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

null

{ALEM} at {CASE} 2021 Task 1: Multilingual Text Classification on News Articles

null

We participated CASE shared task in ACL-IJCNLP 2021. This paper is a summary of our experiments and ideas about this shared task. For each subtask we shared our approach, successful and failed methods and our thoughts about them. We submit our results once for every subtask, except for subtask3, in task submission system and present scores based on our validation set formed from given training samples in this paper. Techniques and models we mentioned includes BERT, Multilingual BERT, oversampling, undersampling, data augmentation and their implications with each other. Most of the experiments we came up with were not completed, as time did not permit, but we share them here as we plan to do them as suggested in the future work part of document.

null

acl

inproceedings

gurel-emin-2021-alem

H{\"urriyeto{\u{glu, Ali

aug

2,021

Online

Association for Computational Linguistics

https://aclanthology.org/2021.case-1.19/

G{\"urel, Alaeddin and Emin, Emre

Proceedings of the 4th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2021)

147--151

null

A preliminary survey for Wolbachia and bacteriophage WO infections in Indian mosquitoes (Diptera: Culicidae).

null

Maternally inherited Wolbachia endosymbiotic bacteria are known to induce various kinds of reproductive alterations in their arthropod hosts. It has been proposed that this bacterium can be used as a tool for gene drive system in mosquitoes and also for the reduction of population size and modulating population age structure in order to reduce disease transmission. In the present study, we carried out a survey to determine the prevalence of Wolbachia and its phage WO infection in Indian mosquitoes and classified Wolbachia infection into groups A and B based on extensive polymerase chain reaction assay using Wolbachia specific wsp and orf7 gene primers. Out of 20 fieldcaught mosquito species, eight species have shown to be infected. Singly infected with Wolbachia A was found in two species and B group found in four species, while double infection with AB group were found in two species. All the screened mosquito species with positive Wolbachia infection were also infected with phage WO. The knowledge of variation in Wolbachia and phage WO infection rates and inferred susceptibility to infection among different mosquito genera has fundamental implications for designing and successful application of Wolbachia based vector-borne disease control strategies.

null

pubmed

null

2,010

null

21,399,578

Tropical biomedicine

2205.15150

Hoang Thien Ly

Thu Nguyen, Hoang Thien Ly, Michael Alexander Riegler, P{\aa}l Halvorsen, Hugo L. Hammer

Principal Component Analysis based frameworks for efficient missing data imputation algorithms

null

cs.LG stat.ML

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

Missing data is a commonly occurring problem in practice. Many imputation methods have been developed to fill in the missing entries. However, not all of them can scale to high-dimensional data, especially the multiple imputation techniques. Meanwhile, the data nowadays tends toward high-dimensional. Therefore, in this work, we propose Principal Component Analysis Imputation (PCAI), a simple but versatile framework based on Principal Component Analysis (PCA) to speed up the imputation process and alleviate memory issues of many available imputation techniques, without sacrificing the imputation quality in term of MSE. In addition, the frameworks can be used even when some or all of the missing features are categorical, or when the number of missing features is large. Next, we introduce PCA Imputation - Classification (PIC), an application of PCAI for classification problems with some adjustments. We validate our approach by experiments on various scenarios, which shows that PCAI and PIC can work with various imputation algorithms, including the state-of-the-art ones and improve the imputation speed significantly, while achieving competitive mean square error/classification accuracy compared to direct imputation (i.e., impute directly on the missing data).

[ { "created": "Mon, 30 May 2022 14:47:27 GMT", "version": "v1" }, { "created": "Tue, 30 Aug 2022 10:35:29 GMT", "version": "v2" }, { "created": "Sun, 19 Mar 2023 18:20:39 GMT", "version": "v3" } ]

2023-03-21

[ [ "Nguyen", "Thu", "" ], [ "Ly", "Hoang Thien", "" ], [ "Riegler", "Michael Alexander", "" ], [ "Halvorsen", "Pål", "" ], [ "Hammer", "Hugo L.", "" ] ]

Mon, 30 May 2022 14:47:27 GMT

2022-05-30T14:47:27Z

[ "cs", "stat" ]

arxiv

null

0710.4133

Alex Chambers

Alex Chambers and Arttu Rajantie

Lattice calculation of non-Gaussianity from preheating

4 pages, 1 figure; erratum added

Phys.Rev.Lett.100:041302,2008; Erratum-ibid.101:149903,2008

10.1103/PhysRevLett.100.041302

Imperial/TP/07/AC/01

astro-ph hep-ph

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

If light scalar fields are present at the end of inflation, their non-equilibrium dynamics such as parametric resonance or a phase transition can produce non-Gaussian density perturbations. We show how these perturbations can be calculated using non-linear lattice field theory simulations and the separate universe approximation. In the massless preheating model, we find that some parameter values are excluded while others lead to acceptable but observable levels of non-Gaussianity. This shows that preheating can be an important factor in assessing the viability of inflationary models.

[ { "created": "Mon, 22 Oct 2007 20:00:11 GMT", "version": "v1" }, { "created": "Tue, 22 Jan 2008 13:38:54 GMT", "version": "v2" }, { "created": "Tue, 30 Sep 2008 10:14:52 GMT", "version": "v3" } ]

2008-11-26

[ [ "Chambers", "Alex", "" ], [ "Rajantie", "Arttu", "" ] ]

Mon, 22 Oct 2007 20:00:11 GMT

2007-10-22T20:00:11Z

[ "physics" ]

arxiv

null

2203.06056

Nikolaj Thams

Nikolaj Thams and Rikke S{\o}ndergaard and Sebastian Weichwald and Jonas Peters

Identifying Causal Effects using Instrumental Time Series: Nuisance IV and Correcting for the Past

null

stat.ME stat.ML

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

Instrumental variable (IV) regression relies on instruments to infer causal effects from observational data with unobserved confounding. We consider IV regression in time series models, such as vector auto-regressive (VAR) processes. Direct applications of i.i.d. techniques are generally inconsistent as they do not correctly adjust for dependencies in the past. In this paper, we outline the difficulties that arise due to time structure and propose methodology for constructing identifying equations that can be used for consistent parametric estimation of causal effects in time series data. One method uses extra nuisance covariates to obtain identifiability (an idea that can be of interest even in the i.i.d. case). We further propose a graph marginalization framework that allows us to apply nuisance IV and other IV methods in a principled way to time series. Our methods make use of a version of the global Markov property, which we prove holds for VAR(p) processes. For VAR(1) processes, we prove identifiability conditions that relate to Jordan forms and are different from the well-known rank conditions in the i.i.d. case (they do not require as many instruments as covariates, for example). We provide methods, prove their consistency, and show how the inferred causal effect can be used for distribution generalization. Simulation experiments corroborate our theoretical results. We provide ready-to-use Python code.

[ { "created": "Fri, 11 Mar 2022 16:29:48 GMT", "version": "v1" }, { "created": "Fri, 15 Mar 2024 20:13:03 GMT", "version": "v2" }, { "created": "Fri, 19 Jul 2024 20:48:15 GMT", "version": "v3" } ]

2024-07-23

[ [ "Thams", "Nikolaj", "" ], [ "Søndergaard", "Rikke", "" ], [ "Weichwald", "Sebastian", "" ], [ "Peters", "Jonas", "" ] ]

Fri, 11 Mar 2022 16:29:48 GMT

2022-03-11T16:29:48Z

[ "stat" ]

arxiv

null

cond-mat/0303578

Peter Mueller

Matthias K\"untzel, Henning L\"owe, Peter M\"uller and Annette Zippelius

Diffusion of gelation clusters in the Zimm model

7 pages, 4 figures; slightly expanded version, typos corrected; to appear in Eur. Phys. J. E

Eur. Phys. J. E 12, 325--331 (2003)

10.1140/epje/i2003-10066-x

null

cond-mat.stat-mech cond-mat.soft

null

Starting from a Zimm model we study selfdiffusion in a solution of crosslinked monomers. We focus on the effects of the hydrodynamic interaction on the dynamics and the critical behaviour at the sol-gel-point. Hydrodynamic interactions cause the clusters' diffusion constant to depend not only on the cluster's size but also on the cluster's shape -- in contrast to the Rouse model. This gives rise to a nontrivial scaling of the Kirkwood diffusion constant averaged over all clusters of fixed size $n$, ${\hat D}_n\sim n^{-{\hat b}}$ with ${\hat b}=1/d_s-1/2$ given in terms of the spectral dimension $d_s$ of critical percolation clusters. The long-time decay of the incoherent scattering function is determined by the diffusive motion of the largest clusters. This implies the critical vanishing $D_{\rm eff}\sim \epsilon^a$ of the cluster-averaged effective diffusion constant at the gel point with exponent $a = (3/2 -\tau +1/d_{s})/\sigma$.

[ { "created": "Thu, 27 Mar 2003 12:21:49 GMT", "version": "v1" }, { "created": "Thu, 9 Oct 2003 10:08:41 GMT", "version": "v2" } ]

2007-05-23

[ [ "Küntzel", "Matthias", "" ], [ "Löwe", "Henning", "" ], [ "Müller", "Peter", "" ], [ "Zippelius", "Annette", "" ] ]

Thu, 27 Mar 2003 12:21:49 GMT

2003-03-27T12:21:49Z

[ "physics" ]

arxiv

null

1201.1728

Italo Dejter Prof

Italo J. Dejter

Worst-case efficient dominating sets in digraphs

13 pages, 3 figures

null

math.CO cs.IT math.IT

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

Let $1\le n\in\Z$. {\it Worst-case efficient dominating sets in digraphs} are conceived so that their presence in certain strong digraphs $\vec{ST}_n$ corresponds to that of efficient dominating sets in star graphs $ST_n$: The fact that the star graphs $ST_n$ form a so-called dense segmental neighborly E-chain is reflected in a corresponding fact for the digraphs $\vec{ST}_n$. Related chains of graphs and open problems are presented as well.

