower Bounds for Fully Retroactive Graph Problems

null

cs.DS

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

Classic dynamic data structure problems maintain a data structure subject to a sequence S of updates and they answer queries using the latest version of the data structure, i.e., the data structure after processing the whole sequence. To handle operations that change the sequence S of updates, Demaine et al. (TALG 2007) introduced retroactive data structures. A retroactive operation modifies the update sequence S in a given position t, called time, and either creates or cancels an update in S at time t. A partially retroactive data structure restricts queries to be executed exclusively in the latest version of the data structure. A fully retroactive data structure supports queries at any time t: a query at time t is answered using only the updates of S up to time t. If the sequence S only consists of insertions, the resulting data structure is an incremental retroactive data structure. While efficient retroactive data structures have been proposed for classic data structures, e.g., stack, priority queue and binary search tree, the retroactive version of graph problems are rarely studied. In this paper we study retroactive graph problems including connectivity, minimum spanning forest (MSF), maximum degree, etc. We provide fully retroactive data structures for maintaining the maximum degree, connectivity and MSF in $\tilde{O}(n)$ time per operation. We also give an algorithm for the incremental fully retroactive connectivity with $\tilde{O}(1)$ time per operation. We compliment our algorithms with almost tight hardness results. We show that under the OMv conjecture (proposed by Henzinger et al. (STOC 2015)), there does not exist fully retroactive data structures maintaining connectivity or MSF, or incremental fully retroactive data structure maintaining the maximum degree with $O(n^{1-\epsilon})$ time per operation, for any constant $\epsilon > 0$.

[ { "created": "Tue, 8 Oct 2019 10:48:29 GMT", "version": "v1" }, { "created": "Thu, 10 Oct 2019 07:47:44 GMT", "version": "v2" }, { "created": "Wed, 1 Dec 2021 02:33:01 GMT", "version": "v3" } ]

2021-12-02

[ [ "Henzinger", "Monika", "" ], [ "Wu", "Xiaowei", "" ] ]

Tue, 8 Oct 2019 10:48:29 GMT

2019-10-08T10:48:29Z

[ "cs" ]

arxiv

null

Weighting Model Based on Group Dynamics to Measure Convergence in Multi-party Dialogue

null

This paper proposes a new weighting method for extending a dyad-level measure of convergence to multi-party dialogues by considering group dynamics instead of simply averaging. Experiments indicate the usefulness of the proposed weighted measure and also show that in general a proper weighting of the dyad-level measures performs better than non-weighted averaging in multiple tasks.

null

acl

inproceedings

rahimi-litman-2018-weighting

Komatani, Kazunori and Litman, Diane and Yu, Kai and Papangelis, Alex and Cavedon, Lawrence and Nakano, Mikio

jul

2,018

Melbourne, Australia

Association for Computational Linguistics

https://aclanthology.org/W18-5046/

Rahimi, Zahra and Litman, Diane

Proceedings of the 19th Annual {SIG}dial Meeting on Discourse and Dialogue

385--390

null

1606.05810

Pablo Pelaez

Amalendu Krishna, Pablo Pelaez

The slice spectral sequence for singular schemes and applications

37 pages. Final version. To appear in Annals of K-theory

Ann. K-Th. 3 (2018) 657-708

10.2140/akt.2018.3.657

null

math.AG math.KT

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

We examine the slice spectral sequence for the cohomology of singular schemes with respect to various motivic T-spectra, especially the motivic cobordism spectrum. When the base field k admits resolution of singularities and X is a scheme of finite type over k, we show that Voevodsky's slice filtration leads to a spectral sequence for MGL(X) whose terms are the motivic cohomology groups of X defined using the cdh-hypercohomology. As a consequence, we establish an isomorphism between certain geometric parts of the motivic cobordism and motivic cohomology of X. A similar spectral sequence for the connective K-theory leads to a cycle class map from the motivic cohomology to the homotopy invariant K-theory of X. We show that this cycle class map is injective for projective schemes. We also deduce applications to the torsion in the motivic cohomology of singular schemes.

[ { "created": "Sat, 18 Jun 2016 22:03:06 GMT", "version": "v1" }, { "created": "Wed, 17 Aug 2016 04:18:17 GMT", "version": "v2" }, { "created": "Fri, 22 Dec 2017 19:43:55 GMT", "version": "v3" }, { "created": "Fri, 1 Jun 2018 08:55:06 GMT", "version": "v4" } ]

2018-12-26

[ [ "Krishna", "Amalendu", "" ], [ "Pelaez", "Pablo", "" ] ]

Sat, 18 Jun 2016 22:03:06 GMT

2016-06-18T22:03:06Z

[ "math" ]

arxiv

null

2101.12493

Matthias Pezzutto

Matthias Pezzutto, Luca Schenato, Subhrakanti Dey

Transmission Power Allocation for Remote Estimation with Multi-packet Reception Capabilities

Preprint submitted to Automatica

null

eess.SY cs.SY

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

In this paper we consider the problem of transmission power allocation for remote estimation of a dynamical system in the case where the estimator is able to simultaneously receive packets from multiple interfering sensors, as it is possible e.g. with the latest wireless technologies such as 5G and WiFi. To this end we introduce a general model where packet arrival probabilities are determined based on the received Signal-to-Interference-and-Noise Ratio and with two different receivers design schemes, one implementing standard multi-packet reception technique and one implementing Successive Interference Cancellation decoding algorithm in addition. Then we cast the power allocation problem as an optimization task where the mean error covariance at the remote estimator is minimized, while penalizing the mean transmission power consumption. For the infinite-horizon problem we show the existence of a stationary optimal policy, while for the finite-horizon case we derive some structural properties under the special scenario where the overall system to be estimated can be seen as a set of independent subsystems. Numerical simulations illustrate the improvement given by the proposed receivers over orthogonal schemes that schedules only one sensor transmission at a time in order to avoid interference.

[ { "created": "Fri, 29 Jan 2021 10:00:46 GMT", "version": "v1" }, { "created": "Tue, 27 Jul 2021 08:50:10 GMT", "version": "v2" }, { "created": "Wed, 12 Jan 2022 15:56:14 GMT", "version": "v3" } ]

2022-01-13

[ [ "Pezzutto", "Matthias", "" ], [ "Schenato", "Luca", "" ], [ "Dey", "Subhrakanti", "" ] ]

Fri, 29 Jan 2021 10:00:46 GMT

2021-01-29T10:00:46Z

[ "cs", "eess" ]

arxiv

null

1503.08676

Konstantin Blyuss

N. Sherborne, K.B. Blyuss, I.Z. Kiss

Dynamics of multi-stage infections on networks

23 pages, 8 figures

Bull. Math. Biol. 77, 1909-1933 (2015)

10.1007/s11538-015-0109-1

null

physics.soc-ph nlin.AO q-bio.PE

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

This paper investigates the dynamics of infectious diseases with a non-exponentially distributed infectious period. This is achieved by considering a multi-stage infection model on networks. Using pairwise approximation with a standard closure, a number of important characteristics of disease dynamics are derived analytically, including the final size of an epidemic and a threshold for epidemic outbreaks, and it is shown how these quantities depend on disease characteristics, as well as the number of disease stages. Stochastic simulations of dynamics on networks are performed and compared to the results of pairwise models for several realistic examples of infectious diseases to illustrate the role played by the number of stages in the disease dynamics. These results show that a higher number of disease stages results in faster epidemic outbreaks with a higher peak prevalence and a larger final size of the epidemic. The agreement between the pairwise and simulation methods is excellent in the cases we consider.

[ { "created": "Mon, 30 Mar 2015 14:04:21 GMT", "version": "v1" }, { "created": "Thu, 19 Nov 2015 21:36:54 GMT", "version": "v2" } ]

2015-11-23

[ [ "Sherborne", "N.", "" ], [ "Blyuss", "K. B.", "" ], [ "Kiss", "I. Z.", "" ] ]

Mon, 30 Mar 2015 14:04:21 GMT

2015-03-30T14:04:21Z

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

arxiv

null

2104.14019

Ga\"etan Dou\'eneau-Tabot

Pebble transducers with unary output

39 pages

null

cs.FL

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

Boja\'nczyk recently initiated an intensive study of deterministic pebble transducers, which are two-way automata that can drop marks (named "pebbles") on their input word, and produce an output word. They describe functions from words to words. Two natural restrictions of this definition have been investigated: marble transducers by Dou\'eneau-Tabot et al., and comparison-free pebble transducers (that we rename here "blind transducers") by Nguy\^en et al. Here, we study the decidability of membership problems between the classes of functions computed by pebble, marble and blind transducers that produce a unary output. First, we show that pebble and marble transducers have the same expressive power when the outputs are unary (which is false over non-unary outputs). Then, we characterize 1-pebble transducers with unary output that describe a function computable by a blind transducer, and show that the membership problem is decidable. These results can be interpreted in terms of automated simplification of programs.

[ { "created": "Wed, 28 Apr 2021 20:52:04 GMT", "version": "v1" }, { "created": "Fri, 30 Apr 2021 19:27:14 GMT", "version": "v2" }, { "created": "Sun, 11 Jul 2021 10:13:25 GMT", "version": "v3" }, { "created": "Sat, 4 Sep 2021 07:15:53 GMT", "version": "v4" }, { "created": "Tue, 19 Apr 2022 08:40:49 GMT", "version": "v5" }, { "created": "Fri, 30 Sep 2022 15:06:02 GMT", "version": "v6" } ]

2022-10-03

[ [ "Douéneau-Tabot", "Gaëtan", "" ] ]

Wed, 28 Apr 2021 20:52:04 GMT

2021-04-28T20:52:04Z

[ "cs" ]

arxiv

null

Cross-lingual Few-Shot Learning on Unseen Languages

null

Large pre-trained language models (LMs) have demonstrated the ability to obtain good performance on downstream tasks with limited examples in cross-lingual settings. However, this was mostly studied for relatively resource-rich languages, where at least enough unlabeled data is available to be included in pre-training a multilingual language model. In this paper, we explore the problem of cross-lingual transfer in unseen languages, where no unlabeled data is available for pre-training a model. We use a downstream sentiment analysis task across 12 languages, including 8 unseen languages, to analyze the effectiveness of several few-shot learning strategies across the three major types of model architectures and their learning dynamics. We also compare strategies for selecting languages for transfer and contrast findings across languages seen in pre-training compared to those that are not. Our findings contribute to the body of knowledge on cross-lingual models for low-resource settings that is paramount to increasing coverage, diversity, and equity in access to NLP technology. We show that, in few-shot learning, linguistically similar and geographically similar languages are useful for cross-lingual adaptation, but taking the context from a mixture of random source languages is surprisingly more effective. We also compare different model architectures and show that the encoder-only model, XLM-R, gives the best downstream task performance.

null

acl

inproceedings

winata-etal-2022-cross

He, Yulan and Ji, Heng and Li, Sujian and Liu, Yang and Chang, Chua-Hui

nov

2,022

Online only

Association for Computational Linguistics

https://aclanthology.org/2022.aacl-main.59/

Winata, Genta and Wu, Shijie and Kulkarni, Mayank and Solorio, Thamar and Preotiuc-Pietro, Daniel

Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)

777--791

null

1703.10092

Sergey Slavnov

On noncommutative extensions of linear logic

null

Logical Methods in Computer Science, Volume 15, Issue 3 (September 20, 2019) lmcs:3765

10.23638/LMCS-15(3:30)2019

null

cs.LO math.LO

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

Pomset logic introduced by Retor\'e is an extension of linear logic with a self-dual noncommutative connective. The logic is defined by means of proof-nets, rather than a sequent calculus. Later a deep inference system BV was developed with an eye to capturing Pomset logic, but equivalence of system has not been proven up to now. As for a sequent calculus formulation, it has not been known for either of these logics, and there are convincing arguments that such a sequent calculus in the usual sense simply does not exist for them. In an on-going work on semantics we discovered a system similar to Pomset logic, where a noncommutative connective is no longer self-dual. Pomset logic appears as a degeneration, when the class of models is restricted. Motivated by these semantic considerations, we define in the current work a semicommutative multiplicative linear logic}, which is multiplicative linear logic extended with two nonisomorphic noncommutative connectives (not to be confused with very different Abrusci-Ruet noncommutative logic). We develop a syntax of proof-nets and show how this logic degenerates to Pomset logic. However, a more interesting problem than just finding yet another noncommutative logic is to find a sequent calculus for this logic. We introduce decorated sequents, which are sequents equipped with an extra structure of a binary relation of reachability on formulas. We define a decorated sequent calculus for semicommutative logic and prove that it is cut-free, sound and complete. This is adapted to "degenerate" variations, including Pomset logic. Thus, in particular, we give a variant of sequent calculus formulation for Pomset logic, which is one of the key results of the paper.

[ { "created": "Wed, 29 Mar 2017 15:17:26 GMT", "version": "v1" }, { "created": "Thu, 15 Jun 2017 13:40:40 GMT", "version": "v2" }, { "created": "Mon, 3 Jul 2017 14:50:42 GMT", "version": "v3" }, { "created": "Tue, 16 Oct 2018 20:22:11 GMT", "version": "v4" }, { "created": "Mon, 26 Aug 2019 20:18:21 GMT", "version": "v5" }, { "created": "Thu, 19 Sep 2019 09:48:21 GMT", "version": "v6" } ]

2023-06-22

[ [ "Slavnov", "Sergey", "" ] ]

Wed, 29 Mar 2017 15:17:26 GMT

2017-03-29T15:17:26Z

[ "cs", "math" ]

arxiv

null

1712.04594

Michal Koles\'ar

Timothy B. Armstrong and Michal Koles\'ar

Finite-Sample Optimal Estimation and Inference on Average Treatment Effects Under Unconfoundedness

45 pages, plus supplemental materials (11 pages)

Econometrica, Volume 89, Issue 3, May 2021, pages 1141-1177

10.3982/ECTA16907

null

stat.AP econ.EM stat.ME

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

We consider estimation and inference on average treatment effects under unconfoundedness conditional on the realizations of the treatment variable and covariates. Given nonparametric smoothness and/or shape restrictions on the conditional mean of the outcome variable, we derive estimators and confidence intervals (CIs) that are optimal in finite samples when the regression errors are normal with known variance. In contrast to conventional CIs, our CIs use a larger critical value that explicitly takes into account the potential bias of the estimator. When the error distribution is unknown, feasible versions of our CIs are valid asymptotically, even when $\sqrt{n}$-inference is not possible due to lack of overlap, or low smoothness of the conditional mean. We also derive the minimum smoothness conditions on the conditional mean that are necessary for $\sqrt{n}$-inference. When the conditional mean is restricted to be Lipschitz with a large enough bound on the Lipschitz constant, the optimal estimator reduces to a matching estimator with the number of matches set to one. We illustrate our methods in an application to the National Supported Work Demonstration.

[ { "created": "Wed, 13 Dec 2017 02:57:02 GMT", "version": "v1" }, { "created": "Mon, 17 Dec 2018 15:47:17 GMT", "version": "v2" }, { "created": "Mon, 20 Jul 2020 16:38:30 GMT", "version": "v3" }, { "created": "Wed, 11 Nov 2020 16:16:48 GMT", "version": "v4" }, { "created": "Mon, 18 Jan 2021 15:59:32 GMT", "version": "v5" } ]

2022-10-04

[ [ "Armstrong", "Timothy B.", "" ], [ "Kolesár", "Michal", "" ] ]

Wed, 13 Dec 2017 02:57:02 GMT

2017-12-13T02:57:02Z

[ "stat", "econ" ]

arxiv

null

2106.10067

Nicolas Gast

Nicolas Gast (POLARIS), Bruno Gaujal (POLARIS), Chen Yan (POLARIS)

LP-based policies for restless bandits: necessary and sufficient conditions for (exponentially fast) asymptotic optimality

Mathematics of Operations Research, 2023

null

10.1287/moor.2022.0101

null

math.OC math.PR

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

We provide a framework to analyse control policies for the restless Markovian bandit model, under both finite and infinite time horizon. We show that when the population of arms goes to infinity, the value of the optimal control policy converges to the solution of a linear program (LP). We provide necessary and sufficient conditions for a generic control policy to be: i) asymptotically optimal; ii) asymptotically optimal with square root convergence rate; iii) asymptotically optimal with exponential rate. We then construct the LP-index policy that is asymptotically optimal with square root convergence rate on all models, and with exponential rate if the model is non-degenerate in finite horizon, and satisfies a uniform global attractor property in infinite horizon. We next define the LP-update policy, which is essentially a repeated LP-index policy that solves a new linear program at each decision epoch. We provide numerical experiments to compare the efficiency of LP-based policies. We compare the performance of the LP-index policy and the LP-update policy with other heuristics. Our result demonstrates that the LP-update policy outperforms the LP-index policy in general, and can have a significant advantage when the transition matrices are wrongly estimated.

[ { "created": "Fri, 18 Jun 2021 11:28:27 GMT", "version": "v1" }, { "created": "Wed, 14 Sep 2022 08:42:02 GMT", "version": "v2" }, { "created": "Fri, 22 Dec 2023 08:42:41 GMT", "version": "v3" } ]

2023-12-25

[ [ "Gast", "Nicolas", "", "POLARIS" ], [ "Gaujal", "Bruno", "", "POLARIS" ], [ "Yan", "Chen", "", "POLARIS" ] ]

Fri, 18 Jun 2021 11:28:27 GMT

2021-06-18T11:28:27Z

[ "math" ]

arxiv

null

2012.06777

Dr. Suryansh Kumar

Berk Kaya, Suryansh Kumar, Carlos Oliveira, Vittorio Ferrari, Luc Van Gool

Uncalibrated Neural Inverse Rendering for Photometric Stereo of General Surfaces

Accepted for publication at CVPR 2021. Document info: 18 pages, 21 Figures, 5 tables. (Minor typo corrected)

null

cs.CV

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

This paper presents an uncalibrated deep neural network framework for the photometric stereo problem. For training models to solve the problem, existing neural network-based methods either require exact light directions or ground-truth surface normals of the object or both. However, in practice, it is challenging to procure both of this information precisely, which restricts the broader adoption of photometric stereo algorithms for vision application. To bypass this difficulty, we propose an uncalibrated neural inverse rendering approach to this problem. Our method first estimates the light directions from the input images and then optimizes an image reconstruction loss to calculate the surface normals, bidirectional reflectance distribution function value, and depth. Additionally, our formulation explicitly models the concave and convex parts of a complex surface to consider the effects of interreflections in the image formation process. Extensive evaluation of the proposed method on the challenging subjects generally shows comparable or better results than the supervised and classical approaches.

[ { "created": "Sat, 12 Dec 2020 10:33:08 GMT", "version": "v1" }, { "created": "Thu, 1 Apr 2021 13:26:46 GMT", "version": "v2" }, { "created": "Sat, 17 Apr 2021 22:10:57 GMT", "version": "v3" } ]

2021-04-20

[ [ "Kaya", "Berk", "" ], [ "Kumar", "Suryansh", "" ], [ "Oliveira", "Carlos", "" ], [ "Ferrari", "Vittorio", "" ], [ "Van Gool", "Luc", "" ] ]

Sat, 12 Dec 2020 10:33:08 GMT

2020-12-12T10:33:08Z

[ "cs" ]

arxiv

null

hep-ph/0509275

Leszek Roszkowski

David G. Cerdeno, Ki-Young Choi, Karsten Jedamzik, Leszek Roszkowski, Roberto Ruiz de Austri

Gravitino Dark Matter in the CMSSM With Improved Constraints from BBN

Version with Erratum. Numerical bug fixed. An upper bound on the reheating temperature strengthened by about an order of magnitude

JCAP 0606:005,2006

10.1088/1475-7516/2006/06/005

null

hep-ph astro-ph

null

In the framework of the Constrained MSSM we re--examine the gravitino as the lightest superpartner and a candidate for cold dark matter in the Universe. Unlike in other recent studies, we include both a thermal contribution to its relic population from scatterings in the plasma and a non--thermal one from neutralino or stau decays after freeze--out. Relative to a previous analysis [1] we update, extend and considerably improve our treatment of constraints from observed light element abundances on additional energy released during BBN in association with late gravitino production. Assuming the gravitino mass in the GeV to TeV range, and for natural ranges of other supersymmetric parameters, the neutralino region is excluded, while for smaller values of the gravitino mass it becomes allowed again. The gravitino relic abundance is consistent with observational constraints on cold dark matter from BBN and CMB in some well defined domains of the stau region but, in most cases, only due to a dominant contribution of the thermal population. This implies, depending on the gravitino mass, a large enough reheating temperature. If $\mgravitino>1$ GeV then $T_R>10^7$ GeV, if allowed by BBN and other constraints but, for light gravitinos, if $\mgravitino>100$ keV then $T_R>3\times 10^3$ GeV. On the other hand, constraints mostly from BBN imply an upper bound $T_R \lsim {a few}x 10^8\times10^9$ GeV which appears inconsistent with thermal leptogenesis. Finally, most of the preferred stau region corresponds to the physical vacuum being a false vacuum. The scenario can be partially probed at the LHC.

[ { "created": "Sun, 25 Sep 2005 22:01:04 GMT", "version": "v1" }, { "created": "Fri, 6 Jul 2007 15:30:30 GMT", "version": "v2" } ]

2011-05-05

[ [ "Cerdeno", "David G.", "" ], [ "Choi", "Ki-Young", "" ], [ "Jedamzik", "Karsten", "" ], [ "Roszkowski", "Leszek", "" ], [ "de Austri", "Roberto Ruiz", "" ] ]

Sun, 25 Sep 2005 22:01:04 GMT

2005-09-25T22:01:04Z

[ "physics" ]

arxiv

null

1911.04628

Alan Yang

Alan Yang and AmirEmad Ghassami and Maxim Raginsky and Negar Kiyavash and Elyse Rosenbaum

Model-Augmented Estimation of Conditional Mutual Information for Feature Selection

Accepted to UAI 2020

null

cs.LG stat.ML

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

Markov blanket feature selection, while theoretically optimal, is generally challenging to implement. This is due to the shortcomings of existing approaches to conditional independence (CI) testing, which tend to struggle either with the curse of dimensionality or computational complexity. We propose a novel two-step approach which facilitates Markov blanket feature selection in high dimensions. First, neural networks are used to map features to low-dimensional representations. In the second step, CI testing is performed by applying the $k$-NN conditional mutual information estimator to the learned feature maps. The mappings are designed to ensure that mapped samples both preserve information and share similar information about the target variable if and only if they are close in Euclidean distance. We show that these properties boost the performance of the $k$-NN estimator in the second step. The performance of the proposed method is evaluated on both synthetic and real data.

[ { "created": "Tue, 12 Nov 2019 01:20:54 GMT", "version": "v1" }, { "created": "Mon, 24 Feb 2020 20:45:53 GMT", "version": "v2" }, { "created": "Fri, 19 Jun 2020 20:38:29 GMT", "version": "v3" } ]

2020-06-23

[ [ "Yang", "Alan", "" ], [ "Ghassami", "AmirEmad", "" ], [ "Raginsky", "Maxim", "" ], [ "Kiyavash", "Negar", "" ], [ "Rosenbaum", "Elyse", "" ] ]

Tue, 12 Nov 2019 01:20:54 GMT

2019-11-12T01:20:54Z

[ "cs", "stat" ]

arxiv

null

Geolocation of Tweets with a {B}i{LSTM} Regression Model

null

Identifying a user`s location can be useful for recommendation systems, demographic analyses, and disaster outbreak monitoring. Although Twitter allows users to voluntarily reveal their location, such information isn`t universally available. Analyzing a tweet can provide a general estimation of a tweet location while giving insight into the dialect of the user and other linguistic markers. Such linguistic attributes can be used to provide a regional approximation of tweet origins. In this paper, we present a neural regression model that can identify the linguistic intricacies of a tweet to predict the location of the user. The final model identifies the dialect embedded in the tweet and predicts the location of the tweet.

null

acl

inproceedings

mishra-2020-geolocation

Zampieri, Marcos and Nakov, Preslav and Ljube{\v{si{\'c, Nikola and Tiedemann, J{\"org and Scherrer, Yves

dec

2,020

Barcelona, Spain (Online)

International Committee on Computational Linguistics (ICCL)

https://aclanthology.org/2020.vardial-1.27/

Mishra, Piyush

Proceedings of the 7th Workshop on NLP for Similar Languages, Varieties and Dialects

283--289

null

0809.2484

Bhuwan Joshi

Bhuwan Joshi, Astrid Veronig, K. -S. Cho, S. -C. Bong, Y. -J. Moon, Jeongwoo Lee, B. V. Somov, P. K. Manoharan, and Y. -H. Kim

Magnetic Reconnection During the Two-Phase Evolution of a Solar Eruptive Flare

30 pages, 13 figures

ApJ, 706, 1438-1450, 2009 December 1

10.1088/0004-637X/706/2/1438

null

astro-ph

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

We present a detailed multi-wavelength analysis and interpretation of the evolution of an M7.6 flare on October 24, 2003. The X-ray observations of the flare taken from the RHESSI spacecraft reveal two phases of the flare evolution. The first phase is characterized by the altitude decrease of the X-ray looptop (LT) source for $\sim$11 minutes. Such a long duration of the descending LT source motion is reported for the first time. The EUV loops, located below the X-ray LT source, also undergo contraction with similar speed ($\sim$15 km s$^{-1}$) in this interval. During the second phase the two distinct hard X-ray footpoints (FP) sources are observed which correlate well with UV and H$\alpha$ flare ribbons. The X-ray LT source now exhibits upward motion. The RHESSI spectra during the first phase are soft and indicative of hot thermal emission from flaring loops with temperatures $T>25$ MK at the early stage. On the other hand, the spectra at high energies ($\varepsilon \gtrsim$25 keV) follow hard power laws during the second phase ($\gamma = 2.6-2.8$). We show that the observed motion of the LT and FP sources can be understood as a consequence of three-dimensional magnetic reconnection at a separator in the corona. During the first phase of the flare, the reconnection releases an excess of magnetic energy related to the magnetic tensions generated before a flare by the shear flows in the photosphere. The relaxation of the associated magnetic shear in the corona by the reconnection process explains the descending motion of the LT source. During the second phase, the ordinary reconnection process dominates describing the energy release in terms of the standard model of large eruptive flares.

[ { "created": "Mon, 15 Sep 2008 10:52:53 GMT", "version": "v1" }, { "created": "Fri, 13 Nov 2009 11:01:05 GMT", "version": "v2" } ]

2009-11-13

[ [ "Joshi", "Bhuwan", "" ], [ "Veronig", "Astrid", "" ], [ "Cho", "K. -S.", "" ], [ "Bong", "S. -C.", "" ], [ "Moon", "Y. -J.", "" ], [ "Lee", "Jeongwoo", "" ], [ "Somov", "B. V.", "" ], [ "Manoharan", "P. K.", "" ], [ "Kim", "Y. -H.", "" ] ]

Mon, 15 Sep 2008 10:52:53 GMT

2008-09-15T10:52:53Z

[ "physics" ]

arxiv

null

1304.1742

Jan Stockhofe

V. Achilleos, J. Stockhofe, P. G. Kevrekidis, D. J. Frantzeskakis, P. Schmelcher

Matter-wave dark solitons and their excitation spectra in spin-orbit coupled Bose-Einstein condensates

null

EPL 103, 20002 (2013)

10.1209/0295-5075/103/20002

null

cond-mat.quant-gas nlin.PS

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

We present three types of dark solitons in quasi-one-dimensional spin-orbit coupled repulsive Bose-Einstein condensates. Among these families, two are always stable, while the third one is only stable sufficiently close to the linear regime. The solitons' excitation spectra reveal the potential existence of a second anomalous mode. While the first such mode describes the soliton oscillatory motion in a parabolic trap, the second, when present, reflects the double well structure of the underlying single-particle spectrum. This novel mode results in moving density stripes in the vicinity of the soliton core, or in an out-of-phase oscillation of the constituent components, with little effect on the nearly stationary striped total density of the composite soliton.

[ { "created": "Fri, 5 Apr 2013 15:08:30 GMT", "version": "v1" }, { "created": "Mon, 5 Aug 2013 14:25:33 GMT", "version": "v2" } ]

2013-08-06

[ [ "Achilleos", "V.", "" ], [ "Stockhofe", "J.", "" ], [ "Kevrekidis", "P. G.", "" ], [ "Frantzeskakis", "D. J.", "" ], [ "Schmelcher", "P.", "" ] ]

Fri, 5 Apr 2013 15:08:30 GMT

2013-04-05T15:08:30Z

[ "physics", "nlin" ]

arxiv

null

Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity

null

When primed with only a handful of training samples, very large, pretrained language models such as GPT-3 have shown competitive results when compared to fully-supervised, fine-tuned, large, pretrained language models. We demonstrate that the order in which the samples are provided can make the difference between near state-of-the-art and random guess performance: essentially some permutations are {\textquotedblleft}fantastic{\textquotedblright} and some not. We analyse this phenomenon in detail, establishing that: it is present across model sizes (even for the largest current models), it is not related to a specific subset of samples, and that a given good permutation for one model is not transferable to another. While one could use a development set to determine which permutations are performant, this would deviate from the true few-shot setting as it requires additional annotated data. Instead, we use the generative nature of language models to construct an artificial development set and based on entropy statistics of the candidate permutations on this set, we identify performant prompts. Our method yields a 13{\%} relative improvement for GPT-family models across eleven different established text classification tasks.

null

acl

inproceedings

lu-etal-2022-fantastically

Muresan, Smaranda and Nakov, Preslav and Villavicencio, Aline

may

2,022

Dublin, Ireland

Association for Computational Linguistics

https://aclanthology.org/2022.acl-long.556/

Lu, Yao and Bartolo, Max and Moore, Alastair and Riedel, Sebastian and Stenetorp, Pontus

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

8086--8098

null

1804.00484

Branislav Cvetkovi\'c

B. Cvetkovi\'c and D. Simi\'c

Near horizon OTT black hole asymptotic symmetries and soft hair

LaTEx, 15 pages, no figures; v2 minor revisions, one reference added

Chin. Phys. C 43, No. 1 (2019) 013109

10.1088/1674-1137/43/1/013109

null

hep-th gr-qc

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

We study the near horizon geometry of both static and stationary extremal Oliva Tempo Troncoso (OTT) black holes. For each of these cases, a set of consistent asymptotic conditions is introduced. The canonical generator for the static configuration is shown to be regular. For the rotating OTT black hole, the asymptotic symmetry is described by the time reparametrization, the chiral Virasoro and centrally extended $u(1)$ Kac-Moody algebras.