[ { "created": "Mon, 9 Jan 2012 10:25:11 GMT", "version": "v1" }, { "created": "Wed, 11 Jan 2012 18:49:57 GMT", "version": "v2" }, { "created": "Tue, 26 Jun 2012 20:59:00 GMT", "version": "v3" } ]

2012-06-28

[ [ "Dejter", "Italo J.", "" ] ]

Mon, 9 Jan 2012 10:25:11 GMT

2012-01-09T10:25:11Z

[ "cs", "math" ]

arxiv

null

Altered prostanoid signaling contributes to increased skin tumorigenesis in Tpl2 knockout mice.

null

Squamous cell carcinoma is the second most common form of skin cancer with the incidence expected to double over the next 20 years. Inflammation is believed to be a critical component in skin cancer progression. Therefore, understanding genes involved in the regulation of inflammatory pathways is vital to the design of targeted therapies. Numerous studies show cyclooxygenases (COXs) play an essential role in inflammation-associated cancers. Tpl2 (MAP3K8) is a protein kinase in the MAP Kinase signal transduction cascade. Previous research using a two-stage skin carcinogenesis model revealed that Tpl2(-/-) mice have significantly higher tumor incidence and inflammatory response than wild-type (WT) controls. The current study investigates whether cyclooxygenase-2 (COX-2) and COX-2- regulated prostaglandins and prostaglandin receptors drive the highly tumorigenic state of Tpl2(-/-) mice by investigating the relationship between Tpl2 and COX-2. Keratinocytes from newborn WT or Tpl2(-/-) mice were treated with 12-O-tetradecanoylphorbol-13-acetate (TPA) for various times over 24 hours. Western analysis revealed significant differences in COX-2 and COX-2 dependent prostanoids and prostanoid receptors. Additionally, in vivo experiments confirmed that COX-2 and COX-2 downstream factors were elevated in TPA-treated Tpl2(-/-) skin, as well as in papillomas from Tpl2(-/-) mice. Use of the selective COX-2 inhibitor Celecoxib showed the increased tumorigenesis in the Tpl2(-/-) mice to primarily be mediated through COX-2. These experiments illustrate COX-2 induction in the absence of Tpl2 may be responsible for the increased tumorigenesis found in Tpl2(-/-) mice. Defining the relationship between Tpl2 and COX-2 may lead to new ways to downregulate COX-2 through the modulation of Tpl2.

null

pubmed

null

2,013

null

23,457,529

PloS one

2211.11087

Li S. Yifei

Li S. Yifei, Lyle Ungar, Jo\~ao Sedoc

Conceptor-Aided Debiasing of Large Language Models

25 pages

null

cs.CL cs.AI

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

Pre-trained large language models (LLMs) reflect the inherent social biases of their training corpus. Many methods have been proposed to mitigate this issue, but they often fail to debias or they sacrifice model accuracy. We use conceptors--a soft projection method--to identify and remove the bias subspace in LLMs such as BERT and GPT. We propose two methods of applying conceptors (1) bias subspace projection by post-processing by the conceptor NOT operation; and (2) a new architecture, conceptor-intervened BERT (CI-BERT), which explicitly incorporates the conceptor projection into all layers during training. We find that conceptor post-processing achieves state-of-the-art (SoTA) debiasing results while maintaining LLMs' performance on the GLUE benchmark. Further, it is robust in various scenarios and can mitigate intersectional bias efficiently by its AND operation on the existing bias subspaces. Although CI-BERT's training takes all layers' bias into account and can beat its post-processing counterpart in bias mitigation, CI-BERT reduces the language model accuracy. We also show the importance of carefully constructing the bias subspace. The best results are obtained by removing outliers from the list of biased words, combining them (via the OR operation), and computing their embeddings using the sentences from a cleaner corpus.

[ { "created": "Sun, 20 Nov 2022 21:24:48 GMT", "version": "v1" }, { "created": "Wed, 24 May 2023 04:04:06 GMT", "version": "v2" }, { "created": "Mon, 30 Oct 2023 22:00:23 GMT", "version": "v3" } ]

2023-11-01

[ [ "Yifei", "Li S.", "" ], [ "Ungar", "Lyle", "" ], [ "Sedoc", "João", "" ] ]

Sun, 20 Nov 2022 21:24:48 GMT

2022-11-20T21:24:48Z

[ "cs" ]

arxiv

null

Evaluation of Online Dialogue Policy Learning Techniques

null

The number of applied Dialogue Systems is ever increasing in several service providing and other applications as a way to efficiently and inexpensively serve large numbers of customers. A DS that employs some form of adaptation to the environment and its users is called an Adaptive Dialogue System (ADS). A significant part of the research community has lately focused on ADS and many existing or novel techniques are being applied to this problem. One of the most promising techniques is Reinforcement Learning (RL) and especially online RL. This paper focuses on online RL techniques used to achieve adaptation in Dialogue Management and provides an evaluation of various such methods in an effort to aid the designers of ADS in deciding which method to use. To the best of our knowledge there is no other work to compare online RL techniques on the dialogue management problem.

null

acl

inproceedings

papangelis-etal-2012-evaluation

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/L12-1126/

Papangelis, Alexandros and Karkaletsis, Vangelis and Makedon, Fillia

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

1410--1415

null

1103.2177

Harpreet S. Dhillon

Harpreet S. Dhillon, Radha Krishna Ganti, Francois Baccelli and Jeffrey G. Andrews

Modeling and Analysis of K-Tier Downlink Heterogeneous Cellular Networks

IEEE Journal on Selected Areas in Communications, vol. 30, no. 3, pp. 550 - 560, Apr. 2012

null

10.1109/JSAC.2012.120405

null

cs.IT math.IT

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

Cellular networks are in a major transition from a carefully planned set of large tower-mounted base-stations (BSs) to an irregular deployment of heterogeneous infrastructure elements that often additionally includes micro, pico, and femtocells, as well as distributed antennas. In this paper, we develop a tractable, flexible, and accurate model for a downlink heterogeneous cellular network (HCN) consisting of K tiers of randomly located BSs, where each tier may differ in terms of average transmit power, supported data rate and BS density. Assuming a mobile user connects to the strongest candidate BS, the resulting Signal-to-Interference-plus-Noise-Ratio (SINR) is greater than 1 when in coverage, Rayleigh fading, we derive an expression for the probability of coverage (equivalently outage) over the entire network under both open and closed access, which assumes a strikingly simple closed-form in the high SINR regime and is accurate down to -4 dB even under weaker assumptions. For external validation, we compare against an actual LTE network (for tier 1) with the other K-1 tiers being modeled as independent Poisson Point Processes. In this case as well, our model is accurate to within 1-2 dB. We also derive the average rate achieved by a randomly located mobile and the average load on each tier of BSs. One interesting observation for interference-limited open access networks is that at a given SINR, adding more tiers and/or BSs neither increases nor decreases the probability of coverage or outage when all the tiers have the same target-SINR.

[ { "created": "Fri, 11 Mar 2011 00:51:27 GMT", "version": "v1" }, { "created": "Mon, 14 Mar 2011 21:48:55 GMT", "version": "v2" }, { "created": "Thu, 29 Mar 2012 04:09:57 GMT", "version": "v3" }, { "created": "Sun, 23 Dec 2012 23:09:23 GMT", "version": "v4" } ]

2016-11-18

[ [ "Dhillon", "Harpreet S.", "" ], [ "Ganti", "Radha Krishna", "" ], [ "Baccelli", "Francois", "" ], [ "Andrews", "Jeffrey G.", "" ] ]

Fri, 11 Mar 2011 00:51:27 GMT

2011-03-11T00:51:27Z

[ "cs", "math" ]

arxiv

null

2204.01087

Amin Dehghani

Amin Dehghani, Hamid Soltanian-Zadeh, Gholam-Ali Hossein-Zadeh

Neural modulation enhancement using connectivity-based EEG neurofeedback with simultaneous fMRI for emotion regulation

12pages, 5 figures

null

q-bio.NC

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

Emotion regulation plays a key role in human behavior and overall well-being. Neurofeedback is a non-invasive self-brain training technique used for emotion regulation to enhance brain function and treatment of mental disorders through behavioral changes. Previous neurofeedback research often focused on using activity from a single brain region as measured by fMRI or power from one or two EEG electrodes. In a new study, we employed connectivity-based EEG neurofeedback through recalling positive autobiographical memories and simultaneous fMRI to upregulate positive emotion. In our novel approach, the feedback was determined by the coherence of EEG electrodes rather than the power of one or two electrodes. We compared the efficiency of this connectivity-based neurofeedback to traditional activity-based neurofeedback through multiple experiments. The results showed that connectivity-based neurofeedback effectively improved BOLD signal change and connectivity in key emotion regulation regions such as the amygdala, thalamus, and insula, and increased EEG frontal asymmetry, which is a biomarker for emotion regulation and treatment of mental disorders such as PTSD, anxiety, and depression and coherence among EEG channels. The psychometric evaluations conducted both before and after the neurofeedback experiments revealed that participants demonstrated improvements in enhancing positive emotions and reducing negative emotions when utilizing connectivity-based neurofeedback, as compared to traditional activity-based and sham neurofeedback approaches. These findings suggest that connectivity-based neurofeedback may be a superior method for regulating emotions and could be a useful alternative therapy for mental disorders, providing individuals with greater control over their brain and mental functions.

[ { "created": "Sun, 3 Apr 2022 15:06:46 GMT", "version": "v1" }, { "created": "Mon, 14 Aug 2023 15:27:49 GMT", "version": "v2" }, { "created": "Fri, 3 Nov 2023 01:59:56 GMT", "version": "v3" } ]

2023-11-06

[ [ "Dehghani", "Amin", "" ], [ "Soltanian-Zadeh", "Hamid", "" ], [ "Hossein-Zadeh", "Gholam-Ali", "" ] ]

Sun, 3 Apr 2022 15:06:46 GMT

2022-04-03T15:06:46Z

[ "q-bio" ]

arxiv

null

2007.03492

Nicolas Grelier

Computing a maximum clique in geometric superclasses of disk graphs

null

cs.CG

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

In the 90's Clark, Colbourn and Johnson wrote a seminal paper where they proved that maximum clique can be solved in polynomial time in unit disk graphs. Since then, the complexity of maximum clique in intersection graphs of d-dimensional (unit) balls has been investigated. For ball graphs, the problem is NP-hard, as shown by Bonamy et al. (FOCS '18). They also gave an efficient polynomial time approximation scheme (EPTAS) for disk graphs. However, the complexity of maximum clique in this setting remains unknown. In this paper, we show the existence of a polynomial time algorithm for a geometric superclass of unit disk graphs. Moreover, we give partial results toward obtaining an EPTAS for intersection graphs of convex pseudo-disks.