[ { "created": "Mon, 2 Apr 2018 13:33:20 GMT", "version": "v1" }, { "created": "Wed, 26 Dec 2018 11:50:56 GMT", "version": "v2" } ]

2018-12-27

[ [ "Cvetković", "B.", "" ], [ "Simić", "D.", "" ] ]

Mon, 2 Apr 2018 13:33:20 GMT

2018-04-02T13:33:20Z

[ "physics" ]

arxiv

null

Creation of a {J}apanese Adverb Dictionary that Includes Information on the Speaker`s Communicative Intention Using Machine Learning

null

Japanese adverbs are classified as either declarative or normal; the former declare the communicative intention of the speaker, while the latter convey a manner of action, a quantity, or a degree by which the adverb modifies the verb or adjective that it accompanies. We have automatically classified adverbs as either declarative or not declarative using a machine-learning method such as the maximum entropy method. We defined adverbs having positive or negative connotations as the positive data. We classified adverbs in the EDR dictionary and IPADIC used by Chasen using this result and built an adverb dictionary that contains descriptions of the communicative intentions of the speaker.

null

acl

inproceedings

kanamaru-etal-2006-creation

Calzolari, Nicoletta and Choukri, Khalid and Gangemi, Aldo and Maegaard, Bente and Mariani, Joseph and Odijk, Jan and Tapias, Daniel

may

2,006

Genoa, Italy

European Language Resources Association (ELRA)

https://aclanthology.org/L06-1205/

Kanamaru, Toshiyuki and Murata, Masaki and Isahara, Hitoshi

Proceedings of the Fifth International Conference on Language Resources and Evaluation ({LREC}`06)

null

1806.03813

Min Hyung Cho

Spectrally-accurate numerical method for acoustic scattering from doubly-periodic 3D multilayered media

null

10.1016/j.jcp.2019.05.010

null

math.NA physics.comp-ph

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

A periodizing scheme and the method of fundamental solutions are used to solve acoustic wave scattering from doubly-periodic three-dimensional multilayered media. A scattered wave in a unit cell is represented by the sum of the near and distant contribution. The near contribution uses the free-space Green's function and its eight immediate neighbors. The contribution from the distant sources is expressed using proxy source points over a sphere surrounding the unit cell and its neighbors. The Rayleigh-Bloch radiation condition is applied to the top and bottom layers. Extra unknowns produced by the periodizing scheme in the linear system are eliminated using a Schur complement. The proposed numerical method avoids using singular quadratures and the quasi-periodic Green's function or complicated lattice sum techniques. Therefore, the proposed scheme is robust at all scattering parameters including Wood anomalies. The algorithm is also applicable to electromagnetic problems by using the dyadic Green's function. Numerical examples with 10-digit accuracy are provided. Finally, reflection and transmission spectra are computed over a wide range of incident angles for device characterization.

[ { "created": "Mon, 11 Jun 2018 05:21:19 GMT", "version": "v1" }, { "created": "Tue, 31 Jul 2018 17:22:27 GMT", "version": "v2" }, { "created": "Tue, 26 Mar 2019 16:30:33 GMT", "version": "v3" } ]

2019-06-26

[ [ "Cho", "Min Hyung", "" ] ]

Mon, 11 Jun 2018 05:21:19 GMT

2018-06-11T05:21:19Z

[ "physics", "math" ]

arxiv

null

1903.10156

Dibakar Ghosh Dr.

Arnob Ray, Sarbendu Rakshit, Dibakar Ghosh, and Syamal K. Dana

Intermittent large deviation of chaotic trajectory in Ikeda map: Signature of extreme events

11 pages, 9 figures

Chaos 29, 043131 (2019)

10.1063/1.5092741

null

nlin.CD

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

We notice signatures of extreme events-like behavior in a laser based Ikeda map. The trajectory of the system occasionally travels a large distance away from the bounded chaotic region, which appears as intermittent spiking events in the temporal dynamics. The large spiking events satisfy the conditions of extreme events as usually observed in dynamical systems. The probability density function of the large spiking events shows a long-tail distribution consistent with the characteristics of rare events. The inter-event intervals obey a Poisson-like distribution. We locate the parameter regions of extreme events in phase diagrams. Furthermore, we study two Ikeda maps to explore how and when extreme events terminates via mutual interaction. A pure diffusion of information exchange is unable to terminate extreme events where synchronous occurrence of extreme events is only possible even for large interaction. On the other hand, a threshold-activated coupling can terminate extreme events above a critical value of mutual interaction.

[ { "created": "Mon, 25 Mar 2019 07:19:02 GMT", "version": "v1" }, { "created": "Fri, 3 May 2019 10:39:44 GMT", "version": "v2" } ]

2019-05-06

[ [ "Ray", "Arnob", "" ], [ "Rakshit", "Sarbendu", "" ], [ "Ghosh", "Dibakar", "" ], [ "Dana", "Syamal K.", "" ] ]

Mon, 25 Mar 2019 07:19:02 GMT

2019-03-25T07:19:02Z

[ "nlin" ]

arxiv

null

1905.12807

Ari Kobren

Ari Kobren, Pablo Barrio, Oksana Yakhnenko, Johann Hibschman, Ian Langmore

Constructing High Precision Knowledge Bases with Subjective and Factual Attributes

Appears at KDD 2019 Applied Data Science Track, 19 pages

null

cs.AI cs.LG stat.ML

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

Knowledge bases (KBs) are the backbone of many ubiquitous applications and are thus required to exhibit high precision. However, for KBs that store subjective attributes of entities, e.g., whether a movie is "kid friendly", simply estimating precision is complicated by the inherent ambiguity in measuring subjective phenomena. In this work, we develop a method for constructing KBs with tunable precision--i.e., KBs that can be made to operate at a specific false positive rate, despite storing both difficult-to-evaluate subjective attributes and more traditional factual attributes. The key to our approach is probabilistically modeling user consensus with respect to each entity-attribute pair, rather than modeling each pair as either True or False. Uncertainty in the model is explicitly represented and used to control the KB's precision. We propose three neural networks for fitting the consensus model and evaluate each one on data from Google Maps--a large KB of locations and their subjective and factual attributes. The results demonstrate that our learned models are well-calibrated and thus can successfully be used to control the KB's precision. Moreover, when constrained to maintain 95% precision, the best consensus model matches the F-score of a baseline that models each entity-attribute pair as a binary variable and does not support tunable precision. When unconstrained, our model dominates the same baseline by 12% F-score. Finally, we perform an empirical analysis of attribute-attribute correlations and show that leveraging them effectively contributes to reduced uncertainty and better performance in attribute prediction.

[ { "created": "Tue, 28 May 2019 20:43:23 GMT", "version": "v1" }, { "created": "Thu, 13 Jun 2019 23:45:18 GMT", "version": "v2" }, { "created": "Wed, 31 Jul 2019 13:21:06 GMT", "version": "v3" } ]

2019-08-01

[ [ "Kobren", "Ari", "" ], [ "Barrio", "Pablo", "" ], [ "Yakhnenko", "Oksana", "" ], [ "Hibschman", "Johann", "" ], [ "Langmore", "Ian", "" ] ]

Tue, 28 May 2019 20:43:23 GMT

2019-05-28T20:43:23Z

[ "cs", "stat" ]

arxiv

null

chao-dyn/9902007

K. H. Andersen

K. H. Andersen and P. Muratore-Ginanneschi

Shell Model for Time-correlated Random Advection of Passive Scalars

Replaced with final version + updated figures

null

10.1103/PhysRevE.60.6663

null

chao-dyn nlin.CD

null

We study a minimal shell model for the advection of a passive scalar by a Gaussian time correlated velocity field. The anomalous scaling properties of the white noise limit are studied analytically. The effect of the time correlations are investigated using perturbation theory around the white noise limit and non-perturbatively by numerical integration. The time correlation of the velocity field is seen to enhance the intermittency of the passive scalar.

[ { "created": "Tue, 9 Feb 1999 00:54:59 GMT", "version": "v1" }, { "created": "Tue, 6 Jul 1999 08:07:05 GMT", "version": "v2" }, { "created": "Wed, 20 Oct 1999 14:01:35 GMT", "version": "v3" } ]

2009-10-31

[ [ "Andersen", "K. H.", "" ], [ "Muratore-Ginanneschi", "P.", "" ] ]

Tue, 9 Feb 1999 00:54:59 GMT

1999-02-09T00:54:59Z

[ "nlin" ]

arxiv

null

No Association Linking Short-Term Proton Pump Inhibitor Use to Dementia: Systematic Review and Meta-analysis of Observational Studies.

null

Long-term use of proton pump inhibitors (PPIs) has been associated with a wide variety of potentially serious adverse effects including a possible increased risk of dementia. Studies evaluating this association have reached divergent conclusions. We aimed to evaluate this proposed association further and to assess the quality of the evidence in its support.

null

pubmed

null

2,020

null

31,895,707

The American journal of gastroenterology

0709.0217

Tobias Reichenbach

Tobias Reichenbach, Mauro Mobilia and Erwin Frey

Mobility promotes and jeopardizes biodiversity in rock-paper-scissors games

Final submitted version; the printed version can be found at http://dx.doi.org/10.1038/nature06095 Supplementary movies are available at http://www.theorie.physik.uni-muenchen.de/lsfrey/images_content/movie1.AVI and http://www.theorie.physik.uni-muenchen.de/lsfrey/images_content/movie2.AVI

Nature 448, 1046-1049 (2007)

10.1038/nature06095

LMU-ASC 62/07

q-bio.PE cond-mat.stat-mech physics.bio-ph

null

Biodiversity is essential to the viability of ecological systems. Species diversity in ecosystems is promoted by cyclic, non-hierarchical interactions among competing populations. Such non-transitive relations lead to an evolution with central features represented by the `rock-paper-scissors' game, where rock crushes scissors, scissors cut paper, and paper wraps rock. In combination with spatial dispersal of static populations, this type of competition results in the stable coexistence of all species and the long-term maintenance of biodiversity. However, population mobility is a central feature of real ecosystems: animals migrate, bacteria run and tumble. Here, we observe a critical influence of mobility on species diversity. When mobility exceeds a certain value, biodiversity is jeopardized and lost. In contrast, below this critical threshold all subpopulations coexist and an entanglement of travelling spiral waves forms in the course of temporal evolution. We establish that this phenomenon is robust, it does not depend on the details of cyclic competition or spatial environment. These findings have important implications for maintenance and evolution of ecological systems and are relevant for the formation and propagation of patterns in excitable media, such as chemical kinetics or epidemic outbreaks.

[ { "created": "Mon, 3 Sep 2007 14:56:20 GMT", "version": "v1" }, { "created": "Wed, 9 Apr 2008 19:20:49 GMT", "version": "v2" } ]

2008-04-09

[ [ "Reichenbach", "Tobias", "" ], [ "Mobilia", "Mauro", "" ], [ "Frey", "Erwin", "" ] ]

Mon, 3 Sep 2007 14:56:20 GMT

2007-09-03T14:56:20Z

[ "q-bio", "physics" ]

arxiv

null

Learning from Noisy Labels for Entity-Centric Information Extraction

null

Recent information extraction approaches have relied on training deep neural models. However, such models can easily overfit noisy labels and suffer from performance degradation. While it is very costly to filter noisy labels in large learning resources, recent studies show that such labels take more training steps to be memorized and are more frequently forgotten than clean labels, therefore are identifiable in training. Motivated by such properties, we propose a simple co-regularization framework for entity-centric information extraction, which consists of several neural models with identical structures but different parameter initialization. These models are jointly optimized with the task-specific losses and are regularized to generate similar predictions based on an agreement loss, which prevents overfitting on noisy labels. Extensive experiments on two widely used but noisy benchmarks for information extraction, TACRED and CoNLL03, demonstrate the effectiveness of our framework. We release our code to the community for future research.

null

acl

inproceedings

zhou-chen-2021-learning

Moens, Marie-Francine and Huang, Xuanjing and Specia, Lucia and Yih, Scott Wen-tau

nov

2,021

Online and Punta Cana, Dominican Republic

Association for Computational Linguistics

https://aclanthology.org/2021.emnlp-main.437/

Zhou, Wenxuan and Chen, Muhao

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

5381--5392

null

1102.5482

Jacob Ziv

A Note on the Compaction of long Training Sequences for Universal Classification -a Non-Probabilistic Approach

null

cs.IT math.IT

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

One of the central problems in the classification of individual test sequences (e.g. genetic analysis), is that of checking for the similarity of sample test sequences as compared with a set of much longer training sequences. This is done by a set of classifiers for test sequences of length N, where each of the classifiers is trained by the training sequences so as to minimize the classification error rate when fed with each of the training sequences. It should be noted that the storage of long training sequences is considered to be a serious bottleneck in the next generation sequencing for Genome analysis Some popular classification algorithms adopt a probabilistic approach, by assuming that the sequences are realizations of some variable-length Markov process or a hidden Markov process (HMM), thus enabling the imbeding of the training data onto a variable-length Suffix-tree, the size of which is usually linear in $N$, the length of the test sequence. Despite of the fact that it is not assumed here that the sequences are realizations of probabilistic processes (an assumption that does not seem to be fully justified when dealing with biological data), it is demonstrated that "feature-based" classifiers, where particular substrings (called "features" or markers) are sought in a set of "big data" training sequences may be based on a universal compaction of the training data that is contained in a set of $t$ (long) individual training sequences, onto a suffix-tree with no more than O(N) leaves, regardless of how long the training sequence is, at only a vanishing increase in the classification error rate.

[ { "created": "Sun, 27 Feb 2011 08:33:32 GMT", "version": "v1" }, { "created": "Tue, 24 Jan 2012 15:51:39 GMT", "version": "v10" }, { "created": "Tue, 28 Feb 2012 16:01:49 GMT", "version": "v11" }, { "created": "Thu, 1 Mar 2012 19:15:19 GMT", "version": "v12" }, { "created": "Sat, 31 Mar 2012 07:26:45 GMT", "version": "v13" }, { "created": "Fri, 6 Apr 2012 12:34:59 GMT", "version": "v14" }, { "created": "Sat, 5 May 2012 06:13:27 GMT", "version": "v15" }, { "created": "Wed, 23 Jan 2013 18:02:09 GMT", "version": "v16" }, { "created": "Sun, 27 Jan 2013 04:29:47 GMT", "version": "v17" }, { "created": "Sat, 13 Jul 2013 06:46:55 GMT", "version": "v18" }, { "created": "Sat, 3 Aug 2013 19:24:02 GMT", "version": "v19" }, { "created": "Tue, 1 Mar 2011 09:26:50 GMT", "version": "v2" }, { "created": "Mon, 12 Aug 2013 06:50:45 GMT", "version": "v20" }, { "created": "Mon, 9 Sep 2013 07:12:50 GMT", "version": "v21" }, { "created": "Tue, 10 Sep 2013 14:43:34 GMT", "version": "v22" }, { "created": "Fri, 13 Sep 2013 05:46:27 GMT", "version": "v23" }, { "created": "Mon, 17 Mar 2014 18:13:35 GMT", "version": "v24" }, { "created": "Sun, 30 Mar 2014 11:05:39 GMT", "version": "v25" }, { "created": "Sun, 11 May 2014 16:37:00 GMT", "version": "v26" }, { "created": "Sun, 22 Jun 2014 07:05:36 GMT", "version": "v27" }, { "created": "Fri, 11 Mar 2011 08:31:14 GMT", "version": "v3" }, { "created": "Tue, 12 Apr 2011 15:38:32 GMT", "version": "v4" }, { "created": "Sat, 23 Jul 2011 10:49:03 GMT", "version": "v5" }, { "created": "Sat, 10 Dec 2011 16:24:08 GMT", "version": "v6" }, { "created": "Wed, 21 Dec 2011 09:21:41 GMT", "version": "v7" }, { "created": "Tue, 3 Jan 2012 07:09:58 GMT", "version": "v8" }, { "created": "Mon, 23 Jan 2012 13:40:54 GMT", "version": "v9" } ]

2014-06-24

[ [ "Ziv", "Jacob", "" ] ]

Sun, 27 Feb 2011 08:33:32 GMT

2011-02-27T08:33:32Z

[ "cs", "math" ]

arxiv

null

1104.4869

Nikos Kalogeropoulos

Weak Chaos from Tsallis Entropy

19 pages, Standard LaTeX2e, v2: addition of the last paragraph in Section 4. Three additional refs. To be published in QScience Connect

QScience Connect v2012.12

10.5339/connect.2012.12

null

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

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

We present a geometric, model-independent, argument that aims to explain why the Tsallis entropy describes systems exhibiting "weak chaos", namely systems whose underlying dynamics has vanishing largest Lyapunov exponent. Our argument relies on properties of a deformation map of the reals induced by the Tsallis entropy, and its conclusion agrees with all currently known results.