[ { "created": "Tue, 7 Jul 2020 14:27:34 GMT", "version": "v1" }, { "created": "Tue, 8 Dec 2020 11:23:17 GMT", "version": "v2" }, { "created": "Mon, 26 Jul 2021 08:05:53 GMT", "version": "v3" } ]

2021-07-27

[ [ "Grelier", "Nicolas", "" ] ]

Tue, 7 Jul 2020 14:27:34 GMT

2020-07-07T14:27:34Z

[ "cs" ]

arxiv

null

2112.10224

Eugene Ndiaye

Stable Conformal Prediction Sets

Update to ICML version

Proceedings of the 39-th International Conference on Machine Learning, 2022

null

stat.ML cs.LG stat.CO

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

When one observes a sequence of variables $(x_1, y_1), \ldots, (x_n, y_n)$, Conformal Prediction (CP) is a methodology that allows to estimate a confidence set for $y_{n+1}$ given $x_{n+1}$ by merely assuming that the distribution of the data is exchangeable. CP sets have guaranteed coverage for any finite population size $n$. While appealing, the computation of such a set turns out to be infeasible in general, e.g. when the unknown variable $y_{n+1}$ is continuous. The bottleneck is that it is based on a procedure that readjusts a prediction model on data where we replace the unknown target by all its possible values in order to select the most probable one. This requires computing an infinite number of models, which often makes it intractable. In this paper, we combine CP techniques with classical algorithmic stability bounds to derive a prediction set computable with a single model fit. We demonstrate that our proposed confidence set does not lose any coverage guarantees while avoiding the need for data splitting as currently done in the literature. We provide some numerical experiments to illustrate the tightness of our estimation when the sample size is sufficiently large, on both synthetic and real datasets.

[ { "created": "Sun, 19 Dec 2021 18:53:32 GMT", "version": "v1" }, { "created": "Wed, 7 Dec 2022 02:42:46 GMT", "version": "v2" } ]

2022-12-08

[ [ "Ndiaye", "Eugene", "" ] ]

Sun, 19 Dec 2021 18:53:32 GMT

2021-12-19T18:53:32Z

[ "cs", "stat" ]

arxiv

null

Do you have the right scissors? Tailoring Pre-trained Language Models via {M}onte-{C}arlo Methods

null

It has been a common approach to pre-train a language model on a large corpus and fine-tune it on task-specific data. In practice, we observe that fine-tuning a pre-trained model on a small dataset may lead to over- and/or under-estimate problem. In this paper, we propose MC-Tailor, a novel method to alleviate the above issue in text generation tasks by truncating and transferring the probability mass from over-estimated regions to under-estimated ones. Experiments on a variety of text generation datasets show that MC-Tailor consistently and significantly outperforms the fine-tuning approach.

null

acl

inproceedings

miao-etal-2020-right

Jurafsky, Dan and Chai, Joyce and Schluter, Natalie and Tetreault, Joel

jul

2,020

Online

Association for Computational Linguistics

https://aclanthology.org/2020.acl-main.314/

Miao, Ning and Song, Yuxuan and Zhou, Hao and Li, Lei

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

3436--3441

null

2104.05831

Gabriel Suarez

Gabriel Suarez, Juan Raful, Maria A. Luque, Carlos F. Valencia, Alejandro Correa-Bahnsen

Enhancing User' s Income Estimation with Super-App Alternative Data

5 pages, 2 figures

null

cs.LG q-fin.RM

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

This paper presents the advantages of alternative data from Super-Apps to enhance user' s income estimation models. It compares the performance of these alternative data sources with the performance of industry-accepted bureau income estimators that takes into account only financial system information; successfully showing that the alternative data manage to capture information that bureau income estimators do not. By implementing the TreeSHAP method for Stochastic Gradient Boosting Interpretation, this paper highlights which of the customer' s behavioral and transactional patterns within a Super-App have a stronger predictive power when estimating user' s income. Ultimately, this paper shows the incentive for financial institutions to seek to incorporate alternative data into constructing their risk profiles.

[ { "created": "Mon, 12 Apr 2021 21:34:44 GMT", "version": "v1" }, { "created": "Fri, 30 Jul 2021 22:28:27 GMT", "version": "v2" }, { "created": "Tue, 3 Aug 2021 14:21:40 GMT", "version": "v3" } ]

2021-08-04

[ [ "Suarez", "Gabriel", "" ], [ "Raful", "Juan", "" ], [ "Luque", "Maria A.", "" ], [ "Valencia", "Carlos F.", "" ], [ "Correa-Bahnsen", "Alejandro", "" ] ]

Mon, 12 Apr 2021 21:34:44 GMT

2021-04-12T21:34:44Z

[ "cs", "q-fin" ]

arxiv

null

1905.02939

Saifuddin Syed

Saifuddin Syed, Alexandre Bouchard-C\^ot\'e, George Deligiannidis, Arnaud Doucet

Non-Reversible Parallel Tempering: a Scalable Highly Parallel MCMC Scheme

74 pages, 30 figures. The method is implemented in an open source probabilistic programming available at https://github.com/UBC-Stat-ML/blangSDK

null

10.1111/rssb.12464

null

stat.CO

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

Parallel tempering (PT) methods are a popular class of Markov chain Monte Carlo schemes used to sample complex high-dimensional probability distributions. They rely on a collection of $N$ interacting auxiliary chains targeting tempered versions of the target distribution to improve the exploration of the state-space. We provide here a new perspective on these highly parallel algorithms and their tuning by identifying and formalizing a sharp divide in the behaviour and performance of reversible versus non-reversible PT schemes. We show theoretically and empirically that a class of non-reversible PT methods dominates its reversible counterparts and identify distinct scaling limits for the non-reversible and reversible schemes, the former being a piecewise-deterministic Markov process and the latter a diffusion. These results are exploited to identify the optimal annealing schedule for non-reversible PT and to develop an iterative scheme approximating this schedule. We provide a wide range of numerical examples supporting our theoretical and methodological contributions. The proposed methodology is applicable to sample from a distribution $\pi$ with a density $L$ with respect to a reference distribution $\pi_0$ and compute the normalizing constant. A typical use case is when $\pi_0$ is a prior distribution, $L$ a likelihood function and $\pi$ the corresponding posterior.

[ { "created": "Wed, 8 May 2019 07:22:30 GMT", "version": "v1" }, { "created": "Mon, 7 Oct 2019 03:21:25 GMT", "version": "v2" }, { "created": "Sun, 22 Nov 2020 23:48:17 GMT", "version": "v3" }, { "created": "Mon, 26 Jul 2021 23:28:08 GMT", "version": "v4" } ]

2021-07-28

[ [ "Syed", "Saifuddin", "" ], [ "Bouchard-Côté", "Alexandre", "" ], [ "Deligiannidis", "George", "" ], [ "Doucet", "Arnaud", "" ] ]

Wed, 8 May 2019 07:22:30 GMT

2019-05-08T07:22:30Z

[ "stat" ]

arxiv

null

{FAKTA}: An Automatic End-to-End Fact Checking System

null

We present FAKTA which is a unified framework that integrates various components of a fact-checking process: document retrieval from media sources with various types of reliability, stance detection of documents with respect to given claims, evidence extraction, and linguistic analysis. FAKTA predicts the factuality of given claims and provides evidence at the document and sentence level to explain its predictions.

null

acl

inproceedings

nadeem-etal-2019-fakta

Ammar, Waleed and Louis, Annie and Mostafazadeh, Nasrin

jun

2,019

Minneapolis, Minnesota

Association for Computational Linguistics

https://aclanthology.org/N19-4014/

Nadeem, Moin and Fang, Wei and Xu, Brian and Mohtarami, Mitra and Glass, James

Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics (Demonstrations)

78--83

null

{COMMIT} at {S}em{E}val-2017 Task 5: Ontology-based Method for Sentiment Analysis of Financial Headlines

null

This paper describes our submission to Task 5 of SemEval 2017, Fine-Grained Sentiment Analysis on Financial Microblogs and News, where we limit ourselves to performing sentiment analysis on news headlines only (track 2). The approach presented in this paper uses a Support Vector Machine to do the required regression, and besides unigrams and a sentiment tool, we use various ontology-based features. To this end we created a domain ontology that models various concepts from the financial domain. This allows us to model the sentiment of actions depending on which entity they are affecting (e.g., {\textquoteleft}decreasing debt' is positive, but {\textquoteleft}decreasing profit' is negative). The presented approach yielded a cosine distance of 0.6810 on the official test data, resulting in the 12th position.

null

acl

inproceedings

schouten-etal-2017-commit

Bethard, Steven and Carpuat, Marine and Apidianaki, Marianna and Mohammad, Saif M. and Cer, Daniel and Jurgens, David

aug

2,017

Vancouver, Canada

Association for Computational Linguistics

https://aclanthology.org/S17-2151/

Schouten, Kim and Frasincar, Flavius and de Jong, Franciska

Proceedings of the 11th International Workshop on Semantic Evaluation ({S}em{E}val-2017)

883--887

null

1501.07586

Christos Pappas Mr

Christos Pappas, Raphael M. Reischuk and Adrian Perrig

FAIR: Forwarding Accountability for Internet Reputability

16 pages, 12 figures

null

10.1109/ICNP.2015.22

null

cs.NI

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

This paper presents FAIR, a forwarding accountability mechanism that incentivizes ISPs to apply stricter security policies to their customers. The Autonomous System (AS) of the receiver specifies a traffic profile that the sender AS must adhere to. Transit ASes on the path mark packets. In case of traffic profile violations, the marked packets are used as a proof of misbehavior. FAIR introduces low bandwidth overhead and requires no per-packet and no per-flow state for forwarding. We describe integration with IP and demonstrate a software switch running on commodity hardware that can switch packets at a line rate of 120 Gbps, and can forward 140M minimum-sized packets per second, limited by the hardware I/O subsystem. Moreover, this paper proposes a "suspicious bit" for packet headers - an application that builds on top of FAIR's proofs of misbehavior and flags packets to warn other entities in the network.