[ { "created": "Tue, 26 Apr 2011 09:12:54 GMT", "version": "v1" }, { "created": "Thu, 29 Nov 2012 05:45:07 GMT", "version": "v2" } ]

2012-12-11

[ [ "Kalogeropoulos", "Nikos", "" ] ]

Tue, 26 Apr 2011 09:12:54 GMT

2011-04-26T09:12:54Z

[ "physics", "math", "nlin" ]

arxiv

null

2211.08911

Antonio M. Sudoso

Marco Locatelli, Veronica Piccialli, Antonio M. Sudoso

Fix and Bound: An efficient approach for solving large-scale quadratic programming problems with box constraints

null

Mathematical Programming Computation, 2024

10.1007/s12532-024-00270-y

null

math.OC

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

In this paper, we propose a branch-and-bound algorithm for solving nonconvex quadratic programming problems with box constraints (BoxQP). Our approach combines existing tools, such as semidefinite programming (SDP) bounds strengthened through valid inequalities, with a new class of optimality-based linear cuts which leads to variable fixing. The most important effect of fixing the value of some variables is the size reduction along the branch-and-bound tree, allowing to compute bounds by solving SDPs of smaller dimension. Extensive computational experiments over large dimensional (up to $n=200$) test instances show that our method is the state-of-the-art solver on large-scale BoxQPs. Furthermore, we test the proposed approach on the class of binary QP problems, where it exhibits competitive performance with state-of-the-art solvers.

[ { "created": "Wed, 16 Nov 2022 13:42:39 GMT", "version": "v1" }, { "created": "Wed, 2 Oct 2024 15:11:35 GMT", "version": "v2" } ]

2024-11-06

[ [ "Locatelli", "Marco", "" ], [ "Piccialli", "Veronica", "" ], [ "Sudoso", "Antonio M.", "" ] ]

Wed, 16 Nov 2022 13:42:39 GMT

2022-11-16T13:42:39Z

[ "math" ]

arxiv

null

1911.03380

Guillermo Angeris

Guillermo Angeris, Hsien-Tang Kao, Rei Chiang, Charlie Noyes, Tarun Chitra

An analysis of Uniswap markets

null

q-fin.TR cs.MA math.OC

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

Uniswap -- and other constant product markets -- appear to work well in practice despite their simplicity. In this paper, we give a simple formal analysis of constant product markets and their generalizations, showing that, under some common conditions, these markets must closely track the reference market price. We also show that Uniswap satisfies many other desirable properties and numerically demonstrate, via a large-scale agent-based simulation, that Uniswap is stable under a wide range of market conditions.

[ { "created": "Fri, 8 Nov 2019 17:00:11 GMT", "version": "v1" }, { "created": "Mon, 11 Nov 2019 19:50:58 GMT", "version": "v2" }, { "created": "Wed, 13 Nov 2019 21:20:52 GMT", "version": "v3" }, { "created": "Thu, 21 Nov 2019 21:34:39 GMT", "version": "v4" }, { "created": "Sat, 1 Feb 2020 19:02:01 GMT", "version": "v5" }, { "created": "Fri, 15 May 2020 21:05:55 GMT", "version": "v6" }, { "created": "Tue, 9 Feb 2021 20:19:12 GMT", "version": "v7" } ]

2021-02-11

[ [ "Angeris", "Guillermo", "" ], [ "Kao", "Hsien-Tang", "" ], [ "Chiang", "Rei", "" ], [ "Noyes", "Charlie", "" ], [ "Chitra", "Tarun", "" ] ]

Fri, 8 Nov 2019 17:00:11 GMT

2019-11-08T17:00:11Z

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

arxiv

null

Staphylococcus aureus screening and decolonization in orthopaedic surgery and reduction of surgical site infections.

null

Staphylococcus aureus is the most common organism responsible for orthopaedic surgical site infections (SSIs). Patients who are carriers for methicillin-sensitive S. aureus or methicillin-resistant S. aureus (MRSA) have a higher likelihood of having invasive S. aureus infections. Although some have advocated screening for S. aureus and decolonizing it is unclear whether these efforts reduce SSIs.

null

pubmed

null

2,013

null

23,463,284

Clinical orthopaedics and related research

1203.6560

Guy Shinar PhD

Guy Shinar and Martin Feinberg

Concordant Chemical Reaction Networks and the Species-Reaction Graph

71 pages, 9 figures

null

q-bio.MN math.DS

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

In a recent paper it was shown that, for chemical reaction networks possessing a subtle structural property called concordance, dynamical behavior of a very circumscribed (and largely stable) kind is enforced, so long as the kinetics lies within the very broad and natural weakly monotonic class. In particular, multiple equilibria are precluded, as are degenerate positive equilibria. Moreover, under certain circumstances, also related to concordance, all real eigenvalues associated with a positive equilibrium are negative. Although concordance of a reaction network can be decided by readily available computational means, we show here that, when a nondegenerate network's Species-Reaction Graph satisfies certain mild conditions, concordance and its dynamical consequences are ensured. These conditions are weaker than earlier ones invoked to establish kinetic system injectivity, which, in turn, is just one ramification of network concordance. Because the Species-Reaction Graph resembles pathway depictions often drawn by biochemists, results here expand the possibility of inferring signicant dynamical information directly from standard biochemical reaction diagrams.

[ { "created": "Thu, 29 Mar 2012 15:40:12 GMT", "version": "v1" }, { "created": "Thu, 5 Apr 2012 12:04:43 GMT", "version": "v2" }, { "created": "Wed, 25 Apr 2012 15:01:04 GMT", "version": "v3" } ]

2012-04-26

[ [ "Shinar", "Guy", "" ], [ "Feinberg", "Martin", "" ] ]

Thu, 29 Mar 2012 15:40:12 GMT

2012-03-29T15:40:12Z

[ "q-bio", "math" ]

arxiv

null

2109.04261

Nicolas Deperrois

Nicolas Deperrois, Mihai A. Petrovici, Walter Senn, and Jakob Jordan

Learning cortical representations through perturbed and adversarial dreaming

35 pages, 15 figures; ; Jakob Jordan and Walter Senn share senior authorship

null

q-bio.NC cs.LG

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

Humans and other animals learn to extract general concepts from sensory experience without extensive teaching. This ability is thought to be facilitated by offline states like sleep where previous experiences are systemically replayed. However, the characteristic creative nature of dreams suggests that learning semantic representations may go beyond merely replaying previous experiences. We support this hypothesis by implementing a cortical architecture inspired by generative adversarial networks (GANs). Learning in our model is organized across three different global brain states mimicking wakefulness, NREM and REM sleep, optimizing different, but complementary objective functions. We train the model on standard datasets of natural images and evaluate the quality of the learned representations. Our results suggest that generating new, virtual sensory inputs via adversarial dreaming during REM sleep is essential for extracting semantic concepts, while replaying episodic memories via perturbed dreaming during NREM sleep improves the robustness of latent representations. The model provides a new computational perspective on sleep states, memory replay and dreams and suggests a cortical implementation of GANs.

[ { "created": "Thu, 9 Sep 2021 13:31:13 GMT", "version": "v1" }, { "created": "Mon, 3 Jan 2022 15:49:18 GMT", "version": "v2" }, { "created": "Fri, 18 Feb 2022 18:01:48 GMT", "version": "v3" } ]

2022-02-21

[ [ "Deperrois", "Nicolas", "" ], [ "Petrovici", "Mihai A.", "" ], [ "Senn", "Walter", "" ], [ "Jordan", "Jakob", "" ] ]

Thu, 9 Sep 2021 13:31:13 GMT

2021-09-09T13:31:13Z

[ "q-bio", "cs" ]

arxiv

null

1711.08947

Elsa Cazelles

J\'er\'emie Bigot, Elsa Cazelles and Nicolas Papadakis

Central limit theorems for entropy-regularized optimal transport on finite spaces and statistical applications

null

math.ST stat.TH

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

The notion of entropy-regularized optimal transport, also known as Sinkhorn divergence, has recently gained popularity in machine learning and statistics, as it makes feasible the use of smoothed optimal transportation distances for data analysis. The Sinkhorn divergence allows the fast computation of an entropically regularized Wasserstein distance between two probability distributions supported on a finite metric space of (possibly) high-dimension. For data sampled from one or two unknown probability distributions, we derive the distributional limits of the empirical Sinkhorn divergence and its centered version (Sinkhorn loss). We also propose a bootstrap procedure which allows to obtain new test statistics for measuring the discrepancies between multivariate probability distributions. Our work is inspired by the results of Sommerfeld and Munk (2016) on the asymptotic distribution of empirical Wasserstein distance on finite space using unregularized transportation costs. Incidentally we also analyze the asymptotic distribution of entropy-regularized Wasserstein distances when the regularization parameter tends to zero. Simulated and real datasets are used to illustrate our approach.

[ { "created": "Fri, 24 Nov 2017 12:46:58 GMT", "version": "v1" }, { "created": "Thu, 7 Feb 2019 18:18:26 GMT", "version": "v2" }, { "created": "Sun, 3 Nov 2019 21:21:58 GMT", "version": "v3" } ]

2019-11-05

[ [ "Bigot", "Jérémie", "" ], [ "Cazelles", "Elsa", "" ], [ "Papadakis", "Nicolas", "" ] ]

Fri, 24 Nov 2017 12:46:58 GMT

2017-11-24T12:46:58Z

[ "math", "stat" ]

arxiv

null

[Assessment of psychological consequences of trauma].

null

After accidents psychological consequences are often asserted. This may be due to the traumatic experience itself, the experience of severe physical injury or to the experience of a traumatic hospital treatment. The pivotal point of the medicolegal assessment is the evidence of primary mental or physical damage that has to be documented beyond a reasonable doubt in all areas of law. A medicolegal assessment in five steps is proposed.

null

pubmed

null

2,011

null

22,038,383

Der Nervenarzt

null

{S}outh {A}frican National Centre for Digital Language Resources

null

This presentation introduces the imminent establishment of a new language resource infrastructure focusing on languages spoken in Southern Africa, with an eventual aim to become a hub for digital language resources within Sub-Saharan Africa. The Constitution of South Africa makes provision for 11 official languages all with equal status. The current language Resource Management Agency will be merged with the new Centre, which will have a wider focus than that of data acquisition, management and distribution. The Centre will entertain two main programs: Digitisation and Digital Humanities. The digitisation program will focus on the systematic digitisation of relevant text, speech and multi-modal data across the official languages. Relevancy will be determined by a Scientific Advisory Board. This will take place on a continuous basis through specified projects allocated to national members of the Centre, as well as through open-calls aimed at the academic as well as local communities. The digital resources will be managed and distributed through a dedicated web-based portal. The development of the Digital Humanities program will entail extensive academic support for projects implementing digital language based data. The Centre will function as an enabling research infrastructure primarily supported by national government and hosted by the North-West University.

null

acl

inproceedings

roux-2016-south

Calzolari, Nicoletta and Choukri, Khalid and Declerck, Thierry and Goggi, Sara and Grobelnik, Marko and Maegaard, Bente and Mariani, Joseph and Mazo, Helene and Moreno, Asuncion and Odijk, Jan and Piperidis, Stelios

may

2,016

Portoro{\v{z}}, Slovenia

European Language Resources Association (ELRA)

https://aclanthology.org/L16-1391/

Roux, Justus

Proceedings of the Tenth International Conference on Language Resources and Evaluation ({LREC}`16)

2467--2470

null

1812.10761

Shen-Huan Lyu

Shen-Huan Lyu, Lu Wang, Zhi-Hua Zhou

Improving Generalization of Deep Neural Networks by Leveraging Margin Distribution

25 pages, 7 figures, 1 table. Accepted by Neural Networks

Neural Networks, 151:48-60, 2022

10.1016/j.neunet.2022.03.019.

null

cs.LG stat.ML

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

Recent research has used margin theory to analyze the generalization performance for deep neural networks (DNNs). The existed results are almost based on the spectrally-normalized minimum margin. However, optimizing the minimum margin ignores a mass of information about the entire margin distribution, which is crucial to generalization performance. In this paper, we prove a generalization upper bound dominated by the statistics of the entire margin distribution. Compared with the minimum margin bounds, our bound highlights an important measure for controlling the complexity, which is the ratio of the margin standard deviation to the expected margin. We utilize a convex margin distribution loss function on the deep neural networks to validate our theoretical results by optimizing the margin ratio. Experiments and visualizations confirm the effectiveness of our approach and the correlation between generalization gap and margin ratio.

[ { "created": "Thu, 27 Dec 2018 16:34:54 GMT", "version": "v1" }, { "created": "Mon, 7 Oct 2019 15:02:14 GMT", "version": "v2" }, { "created": "Tue, 29 Mar 2022 02:00:48 GMT", "version": "v3" } ]

2024-07-10

[ [ "Lyu", "Shen-Huan", "" ], [ "Wang", "Lu", "" ], [ "Zhou", "Zhi-Hua", "" ] ]

Thu, 27 Dec 2018 16:34:54 GMT

2018-12-27T16:34:54Z

[ "cs", "stat" ]

arxiv

null

Selective synthesis of β-alkylpyrroles.

null

β-Alkylpyrroles are key structural motifs found in many natural products and biologically active compounds as well as functional organic materials. For this reason, synthetic chemists continue to be interested in construction of the framework of β-alkylpyrroles. Due to sufficient aromaticity and π-excessive nature of pyrroles, a straightforward approach to β-alkylpyrroles should be electrophilic aromatic substitution (S(E)Ar) toward the pyrrole ring. However, since a primary nucleophilic site of pyrroles is an α-position, some "trick" is required to direct incoming alkyl electrophiles toward a β-position. This Concept article focuses on presenting previous efforts that have been devoted to the synthesis of β-alkylpyrroles, mainly through the S(E)Ar route.

null

pubmed

null

2,011

null

21,404,342

Chemistry (Weinheim an der Bergstrasse, Germany)

null

Idiosyncratic but not Arbitrary: Learning Idiolects in Online Registers Reveals Distinctive yet Consistent Individual Styles

null

An individual`s variation in writing style is often a function of both social and personal attributes. While structured social variation has been extensively studied, e.g., gender based variation, far less is known about how to characterize individual styles due to their idiosyncratic nature. We introduce a new approach to studying idiolects through a massive cross-author comparison to identify and encode stylistic features. The neural model achieves strong performance at authorship identification on short texts and through an analogy-based probing task, showing that the learned representations exhibit surprising regularities that encode qualitative and quantitative shifts of idiolectal styles. Through text perturbation, we quantify the relative contributions of different linguistic elements to idiolectal variation. Furthermore, we provide a description of idiolects through measuring inter- and intra-author variation, showing that variation in idiolects is often distinctive yet consistent.

null

acl

inproceedings

zhu-jurgens-2021-idiosyncratic

Moens, Marie-Francine and Huang, Xuanjing and Specia, Lucia and Yih, Scott Wen-tau

nov

2,021

Online and Punta Cana, Dominican Republic

Association for Computational Linguistics

https://aclanthology.org/2021.emnlp-main.25/

Zhu, Jian and Jurgens, David

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

279--297

null

2006.13888

\c{C}a\u{g}lar G\"ul\c{c}ehre

Caglar Gulcehre, Ziyu Wang, Alexander Novikov, Tom Le Paine, Sergio Gomez Colmenarejo, Konrad Zolna, Rishabh Agarwal, Josh Merel, Daniel Mankowitz, Cosmin Paduraru, Gabriel Dulac-Arnold, Jerry Li, Mohammad Norouzi, Matt Hoffman, Ofir Nachum, George Tucker, Nicolas Heess and Nando de Freitas

RL Unplugged: A Suite of Benchmarks for Offline Reinforcement Learning

NeurIPS paper. 21 pages including supplementary material, the github link for the datasets: https://github.com/deepmind/deepmind-research/rl_unplugged

null

cs.LG stat.ML

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

Offline methods for reinforcement learning have a potential to help bridge the gap between reinforcement learning research and real-world applications. They make it possible to learn policies from offline datasets, thus overcoming concerns associated with online data collection in the real-world, including cost, safety, or ethical concerns. In this paper, we propose a benchmark called RL Unplugged to evaluate and compare offline RL methods. RL Unplugged includes data from a diverse range of domains including games (e.g., Atari benchmark) and simulated motor control problems (e.g., DM Control Suite). The datasets include domains that are partially or fully observable, use continuous or discrete actions, and have stochastic vs. deterministic dynamics. We propose detailed evaluation protocols for each domain in RL Unplugged and provide an extensive analysis of supervised learning and offline RL methods using these protocols. We will release data for all our tasks and open-source all algorithms presented in this paper. We hope that our suite of benchmarks will increase the reproducibility of experiments and make it possible to study challenging tasks with a limited computational budget, thus making RL research both more systematic and more accessible across the community. Moving forward, we view RL Unplugged as a living benchmark suite that will evolve and grow with datasets contributed by the research community and ourselves. Our project page is available on https://git.io/JJUhd.

[ { "created": "Wed, 24 Jun 2020 17:14:51 GMT", "version": "v1" }, { "created": "Thu, 2 Jul 2020 18:01:52 GMT", "version": "v2" }, { "created": "Tue, 21 Jul 2020 17:31:01 GMT", "version": "v3" }, { "created": "Fri, 12 Feb 2021 18:34:07 GMT", "version": "v4" } ]

2021-02-18

[ [ "Gulcehre", "Caglar", "" ], [ "Wang", "Ziyu", "" ], [ "Novikov", "Alexander", "" ], [ "Paine", "Tom Le", "" ], [ "Colmenarejo", "Sergio Gomez", "" ], [ "Zolna", "Konrad", "" ], [ "Agarwal", "Rishabh", "" ], [ "Merel", "Josh", "" ], [ "Mankowitz", "Daniel", "" ], [ "Paduraru", "Cosmin", "" ], [ "Dulac-Arnold", "Gabriel", "" ], [ "Li", "Jerry", "" ], [ "Norouzi", "Mohammad", "" ], [ "Hoffman", "Matt", "" ], [ "Nachum", "Ofir", "" ], [ "Tucker", "George", "" ], [ "Heess", "Nicolas", "" ], [ "de Freitas", "Nando", "" ] ]

Wed, 24 Jun 2020 17:14:51 GMT

2020-06-24T17:14:51Z

[ "cs", "stat" ]

arxiv

null

1608.01912

Toan T. Nguyen

Nguyen Viet Duc, Toan T. Nguyen, and Paolo Carloni

DNA like$-$charge attraction and overcharging by divalent counterions in the presence of divalent co$-$ions

10 pages, 6 figures

J. Biol. Phys. (2017)

10.1007/s10867-017-9443-x

null

q-bio.BM cond-mat.soft physics.bio-ph

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

Strongly correlated electrostatics of DNA systems has drawn the interest of many groups, especially the condensation and overcharging of DNA by multivalent counterions. By adding counterions of different valencies and shapes, one can enhance or reduce DNA overcharging. In this papers, we focus on the effect of multivalent co-ions, specifically divalent co-ions such as SO$_4^{2-}$. A computational experiment of DNA condensation using Monte$-$Carlo simulation in grand canonical ensemble is carried out where DNA system is in equilibrium with a bulk solution containing a mixture of salt of different valency of co-ions. Compared to system with purely monovalent co-ions, the influence of divalent co-ions shows up in multiple aspects. Divalent co-ions lead to an increase of monovalent salt in the DNA condensate. Because monovalent salts mostly participate in linear screening of electrostatic interactions in the system, more monovalent salt molecules enter the condensate leads to screening out of short-range DNA$-$DNA like charge attraction and weaker DNA condensation free energy. The overcharging of DNA by multivalent counterions is also reduced in the presence of divalent co$-$ions. Strong repulsions between DNA and divalent co-ions and among divalent co-ions themselves leads to a {\em depletion} of negative ions near DNA surface as compared to the case without divalent co-ions. At large distance, the DNA$-$DNA repulsive interaction is stronger in the presence of divalent co$-$ions, suggesting that divalent co$-$ions role is not only that of simple stronger linear screening.

[ { "created": "Fri, 5 Aug 2016 15:17:43 GMT", "version": "v1" }, { "created": "Tue, 16 Aug 2016 04:40:44 GMT", "version": "v2" }, { "created": "Wed, 24 May 2017 16:20:57 GMT", "version": "v3" } ]

2017-05-25

[ [ "Duc", "Nguyen Viet", "" ], [ "Nguyen", "Toan T.", "" ], [ "Carloni", "Paolo", "" ] ]

Fri, 5 Aug 2016 15:17:43 GMT

2016-08-05T15:17:43Z

[ "q-bio", "physics" ]

arxiv

null

1311.2376

Giorgio Ottaviani

Giorgio Ottaviani, Pierre-Jean Spaenlehauer and Bernd Sturmfels

Exact Solutions in Structured Low-Rank Approximation

22 pages; theorem numbering fits with the journal version

SIAM Journal on Matrix Analysis and Applications, 35 (4) (2014), 1521-1542

null

math.OC cs.SC math.AG stat.CO

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

Structured low-rank approximation is the problem of minimizing a weighted Frobenius distance to a given matrix among all matrices of fixed rank in a linear space of matrices. We study exact solutions to this problem by way of computational algebraic geometry. A particular focus lies on Hankel matrices, Sylvester matrices and generic linear spaces.

[ { "created": "Mon, 11 Nov 2013 08:22:26 GMT", "version": "v1" }, { "created": "Thu, 24 Apr 2014 15:37:16 GMT", "version": "v2" }, { "created": "Wed, 22 Feb 2017 17:31:11 GMT", "version": "v3" } ]

2017-02-23

[ [ "Ottaviani", "Giorgio", "" ], [ "Spaenlehauer", "Pierre-Jean", "" ], [ "Sturmfels", "Bernd", "" ] ]

Mon, 11 Nov 2013 08:22:26 GMT

2013-11-11T08:22:26Z

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

arxiv

null

0804.2706

Yaroslav Tserkovnyak

Yaroslav Tserkovnyak and Shimul Akhanjee

Spin-selective localization due to intrinsic spin-orbit coupling

7 pages, 1 figure

Phys. Rev. B 79 (2009) 085114

10.1103/PhysRevB.79.085114

null

cond-mat.mes-hall cond-mat.mtrl-sci

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

We study spin-dependent diffusive transport in the presence of a tunable spin-orbit (SO) interaction in a two-dimensional electron system. The spin precession of an electron in the SO coupling field is expressed in terms of a covariant curvature, affecting the quantum interference between different electronic trajectories. Controlling this curvature field by modulating the SO coupling strength and its gradients by, e.g., electric or elastic means, opens intriguing possibilities for exploring spin-selective localization physics. In particular, applying a weak magnetic field allows the control of the electron localization independently for two spin directions, with the spin-quantization axis that could be "engineered" by appropriate SO interaction gradients.

[ { "created": "Thu, 17 Apr 2008 00:40:55 GMT", "version": "v1" }, { "created": "Thu, 17 Apr 2008 23:42:24 GMT", "version": "v2" }, { "created": "Thu, 19 Feb 2009 22:52:20 GMT", "version": "v3" } ]

2009-11-13

[ [ "Tserkovnyak", "Yaroslav", "" ], [ "Akhanjee", "Shimul", "" ] ]

Thu, 17 Apr 2008 00:40:55 GMT

2008-04-17T00:40:55Z

[ "physics" ]

arxiv

null

Target Really Matters: Target-aware Contrastive Learning and Consistency Regularization for Few-shot Stance Detection

null

Stance detection aims to identify the attitude from an opinion towards a certain target. Despite the significant progress on this task, it is extremely time-consuming and budget-unfriendly to collect sufficient high-quality labeled data for every new target under fully-supervised learning, whereas unlabeled data can be collected easier. Therefore, this paper is devoted to few-shot stance detection and investigating how to achieve satisfactory results in semi-supervised settings. As a target-oriented task, the core idea of semi-supervised few-shot stance detection is to make better use of target-relevant information from labeled and unlabeled data. Therefore, we develop a novel target-aware semi-supervised framework. Specifically, we propose a target-aware contrastive learning objective to learn more distinguishable representations for different targets. Such an objective can be easily applied with or without unlabeled data. Furthermore, to thoroughly exploit the unlabeled data and facilitate the model to learn target-relevant stance features in the opinion content, we explore a simple but effective target-aware consistency regularization combined with a self-training strategy. The experimental results demonstrate that our approach can achieve state-of-the-art performance on multiple benchmark datasets in the few-shot setting.

null

acl

inproceedings

liu-etal-2022-target

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-1.605/

Liu, Rui and Lin, Zheng and Ji, Huishan and Li, Jiangnan and Fu, Peng and Wang, Weiping

Proceedings of the 29th International Conference on Computational Linguistics

6944--6954

null

2106.12515

Yian Chen

Yian Chen, Jeremy Hoskins, Yuehaw Khoo, Michael Lindsey

Committor functions via tensor networks

null

math.NA cs.NA

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

We propose a novel approach for computing committor functions, which describe transitions of a stochastic process between metastable states. The committor function satisfies a backward Kolmogorov equation, and in typical high-dimensional settings of interest, it is intractable to compute and store the solution with traditional numerical methods. By parametrizing the committor function in a matrix product state/tensor train format and using a similar representation for the equilibrium probability density, we solve the variational formulation of the backward Kolmogorov equation with linear time and memory complexity in the number of dimensions. This approach bypasses the need for sampling the equilibrium distribution, which can be difficult when the distribution has multiple modes. Numerical results demonstrate the effectiveness of the proposed method for high-dimensional problems.

[ { "created": "Wed, 23 Jun 2021 16:32:54 GMT", "version": "v1" }, { "created": "Thu, 24 Jun 2021 04:57:29 GMT", "version": "v2" }, { "created": "Tue, 3 Aug 2021 02:07:19 GMT", "version": "v3" } ]

2021-08-04

[ [ "Chen", "Yian", "" ], [ "Hoskins", "Jeremy", "" ], [ "Khoo", "Yuehaw", "" ], [ "Lindsey", "Michael", "" ] ]

Wed, 23 Jun 2021 16:32:54 GMT

2021-06-23T16:32:54Z

[ "cs", "math" ]

arxiv

null

2106.07600

Davide Fioravanti

Davide Fioravanti and Marco Rossi

On the origin of the correspondence between classical and quantum integrable theories

More comprehensive and explicit version (v2 is not to be considered as appeared wrongly preceded by v1)

null

10.1016/j.physletb.2023.137706

null

hep-th math-ph math.MP nlin.SI

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

If we start from certain functional relations as definition of a quantum integrable theory, then we can derive from them a linear integral equation. It can be extended, by introducing dynamical variables, to become an equation with the form of Marchenko's. Then, we derive from the latter a classical (differential) Lax pair. We exemplify our method by focusing on the massive version of the ODE/IM (Ordinary Differential Equations/Integrable Models) correspondence from Quantum sine-Gordon (sG) with many moduli/masses to the classical sinh-Gordon (shG) equation, so describing, in a particular case, some super-symmetric gauge theories and the $AdS_3$ strong coupling scattering amplitudes/Wilson loops. Yet, we present it in a way which reveals its generality of application. In fact, we give some hints on how it works for spin chains.

[ { "created": "Mon, 14 Jun 2021 17:07:13 GMT", "version": "v1" }, { "created": "Wed, 17 Aug 2022 13:52:52 GMT", "version": "v2" }, { "created": "Thu, 18 Aug 2022 11:45:03 GMT", "version": "v3" } ]

2023-01-25

[ [ "Fioravanti", "Davide", "" ], [ "Rossi", "Marco", "" ] ]

Mon, 14 Jun 2021 17:07:13 GMT

2021-06-14T17:07:13Z

[ "physics", "math", "nlin" ]

arxiv

null

{S}amsung {R}{\&}{D} Institute {P}oland Participation in {WMT} 2022

null

This paper presents the system description of Samsung R{\&}D Institute Poland participation in WMT 2022 for General MT solution for medium and low resource languages: Russian and Croatian. Our approach combines iterative noised/tagged back-translation and iterative distillation. We investigated different monolingual resources and compared their influence on final translations. We used available BERT-likemodels for text classification and for extracting domains of texts. Then we prepared an ensemble of NMT models adapted to multiple domains. Finally we attempted to predict ensemble weight vectors from the BERT-based domain classifications for individual sentences. Our final trained models reached quality comparable to best online translators using only limited constrained resources during training.

null

acl

inproceedings

dobrowolski-etal-2022-samsung

Koehn, Philipp and Barrault, Lo{\"ic and Bojar, Ond{\v{rej and Bougares, Fethi and Chatterjee, Rajen and Costa-juss{\`a, Marta R. and Federmann, Christian and Fishel, Mark and Fraser, Alexander and Freitag, Markus and Graham, Yvette and Grundkiewicz, Roman and Guzman, Paco and Haddow, Barry and Huck, Matthias and Jimeno Yepes, Antonio and Kocmi, Tom and Martins, Andr{\'e and Morishita, Makoto and Monz, Christof and Nagata, Masaaki and Nakazawa, Toshiaki and Negri, Matteo and N{\'ev{\'eol, Aur{\'elie and Neves, Mariana and Popel, Martin and Turchi, Marco and Zampieri, Marcos

dec

2,022

Abu Dhabi, United Arab Emirates (Hybrid)

Association for Computational Linguistics

https://aclanthology.org/2022.wmt-1.17/

Dobrowolski, Adam and Klimaszewski, Mateusz and My{\'s}liwy, Adam and Szyma{\'n}ski, Marcin and Kowalski, Jakub and Szypu{\l}a, Kornelia and Przew{\l}ocki, Pawe{\l} and Przybysz, Pawe{\l}

Proceedings of the Seventh Conference on Machine Translation (WMT)

251--259

null

cond-mat/0403640

Rosa DiFelice

Danny Porath, Gianaurelio Cuniberti, Rosa Di Felice

Charge Transport in DNA-Based Devices

A single pdf file of 52 pages, containing the text and 23 figures. Review about direct measurements of DNA conductivity and related theoretical studies. For higher-resolution figures contact the authors or retrieve the original publications cited in the captions

Topics in Current Chemistry 237, p. 183-227 (2004), ed. Gary Schuster, Springer-Verlag, Heidelberg

10.1007/3-540-31514-4_16

null

cond-mat.mtrl-sci cond-mat.soft q-bio.BM

null

Charge migration along DNA molecules has attracted scientific interest for over half a century. Reports on possible high rates of charge transfer between donor and acceptor through the DNA, obtained in the last decade from solution chemistry experiments on large numbers of molecules, triggered a series of direct electrical transport measurements through DNA single molecules, bundles and networks. These measurements are reviewed and presented here. From these experiments we conclude that electrical transport is feasible in short DNA molecules, in bundles and networks, but blocked in long single molecules that are attached to surfaces. The experimental background is complemented by an account of the theoretical/computational schemes that are applied to study the electronic and transport properties of DNA-based nanowires. Examples of selected applications are given, to show the capabilities and limits of current theoretical approaches to accurately describe the wires, interpret the transport measurements, and predict suitable strategies to enhance the conductivity of DNA nanostructures.