[ { "created": "Thu, 29 Jan 2015 20:47:18 GMT", "version": "v1" }, { "created": "Mon, 2 Feb 2015 18:51:02 GMT", "version": "v2" }, { "created": "Tue, 3 Feb 2015 17:55:09 GMT", "version": "v3" }, { "created": "Mon, 10 Aug 2015 23:29:40 GMT", "version": "v4" }, { "created": "Fri, 6 May 2016 14:19:07 GMT", "version": "v5" } ]

2016-05-09

[ [ "Pappas", "Christos", "" ], [ "Reischuk", "Raphael M.", "" ], [ "Perrig", "Adrian", "" ] ]

Thu, 29 Jan 2015 20:47:18 GMT

2015-01-29T20:47:18Z

[ "cs" ]

arxiv

null

1712.07309

James Van Zandt

James R. Van Zandt

Efficient Cubature Rules

26 pages, 4 figures, 95 supplemental files. The 10 point rule for E_2^{r^2} was known, but had not been published. It was found by Jankewitz (private communication to R. Cools, 1998). Both the 15 and 16 point formulas of degree 4 for the sphere in 4 dimensions are new, and different from the 16 point formula found by Mysovskih

Electronic Transactions on Numerical Analysis, vol 51, pp 219-239, 2019

10.1553/etna_vol51s219

null

math.NA cs.NA

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

73 new cubature rules are found for three standard multidimensional integrals with spherically symmetric regions and weights, using direct search with a numerical zero-finder. All but four of the new rules have fewer integration points than known rules of the same degree, and twenty are within three points of M{\"o}ller's lower bound. Most have all positive coefficients and most have some symmetry, including some supported by one or two concentric spheres. They include degree 7 formulas for integration over the sphere and Gaussian-weighted integrals over all space, each in 6 and 7 dimensions, with 127 and 183 points, respectively.

[ { "created": "Wed, 20 Dec 2017 03:26:30 GMT", "version": "v1" }, { "created": "Tue, 3 Apr 2018 01:05:54 GMT", "version": "v2" }, { "created": "Sun, 16 Jun 2019 02:12:08 GMT", "version": "v3" } ]

2019-08-09

[ [ "Van Zandt", "James R.", "" ] ]

Wed, 20 Dec 2017 03:26:30 GMT

2017-12-20T03:26:30Z

[ "cs", "math" ]

arxiv

null

hep-th/0604203

Jason Kumar

Jason Kumar and James D. Wells

Multi-Brane Recombination and Standard Model Flux Vacua

6 pages, LaTeX, to be published (in shortened version) in the Proceedings of the Intl. Conf. on Supersymmetry, SUSY-06 (Irvine, 12-17 June 2006)

AIP Conf.Proc.903:509-512,2007

10.1063/1.2735235

null

hep-th hep-ph

null

In a previous work, we noted that vacua with higher flux could be obtained by recombination of a small set of hidden-sector branes, numbering up to the Kahler degrees of freedom left to be fixed in the problem. Here, we discuss the general construction of Type IIB Standard Model flux vacua in which multiple branes participate in brane recombination in the hidden sector. We present several models with 4, 5 and 10 branes in a stack. They are complete with Pati-Salam gauge group, Standard Model chiral matter, and a hidden sector that combines with nonzero flux to satisfy RR tadpole constraints. We also illustrate a puzzle: within the 4-brane recombination approach, we are unable to find a formal cutoff on the number of Standard Model flux vacua. Nevertheless, we show that phenomenological considerations are sure to cut off the number of vacua, albeit at an extraordinarily large value. We comment on the possibility that more formal constraints could further reduce the number of vacua.

[ { "created": "Thu, 27 Apr 2006 15:40:20 GMT", "version": "v1" }, { "created": "Thu, 11 May 2006 18:46:55 GMT", "version": "v2" }, { "created": "Tue, 7 Nov 2006 16:34:00 GMT", "version": "v3" } ]

2011-04-11

[ [ "Kumar", "Jason", "" ], [ "Wells", "James D.", "" ] ]

Thu, 27 Apr 2006 15:40:20 GMT

2006-04-27T15:40:20Z

[ "physics" ]

arxiv

null

Predatory journals: a major threat in orthopaedic research.

null

Predatory publishing is a major threat to contemporary publishing, as it offers 'to unaware scientist's', a quick open-access publication against fees without peer-review procedures.. Lack of peer-review leads to unethical practices, as plagiarism, publication of unscientific falsified data, and even unsafe clinical practices. As these journals threaten the credibility of academic publishing, significant work has been done from many scientific teams, in the last years, in establishing discriminating criteria between predatory and legitimate publishing. In the present review, we include mechanisms used by predatory editors to convince eager researchers to submit to their journals. We also provide useful links giving information about potential predatory journals and publishers, as well as scholarly writing. Joining the efforts of different scientific disciplines which compiled "green" lists with journals in their field, we conducted a "green" list with genuine orthopaedic research journals based on the directory of open-access journals (DOAJ) and Thomson Reuters journal citation reports. Ninety-six legitimate orthopaedic journals were identified based on the Thomson Reuters journal citation reports. One hundred thirty hits were found on the DOAJ site using the keywords "orthopaedics, orthopedics, sports medicine, musculoskeletal, trauma, traumatology, osteoarthritis, osteoporosis, cartilage, bone, hand, shoulder, knee, hip, foot, wound." Twenty-one journals on the DOAJ site occurred overlapping with keywords. Researchers and clinicians in the field of orthopaedics are advised to use all available tools in order to recognize predatory practices and avoid publishing in predatory journals.

null

pubmed

null

2,019

null

30,288,548

International orthopaedics

1102.4930

Jie Hou

Gerhard Kramer, Jie Hou

Short-Message Quantize-Forward Network Coding

null

8th International Workshop on Multi-Carrier Systems & Solutions (invited), Herrsching, Germany, May 2011

null

cs.IT math.IT

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

Recent work for single-relay channels shows that quantize-forward (QF) with long-message encoding achieves the same reliable rates as compress-forward (CF) with short-message encoding. It is shown that short-message QF with backward or pipelined (sliding-window) decoding also achieves the same rates. Similarly, for many relays and sources, short-message QF with backward decoding achieves the same rates as long-message QF. Several practical advantages of short-message encoding are pointed out, e.g., reduced delay and simpler modulation. Furthermore, short-message encoding lets relays use decode-forward (DF) if their channel quality is good, thereby enabling multiinput, multi-output (MIMO) gains that are not possible with long-message encoding. Finally, one may combine the advantages of long- and short-message encoding by hashing a long message to short messages.

[ { "created": "Thu, 24 Feb 2011 08:50:24 GMT", "version": "v1" }, { "created": "Thu, 5 May 2011 13:44:05 GMT", "version": "v2" } ]

2011-05-06

[ [ "Kramer", "Gerhard", "" ], [ "Hou", "Jie", "" ] ]

Thu, 24 Feb 2011 08:50:24 GMT

2011-02-24T08:50:24Z

[ "cs", "math" ]

arxiv

null

1610.09904

Charles-Albert Lehalle

Pierre Cardaliaguet (CEREMADE), Charles-Albert Lehalle

Mean Field Game of Controls and An Application To Trade Crowding

null

q-fin.TR math.OC

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

In this paper we formulate the now classical problem of optimal liquidation (or optimal trading) inside a Mean Field Game (MFG). This is a noticeable change since usually mathematical frameworks focus on one large trader in front of a "background noise" (or "mean field"). In standard frameworks, the interactions between the large trader and the price are a temporary and a permanent market impact terms, the latter influencing the public price. In this paper the trader faces the uncertainty of fair price changes too but not only. He has to deal with price changes generated by other similar market participants, impacting the prices permanently too, and acting strategically. Our MFG formulation of this problem belongs to the class of "extended MFG", we hence provide generic results to address these "MFG of controls", before solving the one generated by the cost function of optimal trading. We provide a closed form formula of its solution, and address the case of "heterogenous preferences" (when each participant has a different risk aversion). Last but not least we give conditions under which participants do not need to instantaneously know the state of the whole system, but can "learn" it day after day, observing others' behaviors.

[ { "created": "Mon, 31 Oct 2016 13:03:37 GMT", "version": "v1" }, { "created": "Fri, 11 Aug 2017 07:37:29 GMT", "version": "v2" }, { "created": "Thu, 21 Sep 2017 08:33:08 GMT", "version": "v3" } ]

2017-09-22

[ [ "Cardaliaguet", "Pierre", "", "CEREMADE" ], [ "Lehalle", "Charles-Albert", "" ] ]

Mon, 31 Oct 2016 13:03:37 GMT

2016-10-31T13:03:37Z

[ "math", "q-fin" ]

arxiv

null

Speculation and Negation detection in {F}rench biomedical corpora

null

In this work, we propose to address the detection of negation and speculation, and of their scope, in French biomedical documents. It has been indeed observed that they play an important role and provide crucial clues for other NLP applications. Our methods are based on CRFs and BiLSTM. We reach up to 97.21 {\%} and 91.30 {\%} F-measure for the detection of negation and speculation cues, respectively, using CRFs. For the computing of scope, we reach up to 90.81 {\%} and 86.73 {\%} F-measure on negation and speculation, respectively, using BiLSTM-CRF fed with word embeddings.

null

acl

inproceedings

dalloux-etal-2019-speculation

Mitkov, Ruslan and Angelova, Galia

sep

2,019

Varna, Bulgaria

INCOMA Ltd.

https://aclanthology.org/R19-1026/

Dalloux, Cl{\'e}ment and Claveau, Vincent and Grabar, Natalia

Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)

223--232

null

1110.2879

G\"uven\c{c} Arslan

A Characterization Based On Products of Order Statistics

8 pages

null

math.ST stat.TH

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

A new characterization for power function distributions is obtained which is based on products of order statistics. This result may be considered as a generalization of some recent results for contractions. We note that in this new result the product consists of order statistics from independent samples. These type of results are related to some scheme of ranked set sampling, for example.