[ { "created": "Thu, 25 Mar 2004 17:47:52 GMT", "version": "v1" }, { "created": "Fri, 16 Apr 2004 07:54:45 GMT", "version": "v2" } ]

2009-11-10

[ [ "Porath", "Danny", "" ], [ "Cuniberti", "Gianaurelio", "" ], [ "Di Felice", "Rosa", "" ] ]

Thu, 25 Mar 2004 17:47:52 GMT

2004-03-25T17:47:52Z

[ "q-bio", "physics" ]

arxiv

null

{RTM} results for Predicting Translation Performance

null

With improved prediction combination using weights based on their training performance and stacking and multilayer perceptrons to build deeper prediction models, RTMs become the 3rd system in general at the sentence-level prediction of translation scores and achieve the lowest RMSE in English to German NMT QET results. For the document-level task, we compare document-level RTM models with sentence-level RTM models obtained with the concatenation of document sentences and obtain similar results.

null

acl

inproceedings

bicici-2018-rtm

Bojar, Ond{\v{r}}ej and Chatterjee, Rajen and Federmann, Christian and Fishel, Mark and Graham, Yvette and Haddow, Barry and Huck, Matthias and Yepes, Antonio Jimeno and Koehn, Philipp and Monz, Christof and Negri, Matteo and N{\'e}v{\'e}ol, Aur{\'e}lie and Neves, Mariana and Post, Matt and Specia, Lucia and Turchi, Marco and Verspoor, Karin

oct

2,018

Belgium, Brussels

Association for Computational Linguistics

https://aclanthology.org/W18-6458/

Bi{\c{c}}ici, Ergun

Proceedings of the Third Conference on Machine Translation: Shared Task Papers

765--769

null

1102.5740

Alberto Zaffaroni

Nick Halmagyi, Michela Petrini, Alberto Zaffaroni

Non-Relativistic Solutions of N=2 Gauged Supergravity

27 pages, v2 refs added, comments regarding string embeddings altered, cleaner discussion of the scaling parameter for Schrodinger solutions; v3 published version, refs added

JHEP 1108:041,2011

10.1007/JHEP08(2011)041

null

hep-th

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

We find infinite families of supersymmetric solutions of four dimensional, N=2 gauged supergravity with Lifshitz, Schrodinger and also AdS symmetries. We focus on the canonical example of a single hypermultiplet and a single vector multiplet and find that the spectrum of solutions depends crucially on whether the gaugings are electric or magnetic but to a far milder extent on the strength of the gaugings. For purely electric or purely magnetic gaugings we generically find Lifshitz solutions, while for a mixed gauging we find Schrodinger and AdS solutions. For some of the gaugings the theory has a known lift to string/M-theory thus giving a higher dimensional embedding of our solutions.

[ { "created": "Mon, 28 Feb 2011 19:00:42 GMT", "version": "v1" }, { "created": "Tue, 29 Mar 2011 14:27:55 GMT", "version": "v2" }, { "created": "Fri, 9 Sep 2011 08:01:58 GMT", "version": "v3" } ]

2011-09-12

[ [ "Halmagyi", "Nick", "" ], [ "Petrini", "Michela", "" ], [ "Zaffaroni", "Alberto", "" ] ]

Mon, 28 Feb 2011 19:00:42 GMT

2011-02-28T19:00:42Z

[ "physics" ]

arxiv

null

1702.02267

Quan Li

David Gamarnik, Quan Li and Hongyi Zhang

Matrix Completion from $O(n)$ Samples in Linear Time

45 pages, 1 figure. Short version accepted for presentation at Conference on Learning Theory (COLT) 2017

null

stat.ML cs.DS cs.LG math.OC

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

We consider the problem of reconstructing a rank-$k$ $n \times n$ matrix $M$ from a sampling of its entries. Under a certain incoherence assumption on $M$ and for the case when both the rank and the condition number of $M$ are bounded, it was shown in \cite{CandesRecht2009, CandesTao2010, keshavan2010, Recht2011, Jain2012, Hardt2014} that $M$ can be recovered exactly or approximately (depending on some trade-off between accuracy and computational complexity) using $O(n \, \text{poly}(\log n))$ samples in super-linear time $O(n^{a} \, \text{poly}(\log n))$ for some constant $a \geq 1$. In this paper, we propose a new matrix completion algorithm using a novel sampling scheme based on a union of independent sparse random regular bipartite graphs. We show that under the same conditions w.h.p. our algorithm recovers an $\epsilon$-approximation of $M$ in terms of the Frobenius norm using $O(n \log^2(1/\epsilon))$ samples and in linear time $O(n \log^2(1/\epsilon))$. This provides the best known bounds both on the sample complexity and computational complexity for reconstructing (approximately) an unknown low-rank matrix. The novelty of our algorithm is two new steps of thresholding singular values and rescaling singular vectors in the application of the "vanilla" alternating minimization algorithm. The structure of sparse random regular graphs is used heavily for controlling the impact of these regularization steps.

[ { "created": "Wed, 8 Feb 2017 03:52:40 GMT", "version": "v1" }, { "created": "Sat, 3 Jun 2017 21:59:54 GMT", "version": "v2" }, { "created": "Tue, 6 Jun 2017 04:14:45 GMT", "version": "v3" }, { "created": "Tue, 22 Aug 2017 04:05:36 GMT", "version": "v4" } ]

2017-08-23

[ [ "Gamarnik", "David", "" ], [ "Li", "Quan", "" ], [ "Zhang", "Hongyi", "" ] ]

Wed, 8 Feb 2017 03:52:40 GMT

2017-02-08T03:52:40Z

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

arxiv

null

1005.5288

Decio Levi

Christian Scimiterna and Decio Levi

C-Integrability Test for Discrete Equations via Multiple Scale Expansions

null

SIGMA 6 (2010), 070, 17 pages

10.3842/SIGMA.2010.070

null

nlin.SI

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

In this paper we are extending the well known integrability theorems obtained by multiple scale techniques to the case of linearizable difference equations. As an example we apply the theory to the case of a differential-difference dispersive equation of the Burgers hierarchy which via a discrete Hopf-Cole transformation reduces to a linear differential difference equation. In this case the equation satisfies the $A_1$, $A_2$ and $A_3$ linearizability conditions. We then consider its discretization. To get a dispersive equation we substitute the time derivative by its symmetric discretization. When we apply to this nonlinear partial difference equation the multiple scale expansion we find out that the lowest order non-secularity condition is given by a non-integrable nonlinear Schr\"odinger equation. Thus showing that this discretized Burgers equation is neither linearizable not integrable.

[ { "created": "Fri, 28 May 2010 13:31:24 GMT", "version": "v1" }, { "created": "Tue, 31 Aug 2010 05:02:44 GMT", "version": "v2" } ]

2010-09-01

[ [ "Scimiterna", "Christian", "" ], [ "Levi", "Decio", "" ] ]

Fri, 28 May 2010 13:31:24 GMT

2010-05-28T13:31:24Z

[ "nlin" ]

arxiv

null

2201.10895

Hirokuni Miyamoto

Hirokuni Miyamoto, Wataru Suda, Hiroaki Kodama, Hideyuki Takahashi, Yumiko Nakanishi, Shigeharu Moriya, Kana Adachi, Nao Kiriyama, Masaya Wada, Daisuke Sudo, Shunsuke Ito, Shunsuke Ito, Minami Shibata, Shinji Wada, Takako Murano, Hitoshi Taguchi, Chie Shindo, Arisa Tsuboi, Naoko Tsuji, Makiko Matsuura, Chitose Ishii, Teruno Nakaguma, Toshiyuki Ito, Toru Okada, Teruo Matsushita, Takashi Satoh, Tamotsu Kato, Atsushi Kurotani, Hideaki Shima, Yudai Inabu, Yukihiro Tashiro, Kenji Sakai, Kenichi Mori, Takashi Satoh, Kenta Suzuki, Takeshi Miura, Hidetoshi Morita, Shinji Fukuda, Jun Kikuchi, Hisashi Miyamoto, Masahira Hattori, Naoki Yamamoto, Hiroshi Ohno

A novel sustainable role of compost as a universal protective substitute for fish, chicken, pig, and cattle, and its estimation by structural equation modeling

null

q-bio.QM

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

Natural decomposition of organic matter is essential in food systems, and compost is used worldwide as an organic fermented fertilizer. However, as a feature of the ecosystem, its effects on the animals are poorly understood. Here we show that oral administration of compost and/or its derived thermophilic Bacillaceae, i.e., Caldibacillus hisashii and Weizmannia coagulans, can modulate the prophylactic activities of various industrial animals. The fecal omics analyses in the modulatory process showed an improving trend dependent upon animal species, environmental conditions, and administration. However, structural equation modeling (SEM) estimated the grouping candidates of bacteria and metabolites as standard key components beyond the animal species. In particular, the SEM model implied a strong relationship among partly digesting fecal amino acids, increasing genus Lactobacillus as inhabitant beneficial bacteria and 2-aminoisobutyric acid involved in lantibiotics. These results highlight the potential role of compost for sustainable protective control in agriculture, fishery, and livestock industries.

[ { "created": "Wed, 26 Jan 2022 12:19:50 GMT", "version": "v1" }, { "created": "Fri, 28 Jan 2022 21:29:16 GMT", "version": "v2" }, { "created": "Tue, 1 Mar 2022 22:53:17 GMT", "version": "v3" }, { "created": "Sun, 27 Nov 2022 05:45:18 GMT", "version": "v4" } ]

2022-11-29

[ [ "Miyamoto", "Hirokuni", "" ], [ "Suda", "Wataru", "" ], [ "Kodama", "Hiroaki", "" ], [ "Takahashi", "Hideyuki", "" ], [ "Nakanishi", "Yumiko", "" ], [ "Moriya", "Shigeharu", "" ], [ "Adachi", "Kana", "" ], [ "Kiriyama", "Nao", "" ], [ "Wada", "Masaya", "" ], [ "Sudo", "Daisuke", "" ], [ "Ito", "Shunsuke", "" ], [ "Ito", "Shunsuke", "" ], [ "Shibata", "Minami", "" ], [ "Wada", "Shinji", "" ], [ "Murano", "Takako", "" ], [ "Taguchi", "Hitoshi", "" ], [ "Shindo", "Chie", "" ], [ "Tsuboi", "Arisa", "" ], [ "Tsuji", "Naoko", "" ], [ "Matsuura", "Makiko", "" ], [ "Ishii", "Chitose", "" ], [ "Nakaguma", "Teruno", "" ], [ "Ito", "Toshiyuki", "" ], [ "Okada", "Toru", "" ], [ "Matsushita", "Teruo", "" ], [ "Satoh", "Takashi", "" ], [ "Kato", "Tamotsu", "" ], [ "Kurotani", "Atsushi", "" ], [ "Shima", "Hideaki", "" ], [ "Inabu", "Yudai", "" ], [ "Tashiro", "Yukihiro", "" ], [ "Sakai", "Kenji", "" ], [ "Mori", "Kenichi", "" ], [ "Satoh", "Takashi", "" ], [ "Suzuki", "Kenta", "" ], [ "Miura", "Takeshi", "" ], [ "Morita", "Hidetoshi", "" ], [ "Fukuda", "Shinji", "" ], [ "Kikuchi", "Jun", "" ], [ "Miyamoto", "Hisashi", "" ], [ "Hattori", "Masahira", "" ], [ "Yamamoto", "Naoki", "" ], [ "Ohno", "Hiroshi", "" ] ]

Wed, 26 Jan 2022 12:19:50 GMT

2022-01-26T12:19:50Z

[ "q-bio" ]

arxiv

null

2207.02758

Eli Tziperman

Golan Bel and Eli Tziperman

Is the resonant wave interaction approximation consistent with the dynamics of internal wave fields?

null

nlin.CD physics.ao-ph physics.flu-dyn

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

Nonlinear interaction and breaking of internal ocean waves are responsible for much of the interior ocean mixing, affecting ocean carbon storage and the global overturning circulation. These interactions are also believed to dictate the observed Garrett-Munk wave energy spectrum, which is still unexplained after 50 years of studies. According to the resonant wave interaction approximation, used to derive the kinetic equation for the energy spectrum, the dominant interactions are between wave triads whose wavevectors satisfy $\mathbf{k}=\mathbf{p}+\mathbf{q}$, and whose frequencies satisfy $\omega_{\mathbf{k}}=|\omega_{\mathbf{p}}\pm\omega_{\mathbf{q}}|$. In order to test the validity of the resonant wave interaction approximation, we examine several analytical derivations of the theory. The assumptions underlying each derivation are tested using idealized direct 2d numerical simulations, representing near-observed energy levels of the oceanic internal wave field. We show that the assumptions underlying the derivations are not consistent with the simulated dynamics. In addition, most of the triads satisfying the resonant conditions do not contribute significantly to nonlinear wave energy transfer in our simulations, while some interactions that are dominant in nonlinear energy transfers do not satisfy the resonance conditions. We also point to possible self-consistency issues with some derivations found in the literature.

[ { "created": "Wed, 6 Jul 2022 15:50:14 GMT", "version": "v1" }, { "created": "Tue, 2 Aug 2022 13:13:33 GMT", "version": "v2" }, { "created": "Tue, 25 Oct 2022 15:58:52 GMT", "version": "v3" }, { "created": "Fri, 1 Dec 2023 15:44:26 GMT", "version": "v4" } ]

2023-12-04

[ [ "Bel", "Golan", "" ], [ "Tziperman", "Eli", "" ] ]

Wed, 6 Jul 2022 15:50:14 GMT

2022-07-06T15:50:14Z

[ "physics", "nlin" ]

arxiv

null

0809.1259

Dominic Meiser

D. Meiser and M. J. Holland

Robustness of Heisenberg-limited interferometry with balanced Fock states

13 pages, 6 figures; high-resolution version can be obtained from authors

New J. Phys. 10, 073014 (2008).

10.1088/1367-2630/11/3/033002

null

quant-ph

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

Interferometry with Heisenberg limited phase resolution may play an important role in the next generation of atomic clocks, gravitational wave detectors, and in quantum information science. For experimental implementations the robustness of the phase resolution is crucial since any experimental realization will be subject to imperfections. In this article we study the robustness of phase reconstruction with two number states as input subject to fluctuations in the state preparation. We find that the phase resolution is insensitive to fluctuations in the total number of particles and robust against noise in the number difference at the input. The phase resolution depends on the uncertainty in the number difference in a universal way that has a clear physical interpretation: Fundamental noise due to the Heisenberg limit and noise due to state preparation imperfection contribute essentially independently to the total uncertainty in the phase. For number difference uncertainties less than one the first noise source is dominant and the phase resolution is essentially Heisenberg limited. For number difference uncertainties greater than one the noise due to state preparation imperfection is dominant and the phase resolution deteriorates linearly with the number difference uncertainty.

[ { "created": "Sun, 7 Sep 2008 23:09:06 GMT", "version": "v1" }, { "created": "Mon, 27 Oct 2008 19:41:16 GMT", "version": "v2" } ]

2009-11-13

[ [ "Meiser", "D.", "" ], [ "Holland", "M. J.", "" ] ]

Sun, 7 Sep 2008 23:09:06 GMT

2008-09-07T23:09:06Z

[ "physics" ]

arxiv

null

2210.01846

Moritz Nikolaus Laber

Moritz Laber, Peter Klimek, Martin Bruckner, Liuhuaying Yang, and Stefan Thurner

Shock propagation from the Russia-Ukraine conflict on international multilayer food production network determines global food availability

36 pages, 8 figures. Nat Food (2023)

null

10.1038/s43016-023-00771-4

null

econ.GN physics.soc-ph q-fin.EC

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

Dependencies in the global food production network can lead to shortages in numerous regions, as demonstrated by the impacts of the Russia-Ukraine conflict on global food supplies. Here, we reveal the losses of $125$ food products after a localized shock to agricultural production in $192$ countries and territories using a multilayer network model of trade (direct) and conversion of food products (indirect), thereby quantifying $10^8$ shock transmissions. We find that a complete agricultural production loss in Ukraine has heterogeneous impacts on other countries, causing relative losses of up to $89\%$ in sunflower oil and $85\%$ in maize via direct effects, and up to $25\%$ in poultry meat via indirect impacts. Whilst previous studies often treated products in isolation and did not account for product conversion during production, our model studies the global propagation of local supply shocks along both production and trade relations, allowing comparison of different response strategies.

[ { "created": "Tue, 4 Oct 2022 18:28:58 GMT", "version": "v1" }, { "created": "Mon, 8 May 2023 14:01:10 GMT", "version": "v2" }, { "created": "Fri, 16 Jun 2023 13:01:12 GMT", "version": "v3" } ]

2023-06-19

[ [ "Laber", "Moritz", "" ], [ "Klimek", "Peter", "" ], [ "Bruckner", "Martin", "" ], [ "Yang", "Liuhuaying", "" ], [ "Thurner", "Stefan", "" ] ]

Tue, 4 Oct 2022 18:28:58 GMT

2022-10-04T18:28:58Z

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

arxiv

null

1907.02829

Adam Brentnall

Adam R Brentnall, Jack Cuzick

Risk models for breast cancer and their validation

null

Statist. Sci., Volume 35, Number 1 (2020), 14-30

10.1214/19-STS729

null

stat.AP stat.ME

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

Strategies to prevent cancer and diagnose it early when it is most treatable are needed to reduce the public health burden from rising disease incidence. Risk assessment is playing an increasingly important role in targeting individuals in need of such interventions. For breast cancer many individual risk factors have been well understood for a long time, but the development of a fully comprehensive risk model has not been straightforward, in part because there have been limited data where joint effects of an extensive set of risk factors may be estimated with precision. In this article we first review the approach taken to develop the IBIS (Tyrer-Cuzick) model, and describe recent updates. We then review and develop methods to assess calibration of models such as this one, where the risk of disease allowing for competing mortality over a long follow-up time or lifetime is estimated. The breast cancer risk model model and calibration assessment methods are demonstrated using a cohort of 132 139 women attending mammography screening in Washington, USA.

[ { "created": "Fri, 5 Jul 2019 13:43:00 GMT", "version": "v1" }, { "created": "Tue, 9 Jul 2019 16:29:19 GMT", "version": "v2" } ]

2020-03-09

[ [ "Brentnall", "Adam R", "" ], [ "Cuzick", "Jack", "" ] ]

Fri, 5 Jul 2019 13:43:00 GMT

2019-07-05T13:43:00Z

[ "stat" ]

arxiv

null

{G}r{ASP}: A Library for Extracting and Exploring Human-Interpretable Textual Patterns

null

Data exploration is an important step of every data science and machine learning project, including those involving textual data. We provide a novel language tool, in the form of a publicly available Python library for extracting patterns from textual data. The library integrates a first public implementation of the existing GrASP algorithm. It allows users to extract patterns using a number of general-purpose built-in linguistic attributes (such as hypernyms, part-of-speech tags, and syntactic dependency tags), as envisaged for the original algorithm, as well as domain-specific custom attributes which can be incorporated into the library by implementing two functions. The library is equipped with a web-based interface empowering human users to conveniently explore data via the extracted patterns, using complementary pattern-centric and example-centric views: the former includes a reading in natural language and statistics of each extracted pattern; the latter shows applications of each extracted pattern to training examples. We demonstrate the usefulness of the library in classification (spam detection and argument mining), model analysis (machine translation), and artifact discovery in datasets (SNLI and 20Newsgroups).

null

acl

inproceedings

lertvittayakumjorn-etal-2022-grasp

Calzolari, Nicoletta and B{\'e}chet, Fr{\'e}d{\'e}ric and Blache, Philippe and Choukri, Khalid and Cieri, Christopher and Declerck, Thierry and Goggi, Sara and Isahara, Hitoshi and Maegaard, Bente and Mariani, Joseph and Mazo, H{\'e}l{\`e}ne and Odijk, Jan and Piperidis, Stelios

jun

2,022

Marseille, France

European Language Resources Association

https://aclanthology.org/2022.lrec-1.655/

Lertvittayakumjorn, Piyawat and Choshen, Leshem and Shnarch, Eyal and Toni, Francesca

Proceedings of the Thirteenth Language Resources and Evaluation Conference

6093--6103

null

1103.2021

Erwan Le Pennec

Serge Cohen, Erwan Le Pennec (INRIA Saclay - Ile de France, LM-Orsay)

Conditional Density Estimation by Penalized Likelihood Model Selection and Applications

No. RR-7596 (2011)

null

math.ST stat.TH

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

In this technical report, we consider conditional density estimation with a maximum likelihood approach. Under weak assumptions, we obtain a theoretical bound for a Kullback-Leibler type loss for a single model maximum likelihood estimate. We use a penalized model selection technique to select a best model within a collection. We give a general condition on penalty choice that leads to oracle type inequality for the resulting estimate. This construction is applied to two examples of partition-based conditional density models, models in which the conditional density depends only in a piecewise manner from the covariate. The first example relies on classical piecewise polynomial densities while the second uses Gaussian mixtures with varying mixing proportion but same mixture components. We show how this last case is related to an unsupervised segmentation application that has been the source of our motivation to this study.

[ { "created": "Thu, 10 Mar 2011 13:01:11 GMT", "version": "v1" }, { "created": "Fri, 8 Apr 2011 07:39:23 GMT", "version": "v2" }, { "created": "Mon, 8 Aug 2011 16:34:50 GMT", "version": "v3" }, { "created": "Sun, 11 Dec 2011 20:26:53 GMT", "version": "v4" }, { "created": "Mon, 9 Jul 2012 13:07:29 GMT", "version": "v5" } ]

2012-07-11

[ [ "Cohen", "Serge", "", "INRIA Saclay - Ile de France, LM-Orsay" ], [ "Pennec", "Erwan Le", "", "INRIA Saclay - Ile de France, LM-Orsay" ] ]

Thu, 10 Mar 2011 13:01:11 GMT

2011-03-10T13:01:11Z

[ "math", "stat" ]

arxiv

null

1804.03725

James Rankin

J. E. Rankin

Quaternionic Gauge Transformations and Yang-Mills Fields in Weyl Type Geometries

31 pages. The second half of the paper is new material. This uses some material from arXiv:1101.3606

null

RCTP-1801

gr-qc

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

This elementary discussion generalizes a Weyl geometry to allow quaternion valued gauge transformations and classical Yang-Mills geometric fields. This development will assume that the symmetric metric tensor is real in some gauge, and will develop the left and right handed approaches to quaternionic gauge transformations. Quaternionic gauge transformations are shown actually to require the shifting of some of Weyl's nonmetricity into torsion to define a properly transforming gauge field full curvature tensor, which is constructed as an asymmetric sum of left and right handed forms. Natural, gauge invariant, dimensionless variables are defined suitable for physics, and for use as a general formalism to describe these geometries, including General Relativity, in rather general circumstances. The geometry "self measures" these variables. Weyl's original action principle provides an example of an action rephrased in these gauge invariant variables, along with some unexpected possible insights on mechanics promoted by such a formulation of that action. Those include the torsion tensor and nonmetricity being constructed from mechanical energy-momentum. The Weyl form of action is then generalized to a quaternionic gauge field. The insights on mechanics now include spin 1/2 Dirac free fields. For physically reasonable choices of free parameters, the dimensionless Ricci tensor becomes nonnegligible in particle physics at distances much greater than the Planck length, along with limited general relativistic effects.

[ { "created": "Tue, 10 Apr 2018 21:24:13 GMT", "version": "v1" }, { "created": "Tue, 17 Apr 2018 18:02:03 GMT", "version": "v2" }, { "created": "Tue, 29 May 2018 03:25:35 GMT", "version": "v3" }, { "created": "Tue, 23 Oct 2018 16:52:04 GMT", "version": "v4" }, { "created": "Mon, 14 Jan 2019 19:42:19 GMT", "version": "v5" }, { "created": "Wed, 9 Oct 2019 02:08:37 GMT", "version": "v6" } ]

2019-10-10

[ [ "Rankin", "J. E.", "" ] ]

Tue, 10 Apr 2018 21:24:13 GMT

2018-04-10T21:24:13Z

[ "physics" ]

arxiv

null

2203.15206

Fangyuan Wang

Fangyuan Wang and Bo Xu

Shifted Chunk Encoder for Transformer Based Streaming End-to-End ASR

This paper has been accepted by iconip 2022, update the latest version

null

cs.SD cs.CL eess.AS

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

Currently, there are mainly three kinds of Transformer encoder based streaming End to End (E2E) Automatic Speech Recognition (ASR) approaches, namely time-restricted methods, chunk-wise methods, and memory-based methods. Generally, all of them have limitations in aspects of linear computational complexity, global context modeling, and parallel training. In this work, we aim to build a model to take all these three advantages for streaming Transformer ASR. Particularly, we propose a shifted chunk mechanism for the chunk-wise Transformer which provides cross-chunk connections between chunks. Therefore, the global context modeling ability of chunk-wise models can be significantly enhanced while all the original merits inherited. We integrate this scheme with the chunk-wise Transformer and Conformer, and identify them as SChunk-Transformer and SChunk-Conformer, respectively. Experiments on AISHELL-1 show that the SChunk-Transformer and SChunk-Conformer can respectively achieve CER 6.43% and 5.77%. And the linear complexity makes them possible to train with large batches and infer more efficiently. Our models can significantly outperform their conventional chunk-wise counterparts, while being competitive, with only 0.22 absolute CER drop, when compared with U2 which has quadratic complexity. A better CER can be achieved if compared with existing chunk-wise or memory-based methods, such as HS-DACS and MMA. Code is released.

[ { "created": "Tue, 29 Mar 2022 03:02:35 GMT", "version": "v1" }, { "created": "Wed, 15 Jun 2022 03:46:13 GMT", "version": "v2" }, { "created": "Mon, 26 Sep 2022 12:16:29 GMT", "version": "v3" } ]

2022-09-27

[ [ "Wang", "Fangyuan", "" ], [ "Xu", "Bo", "" ] ]

Tue, 29 Mar 2022 03:02:35 GMT

2022-03-29T03:02:35Z

[ "cs", "eess" ]

arxiv

null

Monitoring bacterial growth using tunable resistive pulse sensing with a pore-based technique.

null

A novel bacterial growth monitoring method using a tunable resistive pulse sensor (TRPS) system is introduced in this study for accurate and sensitive measurement of cell size and cell concentration simultaneously. Two model bacterial strains, Bacillus subtilis str.168 (BSU168) and Escherichia coli str.DH5α (DH5α), were chosen for benchmarking the growth-monitoring performance of the system. Results showed that the technique of TRPS is sensitive and accurate relative to widely used methods, with a lower detection limit of cell concentration measurement of 5 × 10⁵ cells/ml; at the same time, the mean coefficient of variation from TRPS was within 2 %. The growth of BSU168 and DH5α in liquid cultures was studied by TRPS, optical density (OD), and colony plating. Compared to OD measurement, TRPS-measured concentration correlates better with colony plating (R = 0.85 vs. R = 0.72), which is often regarded as the gold standard of cell concentration determination. General agreement was also observed by comparing TRPS-derived cell volume measurements and those determined from microscopy. We have demonstrated that TRPS is a reliable method for bacterial growth monitoring, where the study of both cell volume and cell concentration are needed to provide further details about the physical aspects of cell dynamics in real time.

null

pubmed

null

2,014

null

24,287,933

Applied microbiology and biotechnology

1705.05882

Marcel Nutz

Marcel Nutz, Jos\'e A. Scheinkman

Shorting in Speculative Markets

Forthcoming in 'Journal of Finance'

null

q-fin.MF math.OC q-fin.EC

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

We propose a continuous-time model of trading with heterogeneous beliefs. Risk-neutral agents face quadratic costs-of-carry on positions and thus their marginal valuations decrease with the size of their position, as it would be the case for risk-averse agents. In the equilibrium models of heterogeneous beliefs that followed Harrison-Kreps, investors are risk-neutral, short-selling is prohibited and agents face constant marginal costs of carrying positions. The resulting resale option guarantees that the price exceeds the price of the asset when speculation is ruled out; the difference is identified as a bubble. In our model increasing marginal costs entail that the price depends on asset supply. Second, agents also value an option to delay, and this may cause the market to equilibrate below the buy-and-hold price. Third, we introduce the possibility of short-selling. A Hamilton-Jacobi-Bellman equation of a novel form quantifies precisely the influence of the costs-of-carry on the price. An unexpected decrease in shorting costs may lead to the collapse of a bubble; this links the financial innovations that facilitated shorting of MBSs to the subsequent collapse of prices.

[ { "created": "Tue, 16 May 2017 19:36:56 GMT", "version": "v1" }, { "created": "Fri, 18 Aug 2017 16:05:00 GMT", "version": "v2" }, { "created": "Mon, 29 Jul 2019 22:54:40 GMT", "version": "v3" } ]

2019-07-31

[ [ "Nutz", "Marcel", "" ], [ "Scheinkman", "José A.", "" ] ]

Tue, 16 May 2017 19:36:56 GMT

2017-05-16T19:36:56Z

[ "math", "q-fin" ]

arxiv

null

2112.02822

Wang Miao

Wang Miao, Xinyu Li, Ping Zhang and Baoluo Sun

A stableness of resistance model for nonresponse adjustment with callback data

null

stat.ME

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

Nonresponse arises frequently in surveys and follow-ups are routinely made to increase the response rate. In order to monitor the follow-up process, callback data have been used in social sciences and survey studies for decades. In modern surveys, the availability of callback data is increasing because the response rate is decreasing and follow-ups are essential to collect maximum information. Although callback data are helpful to reduce the bias in surveys, such data have not been widely used in statistical analysis until recently. We propose a stableness of resistance assumption for nonresponse adjustment with callback data. We establish the identification and the semiparametric efficiency theory under this assumption, and propose a suite of semiparametric estimation methods including doubly robust estimators, which generalize existing parametric approaches for callback data analysis. We apply the approach to a Consumer Expenditure Survey dataset. The results suggest an association between nonresponse and high housing expenditures.