[ { "created": "Thu, 13 Oct 2011 09:42:36 GMT", "version": "v1" }, { "created": "Mon, 15 Oct 2012 17:28:29 GMT", "version": "v2" }, { "created": "Sun, 28 Oct 2012 10:50:17 GMT", "version": "v3" } ]

2012-10-30

[ [ "Arslan", "Güvenç", "" ] ]

Thu, 13 Oct 2011 09:42:36 GMT

2011-10-13T09:42:36Z

[ "math", "stat" ]

arxiv

null

1609.08486

Yi-Jun Chang

Yi-Jun Chang, Tsvi Kopelowitz, Seth Pettie, Ruosong Wang, Wei Zhan

Exponential Separations in the Energy Complexity of Leader Election

null

cs.DC cs.CC cs.DS

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

Energy is often the most constrained resource for battery-powered wireless devices and the lion's share of energy is often spent on transceiver usage (sending/receiving packets), not on computation. In this paper we study the energy complexity of LeaderElection and ApproximateCounting in several models of wireless radio networks. It turns out that energy complexity is very sensitive to whether the devices can generate random bits and their ability to detect collisions. We consider four collision-detection models: Strong-CD (in which transmitters and listeners detect collisions), Sender-CD and Receiver-CD (in which only transmitters or only listeners detect collisions), and No-CD (in which no one detects collisions.) The take-away message of our results is quite surprising. For randomized LeaderElection algorithms, there is an exponential gap between the energy complexity of Sender-CD and Receiver-CD, and for deterministic LeaderElection algorithms there is another exponential gap, but in the reverse direction. In particular, the randomized energy complexity of LeaderElection is $\Theta(\log^* n)$ in Sender-CD but $\Theta(\log(\log^* n))$ in Receiver-CD, where $n$ is the (unknown) number of devices. Its deterministic complexity is $\Theta(\log N)$ in Receiver-CD but $\Theta(\log\log N)$ in Sender-CD, where $N$ is the (known) size of the devices' ID space. There is a tradeoff between time and energy. We give a new upper bound on the time-energy tradeoff curve for randomized LeaderElection and ApproximateCounting. A critical component of this algorithm is a new deterministic LeaderElection algorithm for dense instances, when $n=\Theta(N)$, with inverse-Ackermann-type ($O(\alpha(N))$) energy complexity.

[ { "created": "Tue, 27 Sep 2016 14:58:35 GMT", "version": "v1" }, { "created": "Thu, 3 Nov 2016 14:05:57 GMT", "version": "v2" }, { "created": "Wed, 5 Sep 2018 19:34:48 GMT", "version": "v3" } ]

2018-09-07

[ [ "Chang", "Yi-Jun", "" ], [ "Kopelowitz", "Tsvi", "" ], [ "Pettie", "Seth", "" ], [ "Wang", "Ruosong", "" ], [ "Zhan", "Wei", "" ] ]

Tue, 27 Sep 2016 14:58:35 GMT

2016-09-27T14:58:35Z

[ "cs" ]

arxiv

null

1601.03120

Sanzhang Xu

Xiaoxiang Zhang, Sanzhang Xu, Jianlong Chen

Core partial order in rings with involution

null

math.RA

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

Let $R$ be a unital ring with involution. We give several characterizations and properties of core partial order in $R$. In particular, we investigate the reverse order law $(ab)^{\tiny\textcircled{\tiny\#}} = b^{\tiny\textcircled{\tiny\#}} a^{\tiny\textcircled{\tiny\#}}$ for two core invertible elements $a,b\in R$. Some relationships between core partial order and other partial orders are obtained.

[ { "created": "Wed, 13 Jan 2016 03:19:03 GMT", "version": "v1" }, { "created": "Thu, 28 Jan 2016 08:14:33 GMT", "version": "v2" }, { "created": "Thu, 25 May 2017 14:56:06 GMT", "version": "v3" } ]

2017-05-26

[ [ "Zhang", "Xiaoxiang", "" ], [ "Xu", "Sanzhang", "" ], [ "Chen", "Jianlong", "" ] ]

Wed, 13 Jan 2016 03:19:03 GMT

2016-01-13T03:19:03Z

[ "math" ]

arxiv

null

Lateral and syndesmotic ankle sprain injuries: a narrative literature review.

null

The purpose of this article is to review the literature that discusses normal anatomy and biomechanics of the foot and ankle, mechanisms that may result in a lateral ankle sprain or syndesmotic sprain, and assessment and diagnostic procedures, and to present a treatment algorithm based on normal ligament healing principles.

null

pubmed

null

2,011

null

22,014,912

Journal of chiropractic medicine

1809.04202

Manik Banik

Sristy Agrawal, Saronath Halder, and Manik Banik

Genuinely entangled subspace with all-encompassing distillable entanglement across every bipartition

Close to published version

Phys. Rev. A 99, 032335 (2019)

10.1103/PhysRevA.99.032335

null

quant-ph math-ph math.MP

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

In a multipartite scenario quantum entanglement manifests its most dramatic form when the state is genuinely entangled. Such a state is more beneficial for information theoretic applications if it contains distillable entanglement in every bipartition. It is, therefore, of significant operational interest to identify subspaces of multipartite quantum systems that contain such properties apriori. In this letter, we introduce the notion of unextendible biseparable bases (UBB) that provides an adequate method to construct genuinely entangled subspaces (GES). We provide an explicit construction of two types of UBBs -- party symmetric and party asymmetric -- for every $3$-{\it qudit} quantum system, with local dimension d\ge 3. Further, we show that the GES resulting from the symmetric construction is indeed a {\it bidistillable} subspace, i.e., all the states supported on it contain distillable entanglement across every bipartition.

[ { "created": "Wed, 12 Sep 2018 00:06:50 GMT", "version": "v1" }, { "created": "Tue, 26 Mar 2019 03:58:38 GMT", "version": "v2" } ]

2019-03-27

[ [ "Agrawal", "Sristy", "" ], [ "Halder", "Saronath", "" ], [ "Banik", "Manik", "" ] ]

Wed, 12 Sep 2018 00:06:50 GMT

2018-09-12T00:06:50Z

[ "physics", "math" ]

arxiv

null

1506.04911

Mauro Mobilia

Nonlinear $q$-voter model with inflexible zealots

11 pages, 10 figures. For Physical Review E (accepted, final version)

Phys. Rev. E 92, 012803 (2015)

10.1103/PhysRevE.92.012803

null

physics.soc-ph cond-mat.stat-mech cs.SI nlin.AO q-bio.PE

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

We study the dynamics of the nonlinear $q$-voter model with inflexible zealots in a finite well-mixed population. In this system, each individual supports one of two parties and is either a susceptible voter or an inflexible zealot. At each time step, a susceptible adopts the opinion of a neighbor if this belongs to a group of $q\geq 2$ neighbors all in the same state, whereas inflexible zealots never change their opinion. In the presence of zealots of both parties the model is characterized by a fluctuating stationary state and, below a zealotry density threshold, the distribution of opinions is bimodal. After a characteristic time, most susceptibles become supporters of the party having more zealots and the opinion distribution is asymmetric. When the number of zealots of both parties is the same, the opinion distribution is symmetric and, in the long run, susceptibles endlessly swing from the state where they all support one party to the opposite state. Above the zealotry density threshold, when there is an unequal number of zealots of each type, the probability distribution is single-peaked and non-Gaussian. These properties are investigated analytically and with stochastic simulations. We also study the mean time to reach a consensus when zealots support only one party.

[ { "created": "Tue, 16 Jun 2015 10:55:33 GMT", "version": "v1" }, { "created": "Thu, 25 Jun 2015 16:10:21 GMT", "version": "v2" } ]

2015-07-09

[ [ "Mobilia", "Mauro", "" ] ]

Tue, 16 Jun 2015 10:55:33 GMT

2015-06-16T10:55:33Z

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

arxiv

null

0904.3756

Michael Goodrich

Wenliang Du, David Eppstein, Michael T. Goodrich, George S. Lueker

On the Approximability of Geometric and Geographic Generalization and the Min-Max Bin Covering Problem

18 pages. Expanded version of paper appearing in 2009 Algorithms and Data Structures Symposium (formerly WADS)

null

cs.DS

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

We study the problem of abstracting a table of data about individuals so that no selection query can identify fewer than k individuals. We show that it is impossible to achieve arbitrarily good polynomial-time approximations for a number of natural variations of the generalization technique, unless P = NP, even when the table has only a single quasi-identifying attribute that represents a geographic or unordered attribute: Zip-codes: nodes of a planar graph generalized into connected subgraphs GPS coordinates: points in R2 generalized into non-overlapping rectangles Unordered data: text labels that can be grouped arbitrarily. In addition to impossibility results, we provide approximation algorithms for these difficult single-attribute generalization problems, which, of course, apply to multiple-attribute instances with one that is quasi-identifying. We show theoretically and experimentally that our approximation algorithms can come reasonably close to optimal solutions. Incidentally, the generalization problem for unordered data can be viewed as a novel type of bin packing problem--min-max bin covering--which may be of independent interest.

[ { "created": "Thu, 23 Apr 2009 21:06:58 GMT", "version": "v1" }, { "created": "Mon, 27 Apr 2009 22:59:44 GMT", "version": "v2" }, { "created": "Tue, 12 May 2009 16:16:10 GMT", "version": "v3" } ]

2009-05-12

[ [ "Du", "Wenliang", "" ], [ "Eppstein", "David", "" ], [ "Goodrich", "Michael T.", "" ], [ "Lueker", "George S.", "" ] ]

Thu, 23 Apr 2009 21:06:58 GMT

2009-04-23T21:06:58Z

[ "cs" ]

arxiv

null

2203.06250

Vittorio Giammarino

Vittorio Giammarino, Matthew F Dunne, Kylie N Moore, Michael E Hasselmo, Chantal E Stern, Ioannis Ch. Paschalidis

Combining imitation and deep reinforcement learning to accomplish human-level performance on a virtual foraging task

24 pages, 15 figures

null

cs.LG cs.SY eess.SY

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

We develop a simple framework to learn bio-inspired foraging policies using human data. We conduct an experiment where humans are virtually immersed in an open field foraging environment and are trained to collect the highest amount of rewards. A Markov Decision Process (MDP) framework is introduced to model the human decision dynamics. Then, Imitation Learning (IL) based on maximum likelihood estimation is used to train Neural Networks (NN) that map human decisions to observed states. The results show that passive imitation substantially underperforms humans. We further refine the human-inspired policies via Reinforcement Learning (RL) using the on-policy Proximal Policy Optimization (PPO) algorithm which shows better stability than other algorithms and can steadily improve the policies pretrained with IL. We show that the combination of IL and RL can match human results and that good performance strongly depends on combining the allocentric information with an egocentric representation of the environment.