[ { "created": "Mon, 6 Dec 2021 07:19:19 GMT", "version": "v1" }, { "created": "Wed, 8 Dec 2021 12:05:05 GMT", "version": "v2" }, { "created": "Tue, 14 Feb 2023 12:31:23 GMT", "version": "v3" }, { "created": "Wed, 4 Sep 2024 13:49:11 GMT", "version": "v4" } ]

2024-09-05

[ [ "Miao", "Wang", "" ], [ "Li", "Xinyu", "" ], [ "Zhang", "Ping", "" ], [ "Sun", "Baoluo", "" ] ]

Mon, 6 Dec 2021 07:19:19 GMT

2021-12-06T07:19:19Z

[ "stat" ]

arxiv

null

2003.13526

Luca Demetrio

Luca Demetrio, Battista Biggio, Giovanni Lagorio, Fabio Roli, Alessandro Armando

Functionality-preserving Black-box Optimization of Adversarial Windows Malware

null

10.1109/TIFS.2021.3082330

null

cs.CR

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

Windows malware detectors based on machine learning are vulnerable to adversarial examples, even if the attacker is only given black-box query access to the model. The main drawback of these attacks is that: (i) they are query-inefficient, as they rely on iteratively applying random transformations to the input malware; and (ii) they may also require executing the adversarial malware in a sandbox at each iteration of the optimization process, to ensure that its intrusive functionality is preserved. In this paper, we overcome these issues by presenting a novel family of black-box attacks that are both query-efficient and functionality-preserving, as they rely on the injection of benign content - which will never be executed - either at the end of the malicious file, or within some newly-created sections. Our attacks are formalized as a constrained minimization problem which also enables optimizing the trade-off between the probability of evading detection and the size of the injected payload. We empirically investigate this trade-off on two popular static Windows malware detectors, and show that our black-box attacks can bypass them with only few queries and small payloads, even when they only return the predicted labels. We also evaluate whether our attacks transfer to other commercial antivirus solutions, and surprisingly find that they can evade, on average, more than 12 commercial antivirus engines. We conclude by discussing the limitations of our approach, and its possible future extensions to target malware classifiers based on dynamic analysis.

[ { "created": "Mon, 30 Mar 2020 14:56:01 GMT", "version": "v1" }, { "created": "Thu, 2 Apr 2020 07:46:51 GMT", "version": "v2" }, { "created": "Wed, 30 Sep 2020 15:16:04 GMT", "version": "v3" }, { "created": "Thu, 18 Feb 2021 15:43:31 GMT", "version": "v4" } ]

2021-05-20

[ [ "Demetrio", "Luca", "" ], [ "Biggio", "Battista", "" ], [ "Lagorio", "Giovanni", "" ], [ "Roli", "Fabio", "" ], [ "Armando", "Alessandro", "" ] ]

Mon, 30 Mar 2020 14:56:01 GMT

2020-03-30T14:56:01Z

[ "cs" ]

arxiv

null

Findings of the Third Workshop on Neural Generation and Translation

null

This document describes the findings of the Third Workshop on Neural Generation and Translation, held in concert with the annual conference of the Empirical Methods in Natural Language Processing (EMNLP 2019). First, we summarize the research trends of papers presented in the proceedings. Second, we describe the results of the two shared tasks 1) efficient neural machine translation (NMT) where participants were tasked with creating NMT systems that are both accurate and efficient, and 2) document generation and translation (DGT) where participants were tasked with developing systems that generate summaries from structured data, potentially with assistance from text in another language.

null

acl

inproceedings

hayashi-etal-2019-findings

Birch, Alexandra and Finch, Andrew and Hayashi, Hiroaki and Konstas, Ioannis and Luong, Thang and Neubig, Graham and Oda, Yusuke and Sudoh, Katsuhito

nov

2,019

Hong Kong

Association for Computational Linguistics

https://aclanthology.org/D19-5601/

Hayashi, Hiroaki and Oda, Yusuke and Birch, Alexandra and Konstas, Ioannis and Finch, Andrew and Luong, Minh-Thang and Neubig, Graham and Sudoh, Katsuhito

Proceedings of the 3rd Workshop on Neural Generation and Translation

1--14

null

Adapting Large Multilingual Machine Translation Models to Unseen Low Resource Languages via Vocabulary Substitution and Neuron Selection

null

We propose a method to adapt large Multilingual Machine Translation models to a low resource language (LRL) that was not included during the pre-training/training phases. We use neuron-ranking analysis to select neurons that are most influential to the high resource language (HRL) and fine-tune only this subset of the deep neural network`s neurons. We experiment with three mechanisms to compute such ranking. To allow for the potential difference in writing scripts between the HRL and LRL we utilize an alignment model to substitute HRL elements of the predefined vocab with appropriate LRL ones. Our method improves on both zero-shot and the stronger baseline of directly fine-tuning the model on the low-resource data by 3 BLEU points in X -{\ensuremath{>}} E and 1.6 points in E -{\ensuremath{>}} X.We also show that as we simulate smaller data amounts, the gap between our method and direct fine-tuning continues to widen.

null

acl

inproceedings

abdelghaffar-etal-2022-adapting

Duh, Kevin and Guzm{\'a}n, Francisco

sep

2,022

Orlando, USA

Association for Machine Translation in the Americas

https://aclanthology.org/2022.amta-research.22/

Abdelghaffar, Mohamed A and El Mogy, Amr and Sharaf, Nada Ahmed

Proceedings of the 15th biennial conference of the Association for Machine Translation in the Americas (Volume 1: Research Track)

287--297

null

2201.12655

Hugo Cui

Hugo Cui, Bruno Loureiro, Florent Krzakala, Lenka Zdeborov\'a

Error Scaling Laws for Kernel Classification under Source and Capacity Conditions

null

Mach. Learn.: Sci. Technol. (2023) 4 035033

10.1088/2632-2153/acf041

null

stat.ML cs.LG

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

We consider the problem of kernel classification. While worst-case bounds on the decay rate of the prediction error with the number of samples are known for some classifiers, they often fail to accurately describe the learning curves of real data sets. In this work, we consider the important class of data sets satisfying the standard source and capacity conditions, comprising a number of real data sets as we show numerically. Under the Gaussian design, we derive the decay rates for the misclassification (prediction) error as a function of the source and capacity coefficients. We do so for two standard kernel classification settings, namely margin-maximizing Support Vector Machines (SVM) and ridge classification, and contrast the two methods. We find that our rates tightly describe the learning curves for this class of data sets, and are also observed on real data. Our results can also be seen as an explicit prediction of the exponents of a scaling law for kernel classification that is accurate on some real datasets.

[ { "created": "Sat, 29 Jan 2022 20:39:58 GMT", "version": "v1" }, { "created": "Wed, 8 Jun 2022 14:02:15 GMT", "version": "v2" }, { "created": "Wed, 6 Sep 2023 12:20:43 GMT", "version": "v3" } ]

2023-09-07

[ [ "Cui", "Hugo", "" ], [ "Loureiro", "Bruno", "" ], [ "Krzakala", "Florent", "" ], [ "Zdeborová", "Lenka", "" ] ]

Sat, 29 Jan 2022 20:39:58 GMT

2022-01-29T20:39:58Z

[ "cs", "stat" ]

arxiv

null

2107.01819

Markus Hegland

Frank de Hoog and Markus Hegland

A Note on Error Bounds for Pseudo Skeleton Approximations of Matrices

8 pages, 1 figure

Lin. Alg. Apps, 2023

10.1016/j.laa.2023.03.024

null

math.NA cs.NA

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

Due to their importance in both data analysis and numerical algorithms, low rank approximations have recently been widely studied. They enable the handling of very large matrices. Tight error bounds for the computationally efficient Gaussian elimination based methods (skeleton approximations) are available. In practice, these bounds are useful for matrices with singular values which decrease quickly. Using the Chebyshev norm, this paper provides improved bounds for the errors of the matrix elements. These bounds are substantially better in the practically relevant cases where the eigenvalues decrease polynomially. Results are proven for general real rectangular matrices. Even stronger bounds are obtained for symmetric positive definite matrices. A simple example is given, comparing these new bounds to earlier ones.

[ { "created": "Mon, 5 Jul 2021 06:59:35 GMT", "version": "v1" }, { "created": "Sun, 14 Aug 2022 10:29:33 GMT", "version": "v2" } ]

2023-04-06

[ [ "de Hoog", "Frank", "" ], [ "Hegland", "Markus", "" ] ]

Mon, 5 Jul 2021 06:59:35 GMT

2021-07-05T06:59:35Z

[ "cs", "math" ]

arxiv

null

2011.14684

Francesco Salvetti

Simone Angarano, Vittorio Mazzia, Francesco Salvetti, Giovanni Fantin and Marcello Chiaberge

Robust Ultra-wideband Range Error Mitigation with Deep Learning at the Edge

Submitted to Engineering Applications of Artificial Intelligence

Engineering Applications of Artificial Intelligence, Volume 102, June 2021, 104278

10.1016/j.engappai.2021.104278

null

cs.LG cs.AI cs.RO eess.SP

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

Ultra-wideband (UWB) is the state-of-the-art and most popular technology for wireless localization. Nevertheless, precise ranging and localization in non-line-of-sight (NLoS) conditions is still an open research topic. Indeed, multipath effects, reflections, refractions, and complexity of the indoor radio environment can easily introduce a positive bias in the ranging measurement, resulting in highly inaccurate and unsatisfactory position estimation. This article proposes an efficient representation learning methodology that exploits the latest advancement in deep learning and graph optimization techniques to achieve effective ranging error mitigation at the edge. Channel Impulse Response (CIR) signals are directly exploited to extract high semantic features to estimate corrections in either NLoS or LoS conditions. Extensive experimentation with different settings and configurations has proved the effectiveness of our methodology and demonstrated the feasibility of a robust and low computational power UWB range error mitigation.

[ { "created": "Mon, 30 Nov 2020 10:52:21 GMT", "version": "v1" }, { "created": "Wed, 28 Apr 2021 09:16:00 GMT", "version": "v2" } ]

2021-05-06

[ [ "Angarano", "Simone", "" ], [ "Mazzia", "Vittorio", "" ], [ "Salvetti", "Francesco", "" ], [ "Fantin", "Giovanni", "" ], [ "Chiaberge", "Marcello", "" ] ]

Mon, 30 Nov 2020 10:52:21 GMT

2020-11-30T10:52:21Z

[ "cs", "eess" ]

arxiv

null

1511.05634

Carlo Baldassi

Carlo Baldassi, Alessandro Ingrosso, Carlo Lucibello, Luca Saglietti and Riccardo Zecchina

Local entropy as a measure for sampling solutions in Constraint Satisfaction Problems

46 pages (main text: 22), 7 figures. This is an author-created, un-copyedited version of an article published in Journal of Statistical Mechanics: Theory and Experiment. IOP Publishing Ltd is not responsible for any errors or omissions in this version of the manuscript or any version derived from it. The Version of Record is available online at http://dx.doi.org/10.1088/1742-5468/2016/02/023301

J. Stat. Mech. 2016 (2) 023301

10.1088/1742-5468/2016/02/023301

null

cond-mat.dis-nn stat.ML

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

We introduce a novel Entropy-driven Monte Carlo (EdMC) strategy to efficiently sample solutions of random Constraint Satisfaction Problems (CSPs). First, we extend a recent result that, using a large-deviation analysis, shows that the geometry of the space of solutions of the Binary Perceptron Learning Problem (a prototypical CSP), contains regions of very high-density of solutions. Despite being sub-dominant, these regions can be found by optimizing a local entropy measure. Building on these results, we construct a fast solver that relies exclusively on a local entropy estimate, and can be applied to general CSPs. We describe its performance not only for the Perceptron Learning Problem but also for the random $K$-Satisfiabilty Problem (another prototypical CSP with a radically different structure), and show numerically that a simple zero-temperature Metropolis search in the smooth local entropy landscape can reach sub-dominant clusters of optimal solutions in a small number of steps, while standard Simulated Annealing either requires extremely long cooling procedures or just fails. We also discuss how the EdMC can heuristically be made even more efficient for the cases we studied.

[ { "created": "Wed, 18 Nov 2015 01:03:59 GMT", "version": "v1" }, { "created": "Thu, 25 Feb 2016 18:54:40 GMT", "version": "v2" } ]

2016-02-26

[ [ "Baldassi", "Carlo", "" ], [ "Ingrosso", "Alessandro", "" ], [ "Lucibello", "Carlo", "" ], [ "Saglietti", "Luca", "" ], [ "Zecchina", "Riccardo", "" ] ]

Wed, 18 Nov 2015 01:03:59 GMT

2015-11-18T01:03:59Z

[ "physics", "stat" ]

arxiv

null

2202.07270

Reem Yassawi

Val\'erie Berth\'e and Paulina Cecchi Bernales and Reem Yassawi

Coboundaries and eigenvalues of finitary S-adic systems

Minor revisions. 41 pages

null

math.DS

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

An S-adic system is a symbolic dynamical system generated by iterating an infinite sequence of substitutions or morphisms, called a directive sequence. A finitary S-adic dynamical system is one where the directive sequence consists of morphisms selected from a finite set. We study eigenvalues and coboundaries for finitary recognizable S-adic dynamical systems, i.e., those where points can be uniquely desubstituted using the given sequence of morphisms. To do this we identify the notions of straightness and essential words, and use them to define a coboundary, inspired by of Host's formalism, which allows us to express necessary and sufficient conditions that a complex number must satisfy in order to be a continuous or measurable eigenvalue. We then apply our results to finitary directive sequences of substitutions of constant length, and show how to create constant-length $S$-adic shifts with non-trivial coboundaries. We show that in this case all continuous eigenvalues are rational and we give a complete description of the rationals that can be an eigenvalue, indicating how this leads to a Cobham-style result for these systems.

[ { "created": "Tue, 15 Feb 2022 09:25:04 GMT", "version": "v1" }, { "created": "Thu, 4 Apr 2024 19:52:05 GMT", "version": "v2" }, { "created": "Mon, 27 Jan 2025 22:57:20 GMT", "version": "v3" } ]

2025-01-29

[ [ "Berthé", "Valérie", "" ], [ "Bernales", "Paulina Cecchi", "" ], [ "Yassawi", "Reem", "" ] ]

Tue, 15 Feb 2022 09:25:04 GMT

2022-02-15T09:25:04Z

[ "math" ]

arxiv

null

Cardiac biomarkers: new tools for heart failure management.

null

The last decade has seen exciting advances in the field of biomarkers used in managing patients with heart failure (HF). Biomarker research has broadened our knowledge base, shedding more light on the underlying pathophysiological mechanisms occurring in patients with both acute and chronic HF. The criterion required by an ideal cardiovascular biomarker has been progressively changing to an era of sensitive assays that can be used to guide treatment. Recent technological advances have made it possible to rapidly measure even minute amounts of these proteins by means of higher sensitivity assays. With a high prevalence of comorbidities associated with HF, an integrated approach utilizing multiple biomarkers have shown promise in predicting mortality, better risk stratification and reducing re-hospitalizations, thus lowering health-care costs. This review provides a brief insight into recent advances in the field of biomarkers currently used in the diagnosis and prognosis of patients with acute and chronic HF.

null

pubmed

null

2,012

null

24,282,708

Cardiovascular diagnosis and therapy

null

Drug use and risk among youth in different rural contexts.

null

This study compared levels of drug use and risk and protective factors among 18,767 adolescent youths from communities of less than 50,000 in population living either on farms, in the country but not on farms, or in towns. Current alcohol use, smokeless tobacco use, inhalant use, and other illicit drug use were more prevalent among high school-aged youths living on farms than among those living in towns. Prevalence of drug use did not significantly vary across youths living in different residential contexts among middle school youths. While risk and protective factors showed associations of similar magnitude with drug use across residential location, high school students living on farms were exposed to greater numbers of risk factors across multiple domains than were students living in towns. The findings suggest that outreach to farm-dwelling youths may be particularly important for interventions seeking to prevent adolescent drug use in rural settings.

null

pubmed

null

2,011

null

21,414,831

Health & place