[ { "created": "Fri, 11 Mar 2022 20:52:30 GMT", "version": "v1" }, { "created": "Wed, 16 Mar 2022 18:34:58 GMT", "version": "v2" }, { "created": "Wed, 30 Mar 2022 19:51:54 GMT", "version": "v3" }, { "created": "Tue, 6 Dec 2022 21:36:23 GMT", "version": "v4" } ]

2022-12-08

[ [ "Giammarino", "Vittorio", "" ], [ "Dunne", "Matthew F", "" ], [ "Moore", "Kylie N", "" ], [ "Hasselmo", "Michael E", "" ], [ "Stern", "Chantal E", "" ], [ "Paschalidis", "Ioannis Ch.", "" ] ]

Fri, 11 Mar 2022 20:52:30 GMT

2022-03-11T20:52:30Z

[ "cs", "eess" ]

arxiv

null

Dietary intakes and risk of lymphoid and myeloid leukemia in the European Prospective Investigation into Cancer and Nutrition (EPIC).

null

The etiology of leukemias cannot entirely be explained by known risk factors, including ionizing radiation, benzene exposure, and infection with human T cell leukemia virus. A number of studies suggested that diet influences the risk of adult leukemias. However, results have been largely inconsistent. We examined the potential association between dietary factors and risk of leukemias among participants of the European Prospective Investigation into Cancer and Nutrition study. Among the 477,325 participants with mean follow-up of 11.34 yr (SD = 2.47), 773 leukemias (373 and 342 cases of lymphoid and myeloid leukemia, respectively) were identified. Diet over the previous 12 mo was assessed at baseline using a validated country-specific dietary questionnaire. Cox proportional hazards regression was used to explore the association between dietary factors that have previously been associated with leukemia risk, including red and processed meat, poultry, offal, fish, dairy products, vegetables, fruits, and seeds/nuts, and risk of both lymphoid and myeloid leukemias. No significant associations were observed between dietary measures and total, lymphoid, and myeloid leukemias. Additional subtype analyses showed no dietary association with risk of major subtypes of leukemias. In summary, this study did not support a possible link between selected dietary factors and risk of leukemias.

null

pubmed

null

2,014

null

24,279,598

Nutrition and cancer

1503.03888

Andrey Nikolaev

Jeremy Macdonald, Alexei Myasnikov, Andrey Nikolaev, Svetla Vassileva

Logspace and compressed-word computations in nilpotent groups

Accepted in Transactions of the AMS

null

math.GR cs.CC

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

For finitely generated nilpotent groups, we employ Mal'cev coordinates to solve several classical algorithmic problems efficiently. Computation of normal forms, the membership problem, the conjugacy problem, and computation of presentations for subgroups are solved using only logarithmic space and quasilinear time. Logarithmic space presentation-uniform versions of these algorithms are provided. Compressed-word versions of the same problems, in which each input word is provided as a straight-line program, are solved in polynomial time.

[ { "created": "Thu, 12 Mar 2015 20:59:43 GMT", "version": "v1" }, { "created": "Sun, 10 May 2015 20:41:55 GMT", "version": "v2" }, { "created": "Sun, 19 Dec 2021 19:25:20 GMT", "version": "v3" } ]

2021-12-21

[ [ "Macdonald", "Jeremy", "" ], [ "Myasnikov", "Alexei", "" ], [ "Nikolaev", "Andrey", "" ], [ "Vassileva", "Svetla", "" ] ]

Thu, 12 Mar 2015 20:59:43 GMT

2015-03-12T20:59:43Z

[ "cs", "math" ]

arxiv

null

1207.5237

Alex Bernardini Dr.

Alex E. Bernardini and Roldao da Rocha

Cyclically deformed defects and topological mass constraints

36 pages, 9 figures

Advances in High Energy Physics, vol.2013, Article ID 304980, 18 pages (2013)

10.1155/2013/304980

null

hep-th cond-mat.stat-mech math-ph math.MP nlin.PS

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

A systematic procedure for obtaining defect structures through cyclic deformation chains is introduced and explored in detail. The procedure outlines a set of rules for analytically constructing constraint equations that involve the finite localized energy of cyclically generated defects. The idea of obtaining cyclically deformed defects concerns the possibility of regenerating a primitive (departing) defect structure through successive, unidirectional, and eventually irreversible, deformation processes. Our technique is applied on kink-like and lump-like solutions in models described by a single real scalar field, such that extensions to quantum mechanics follow the usual theory of deformed defects. The preliminary results show that the cyclic device supports simultaneously kink-like and lump-like defects into 3- and 4-cyclic deformation chains with topological mass values closed by trigonometric and hyperbolic deformations. In a straightforward generalization, results concerning with the analytical calculation of $N$-cyclic deformations are obtained and lessons regarding extensions from more elaborated primitive defects are depicted.

[ { "created": "Sun, 22 Jul 2012 16:16:06 GMT", "version": "v1" }, { "created": "Mon, 5 Nov 2012 21:26:30 GMT", "version": "v2" }, { "created": "Wed, 27 Feb 2013 18:30:14 GMT", "version": "v3" } ]

2013-04-09

[ [ "Bernardini", "Alex E.", "" ], [ "da Rocha", "Roldao", "" ] ]

Sun, 22 Jul 2012 16:16:06 GMT

2012-07-22T16:16:06Z

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

arxiv

null

2210.09211

Miguel Garcia-Ortegon

Miguel Garcia-Ortegon, Andreas Bender and Sergio Bacallado

Conditional Neural Processes for Molecules

null

stat.ML cs.LG

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

Neural processes (NPs) are models for transfer learning with properties reminiscent of Gaussian Processes (GPs). They are adept at modelling data consisting of few observations of many related functions on the same input space and are trained by minimizing a variational objective, which is computationally much less expensive than the Bayesian updating required by GPs. So far, most studies of NPs have focused on low-dimensional datasets which are not representative of realistic transfer learning tasks. Drug discovery is one application area that is characterized by datasets consisting of many chemical properties or functions which are sparsely observed, yet depend on shared features or representations of the molecular inputs. This paper applies the conditional neural process (CNP) to DOCKSTRING, a dataset of docking scores for benchmarking ML models. CNPs show competitive performance in few-shot learning tasks relative to supervised learning baselines common in chemoinformatics, as well as an alternative model for transfer learning based on pre-training and refining neural network regressors. We present a Bayesian optimization experiment which showcases the probabilistic nature of CNPs and discuss shortcomings of the model in uncertainty quantification.

[ { "created": "Mon, 17 Oct 2022 16:10:12 GMT", "version": "v1" }, { "created": "Thu, 1 Dec 2022 17:20:28 GMT", "version": "v2" }, { "created": "Thu, 23 Feb 2023 15:48:02 GMT", "version": "v3" } ]

2023-02-24

[ [ "Garcia-Ortegon", "Miguel", "" ], [ "Bender", "Andreas", "" ], [ "Bacallado", "Sergio", "" ] ]

Mon, 17 Oct 2022 16:10:12 GMT

2022-10-17T16:10:12Z

[ "cs", "stat" ]

arxiv

null

1605.05432

Zachary McGuirk

Sajjad Lakzian and Zachary McGuirk

Global Poincar\'e inequality on Graphs via Conical Curvature-Dimension Conditions

15 pages, minor corrections in v2 and generalized harmonic functions are getting their own manuscript so they have been removed from this version

Analysis and Geometry in Metric Spaces 6, no. 1, 32 - 47, first online (2018)

null

math.DG

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

We introduce and study the conical curvature-dimension condition, $CCD(K,N)$, for graphs. We show that $CCD(K,N)$ provides necessary and sufficient conditions for the underlying graph to satisfy a sharp global Poincar\'e inequality which in turn translates to a sharp lower bound for the first eigenvalues of these graphs. Another application of the conical curvature-dimension analysis is finding a sharp estimate on the curvature of complete graphs.

[ { "created": "Wed, 18 May 2016 04:05:35 GMT", "version": "v1" }, { "created": "Tue, 12 Sep 2017 20:11:41 GMT", "version": "v2" } ]

2018-07-26

[ [ "Lakzian", "Sajjad", "" ], [ "McGuirk", "Zachary", "" ] ]

Wed, 18 May 2016 04:05:35 GMT

2016-05-18T04:05:35Z

[ "math" ]

arxiv

null

1603.00946

Goran Radunovi\'c

Michel L. Lapidus (1), Goran Radunovi\'c (2), Darko \v{Z}ubrini\'c (2) ((1) University of California, Riverside, (2) University of Zagreb)

Zeta Functions and Complex Dimensions of Relative Fractal Drums: Theory, Examples and Applications

101 pages, corrected typos

Dissertationes mathematicae, 526 (2017) 1-105

10.4064/dm757-4-2017

null

math-ph math.CV math.MG math.MP

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

In 2009, the first author introduced a new class of zeta functions, called `distance zeta functions', associated with arbitrary compact fractal subsets of Euclidean spaces of arbitrary dimension. It represents a natural, but nontrivial extension of the theory of `geometric zeta functions' of bounded fractal strings. In this memoir, we introduce the class of `relative fractal drums' (or RFDs), which contains the classes of bounded fractal strings and of compact fractal subsets of Euclidean spaces as special cases. Furthermore, the associated (relative) distance zeta functions of RFDs, extend (in a suitable sense) the aforementioned classes of fractal zeta functions. This notion is very general and flexible, enabling us to view practically all of the previously studied aspects of the theory of fractal zeta functions from a unified perspective as well as to go well beyond the previous theory. The abscissa of (absolute) convergence of any relative fractal drum is equal to the relative box dimension of the RFD. We pay particular attention to the question of constructing meromorphic extensions of the distance zeta functions of RFDs, as well as to the construction of transcendentally $\infty$-quasiperiodic RFDs (i.e., roughly, RFDs with infinitely many quasiperiods, all of which are algebraically independent). We also describe a class of RFDs (and, in particular, a new class of bounded sets), called {\em maximal hyperfractals}, such that the critical line of (absolute) convergence consists solely of nonremovable singularities of the associated relative distance zeta functions. Finally, we also describe a class of Minkowski measurable RFDs which possess an infinite sequence of complex dimensions of arbitrary multiplicity $m\ge1$, and even an infinite sequence of essential singularities along the critical line.

[ { "created": "Thu, 3 Mar 2016 02:15:51 GMT", "version": "v1" }, { "created": "Thu, 10 Mar 2016 08:02:42 GMT", "version": "v2" }, { "created": "Tue, 23 Aug 2016 08:57:02 GMT", "version": "v3" } ]

2018-03-21

[ [ "Lapidus", "Michel L.", "", "University of California, Riverside" ], [ "Radunović", "Goran", "", "University of Zagreb" ], [ "Žubrinić", "Darko", "", "University of Zagreb" ] ]

Thu, 3 Mar 2016 02:15:51 GMT

2016-03-03T02:15:51Z

[ "math" ]

arxiv

null

hep-ph/0003296

Jose Antonio Martins Simoes

Y. A. Coutinho, J. A. Martins Simoes, C. M. Porto

Fermion masses in a model for spontaneous parity breaking

7 pages, 3 figures. Accepted in Eur. Phys. J. C

Eur.Phys.J.C18:779-784,2001

10.1007/s100520100525

null

hep-ph

null

In this paper we discuss a left-right symmetric model for elementary particles and their connection with the mass spectrum of elementary fermions. The model is based on the group $SU(2)_L\otimes SU(2)_R\otimes U(1)$. New mirror fermions and a minimal set of Higgs particles that breaks the symmetry down to $U(1)_{em}$ are proposed. The model can accommodate a consistent pattern for charged and neutral fermion masses as well as neutrino oscillations. An important consequence of the model is that the connection between the left and right sectors can be done by the neutral vector gauge bosons Z and a new heavy Z'.

[ { "created": "Thu, 30 Mar 2000 13:46:01 GMT", "version": "v1" }, { "created": "Wed, 13 Sep 2000 19:04:54 GMT", "version": "v2" }, { "created": "Tue, 9 Jan 2001 15:49:20 GMT", "version": "v3" } ]

2009-01-07

[ [ "Coutinho", "Y. A.", "" ], [ "Simoes", "J. A. Martins", "" ], [ "Porto", "C. M.", "" ] ]

Thu, 30 Mar 2000 13:46:01 GMT

2000-03-30T13:46:01Z

[ "physics" ]

arxiv

null

Obstacles to skin self-examination: are frontier adults inclined abstainers?

null

Individuals residing in U.S. frontier counties have limited access to dermatology care and higher melanoma mortality rates. Given these limitations, frontier residents rely disproportionately on skin self-examinations (SSE) for early detection of melanoma, though little is known about their SSE behaviors and barriers to conducting SSEs. The goal of the present study was to identify obstacles to SSE performance via a survey of adults (

null

pubmed

null

2,020

null

31,847,571

Psychology, health & medicine

0708.1756

Alfonsi Aurelien

Aur\'elien Alfonsi (CERMICS), Antje Fruth, Alexander Schied

Optimal execution strategies in limit order books with general shape functions

null

q-fin.TR math.PR

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

We consider optimal execution strategies for block market orders placed in a limit order book (LOB). We build on the resilience model proposed by Obizhaeva and Wang (2005) but allow for a general shape of the LOB defined via a given density function. Thus, we can allow for empirically observed LOB shapes and obtain a nonlinear price impact of market orders. We distinguish two possibilities for modeling the resilience of the LOB after a large market order: the exponential recovery of the number of limit orders, i.e., of the volume of the LOB, or the exponential recovery of the bid-ask spread. We consider both of these resilience modes and, in each case, derive explicit optimal execution strategies in discrete time. Applying our results to a block-shaped LOB, we obtain a new closed-form representation for the optimal strategy, which explicitly solves the recursive scheme given in Obizhaeva and Wang (2005). We also provide some evidence for the robustness of optimal strategies with respect to the choice of the shape function and the resilience-type.

[ { "created": "Mon, 13 Aug 2007 19:11:46 GMT", "version": "v1" }, { "created": "Tue, 25 Sep 2007 13:43:30 GMT", "version": "v2" }, { "created": "Wed, 3 Feb 2010 15:18:25 GMT", "version": "v3" } ]

2010-02-03

[ [ "Alfonsi", "Aurélien", "", "CERMICS" ], [ "Fruth", "Antje", "" ], [ "Schied", "Alexander", "" ] ]

Mon, 13 Aug 2007 19:11:46 GMT

2007-08-13T19:11:46Z

[ "math", "q-fin" ]

arxiv

null

1712.09644

William Mayner

William G. P. Mayner, William Marshall, Larissa Albantakis, Graham Findlay, Robert Marchman, Giulio Tononi

PyPhi: A toolbox for integrated information theory

22 pages, 4 figures, 6 pages of appendices. Supporting information "S1 Calculating Phi" can be found in the ancillary files

PLOS Computational Biology 14(7): e1006343. 2018

10.1371/journal.pcbi.1006343

null

q-bio.NC cs.AI q-bio.QM

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

Integrated information theory provides a mathematical framework to fully characterize the cause-effect structure of a physical system. Here, we introduce PyPhi, a Python software package that implements this framework for causal analysis and unfolds the full cause-effect structure of discrete dynamical systems of binary elements. The software allows users to easily study these structures, serves as an up-to-date reference implementation of the formalisms of integrated information theory, and has been applied in research on complexity, emergence, and certain biological questions. We first provide an overview of the main algorithm and demonstrate PyPhi's functionality in the course of analyzing an example system, and then describe details of the algorithm's design and implementation. PyPhi can be installed with Python's package manager via the command 'pip install pyphi' on Linux and macOS systems equipped with Python 3.4 or higher. PyPhi is open-source and licensed under the GPLv3; the source code is hosted on GitHub at https://github.com/wmayner/pyphi . Comprehensive and continually-updated documentation is available at https://pyphi.readthedocs.io/ . The pyphi-users mailing list can be joined at https://groups.google.com/forum/#!forum/pyphi-users . A web-based graphical interface to the software is available at http://integratedinformationtheory.org/calculate.html .

[ { "created": "Wed, 27 Dec 2017 18:01:12 GMT", "version": "v1" }, { "created": "Fri, 29 Dec 2017 17:45:30 GMT", "version": "v2" }, { "created": "Wed, 27 Jun 2018 09:52:27 GMT", "version": "v3" } ]

2018-08-16

[ [ "Mayner", "William G. P.", "" ], [ "Marshall", "William", "" ], [ "Albantakis", "Larissa", "" ], [ "Findlay", "Graham", "" ], [ "Marchman", "Robert", "" ], [ "Tononi", "Giulio", "" ] ]

Wed, 27 Dec 2017 18:01:12 GMT

2017-12-27T18:01:12Z

[ "q-bio", "cs" ]

arxiv

null

2202.12275

Matthew Ashman

Matthew Ashman, Thang D. Bui, Cuong V. Nguyen, Stratis Markou, Adrian Weller, Siddharth Swaroop and Richard E. Turner

Partitioned Variational Inference: A Framework for Probabilistic Federated Learning

arXiv admin note: substantial text overlap with arXiv:1811.11206

null

stat.ML cs.LG

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

The proliferation of computing devices has brought about an opportunity to deploy machine learning models on new problem domains using previously inaccessible data. Traditional algorithms for training such models often require data to be stored on a single machine with compute performed by a single node, making them unsuitable for decentralised training on multiple devices. This deficiency has motivated the development of federated learning algorithms, which allow multiple data owners to train collaboratively and use a shared model whilst keeping local data private. However, many of these algorithms focus on obtaining point estimates of model parameters, rather than probabilistic estimates capable of capturing model uncertainty, which is essential in many applications. Variational inference (VI) has become the method of choice for fitting many modern probabilistic models. In this paper we introduce partitioned variational inference (PVI), a general framework for performing VI in the federated setting. We develop new supporting theory for PVI, demonstrating a number of properties that make it an attractive choice for practitioners; use PVI to unify a wealth of fragmented, yet related literature; and provide empirical results that showcase the effectiveness of PVI in a variety of federated settings.

[ { "created": "Thu, 24 Feb 2022 18:15:30 GMT", "version": "v1" }, { "created": "Fri, 25 Feb 2022 12:04:46 GMT", "version": "v2" }, { "created": "Mon, 28 Feb 2022 23:40:04 GMT", "version": "v3" }, { "created": "Thu, 28 Apr 2022 14:01:57 GMT", "version": "v4" } ]

2022-04-29

[ [ "Ashman", "Matthew", "" ], [ "Bui", "Thang D.", "" ], [ "Nguyen", "Cuong V.", "" ], [ "Markou", "Stratis", "" ], [ "Weller", "Adrian", "" ], [ "Swaroop", "Siddharth", "" ], [ "Turner", "Richard E.", "" ] ]

Thu, 24 Feb 2022 18:15:30 GMT

2022-02-24T18:15:30Z

[ "cs", "stat" ]

arxiv

null

2010.04244

Jinglin Chen

Huozhi Zhou, Jinglin Chen, Lav R. Varshney, and Ashish Jagmohan

Nonstationary Reinforcement Learning with Linear Function Approximation

null

cs.LG stat.ML

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

We consider reinforcement learning (RL) in episodic Markov decision processes (MDPs) with linear function approximation under drifting environment. Specifically, both the reward and state transition functions can evolve over time but their total variations do not exceed a $\textit{variation budget}$. We first develop $\texttt{LSVI-UCB-Restart}$ algorithm, an optimistic modification of least-squares value iteration with periodic restart, and bound its dynamic regret when variation budgets are known. Then we propose a parameter-free algorithm $\texttt{Ada-LSVI-UCB-Restart}$ that extends to unknown variation budgets. We also derive the first minimax dynamic regret lower bound for nonstationary linear MDPs and as a byproduct establish a minimax regret lower bound for linear MDPs unsolved by Jin et al. (2020). Finally, we provide numerical experiments to demonstrate the effectiveness of our proposed algorithms.

[ { "created": "Thu, 8 Oct 2020 20:07:44 GMT", "version": "v1" }, { "created": "Thu, 15 Oct 2020 04:56:38 GMT", "version": "v2" }, { "created": "Sat, 13 Apr 2024 06:52:10 GMT", "version": "v3" } ]

2024-04-16

[ [ "Zhou", "Huozhi", "" ], [ "Chen", "Jinglin", "" ], [ "Varshney", "Lav R.", "" ], [ "Jagmohan", "Ashish", "" ] ]

Thu, 8 Oct 2020 20:07:44 GMT

2020-10-08T20:07:44Z

[ "cs", "stat" ]

arxiv

null

Combining Translation Memory with Neural Machine Translation

null

In this paper, we report our submission systems (geoduck) to the Timely Disclosure task on the 6th Workshop on Asian Translation (WAT) (Nakazawa et al., 2019). Our system employs a combined approach of translation memory and Neural Machine Translation (NMT) models, where we can select final translation outputs from either a translation memory or an NMT system, when the similarity score of a test source sentence exceeds the predefined threshold. We observed that this combination approach significantly improves the translation performance on the Timely Disclosure corpus, as compared to a standalone NMT system. We also conducted source-based direct assessment on the final output, and we discuss the comparison between human references and each system`s output.

null

acl

inproceedings

eriguchi-etal-2019-combining

Nakazawa, Toshiaki and Ding, Chenchen and Dabre, Raj and Kunchukuttan, Anoop and Doi, Nobushige and Oda, Yusuke and Bojar, Ond{\v{r}}ej and Parida, Shantipriya and Goto, Isao and Mino, Hidaya

nov

2,019

Hong Kong, China

Association for Computational Linguistics

https://aclanthology.org/D19-5214/

Eriguchi, Akiko and Rarrick, Spencer and Matsushita, Hitokazu

Proceedings of the 6th Workshop on Asian Translation

123--130

null

{N}arra{S}um: A Large-Scale Dataset for Abstractive Narrative Summarization

null

Narrative summarization aims to produce a distilled version of a narrative to describe its most salient events and characters. Writing a summary for a narrative is challenging as it requires an understanding of event causality and character behaviors. To encourage research in this direction, we propose NarraSum, a large-scale narrative summarization dataset. It contains 122K narratives, which are collected from the synopses of movies and TV episodes with diverse genres, and their corresponding abstractive summaries. Experiments show that there is a large performance gap between humans and the state-of-the-art summarization models on NarraSum. We hope that this dataset will promote future research in summarization, as well as broader studies of natural language understanding and generation. The dataset is available at https://github.com/zhaochaocs/narrasum.

null

acl

inproceedings

zhao-etal-2022-narrasum

Goldberg, Yoav and Kozareva, Zornitsa and Zhang, Yue

dec

2,022

Abu Dhabi, United Arab Emirates

Association for Computational Linguistics

https://aclanthology.org/2022.findings-emnlp.14/

Zhao, Chao and Brahman, Faeze and Song, Kaiqiang and Yao, Wenlin and Yu, Dian and Chaturvedi, Snigdha

Findings of the Association for Computational Linguistics: EMNLP 2022

182--197

null

1905.13211

Keyulu Xu

Keyulu Xu, Jingling Li, Mozhi Zhang, Simon S. Du, Ken-ichi Kawarabayashi, Stefanie Jegelka

What Can Neural Networks Reason About?

null

cs.LG cs.AI cs.CV cs.NE stat.ML

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

Neural networks have succeeded in many reasoning tasks. Empirically, these tasks require specialized network structures, e.g., Graph Neural Networks (GNNs) perform well on many such tasks, but less structured networks fail. Theoretically, there is limited understanding of why and when a network structure generalizes better than others, although they have equal expressive power. In this paper, we develop a framework to characterize which reasoning tasks a network can learn well, by studying how well its computation structure aligns with the algorithmic structure of the relevant reasoning process. We formally define this algorithmic alignment and derive a sample complexity bound that decreases with better alignment. This framework offers an explanation for the empirical success of popular reasoning models, and suggests their limitations. As an example, we unify seemingly different reasoning tasks, such as intuitive physics, visual question answering, and shortest paths, via the lens of a powerful algorithmic paradigm, dynamic programming (DP). We show that GNNs align with DP and thus are expected to solve these tasks. On several reasoning tasks, our theory is supported by empirical results.

[ { "created": "Thu, 30 May 2019 17:53:30 GMT", "version": "v1" }, { "created": "Fri, 31 May 2019 21:50:31 GMT", "version": "v2" }, { "created": "Sun, 29 Sep 2019 20:42:29 GMT", "version": "v3" }, { "created": "Sat, 15 Feb 2020 06:56:25 GMT", "version": "v4" } ]

2020-02-18

[ [ "Xu", "Keyulu", "" ], [ "Li", "Jingling", "" ], [ "Zhang", "Mozhi", "" ], [ "Du", "Simon S.", "" ], [ "Kawarabayashi", "Ken-ichi", "" ], [ "Jegelka", "Stefanie", "" ] ]

Thu, 30 May 2019 17:53:30 GMT

2019-05-30T17:53:30Z

[ "cs", "stat" ]

arxiv

null

2201.09095

Paul Van den Hof

H. J. Dreef, S. Shi, X. Cheng, M. C. F. Donkers and P. M. J. Van den Hof

Excitation allocation for generic identifiability of linear dynamic networks with fixed modules

null

IEEE Control Systems Letters, Vol. 6, pp. 2587-2592, 2022

10.1109/LCSYS.2022.3171172

null

eess.SY cs.SY

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

Identifiability of linear dynamic networks requires the presence of a sufficient number of external excitation signals. The problem of allocating a minimal number of external signals for guaranteeing generic network identifiability has been recently addressed in the literature. Here we will extend that work by explicitly incorporating the situation that some network modules are known, and thus are fixed in the parametrized model set. The graphical approach introduced earlier is extended to this situation, showing that the presence of fixed modules reduces the required number of external signals. An algorithm is presented that allocates the external signals in a systematic fashion.

[ { "created": "Sat, 22 Jan 2022 16:40:31 GMT", "version": "v1" }, { "created": "Fri, 13 May 2022 09:31:54 GMT", "version": "v2" } ]

2022-05-16

[ [ "Dreef", "H. J.", "" ], [ "Shi", "S.", "" ], [ "Cheng", "X.", "" ], [ "Donkers", "M. C. F.", "" ], [ "Hof", "P. M. J. Van den", "" ] ]

Sat, 22 Jan 2022 16:40:31 GMT

2022-01-22T16:40:31Z

[ "cs", "eess" ]

arxiv

null

Mitochondrion.

null

Diverse biological data may be used to create illustrations of molecules in their cellular context. I describe the scientific results that support a recent textbook illustration of a mitochondrion. The image magnifies a portion of the mitochondrion by one million times, showing the location and form of membranes and individual macromolecules, revealing the molecular basis of its role in energy metabolism and apoptosis. Results from biochemistry, electron microscopy, and X-ray crystallography were used to create the image.

null

pubmed

null

2,010

null

21,567,815

Biochemistry and molecular biology education : a bimonthly publication of the International Union of Biochemistry and Molecular Biology

1802.03916

Zachary Lipton

Zachary C. Lipton, Yu-Xiang Wang, Alex Smola

Detecting and Correcting for Label Shift with Black Box Predictors

Published at the International Conference on Machine Learning (ICML) 2018

null

cs.LG cs.AI cs.NE stat.ML

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

Faced with distribution shift between training and test set, we wish to detect and quantify the shift, and to correct our classifiers without test set labels. Motivated by medical diagnosis, where diseases (targets) cause symptoms (observations), we focus on label shift, where the label marginal $p(y)$ changes but the conditional $p(x| y)$ does not. We propose Black Box Shift Estimation (BBSE) to estimate the test distribution $p(y)$. BBSE exploits arbitrary black box predictors to reduce dimensionality prior to shift correction. While better predictors give tighter estimates, BBSE works even when predictors are biased, inaccurate, or uncalibrated, so long as their confusion matrices are invertible. We prove BBSE's consistency, bound its error, and introduce a statistical test that uses BBSE to detect shift. We also leverage BBSE to correct classifiers. Experiments demonstrate accurate estimates and improved prediction, even on high-dimensional datasets of natural images.

[ { "created": "Mon, 12 Feb 2018 07:16:03 GMT", "version": "v1" }, { "created": "Wed, 28 Feb 2018 17:23:51 GMT", "version": "v2" }, { "created": "Thu, 26 Jul 2018 12:51:37 GMT", "version": "v3" } ]

2018-07-27

[ [ "Lipton", "Zachary C.", "" ], [ "Wang", "Yu-Xiang", "" ], [ "Smola", "Alex", "" ] ]

Mon, 12 Feb 2018 07:16:03 GMT

2018-02-12T07:16:03Z

[ "cs", "stat" ]

arxiv

null

1903.10605

Riley Simmons-Edler

Riley Simmons-Edler, Ben Eisner, Eric Mitchell, Sebastian Seung, Daniel Lee

Q-Learning for Continuous Actions with Cross-Entropy Guided Policies

null

cs.AI

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

Off-Policy reinforcement learning (RL) is an important class of methods for many problem domains, such as robotics, where the cost of collecting data is high and on-policy methods are consequently intractable. Standard methods for applying Q-learning to continuous-valued action domains involve iteratively sampling the Q-function to find a good action (e.g. via hill-climbing), or by learning a policy network at the same time as the Q-function (e.g. DDPG). Both approaches make tradeoffs between stability, speed, and accuracy. We propose a novel approach, called Cross-Entropy Guided Policies, or CGP, that draws inspiration from both classes of techniques. CGP aims to combine the stability and performance of iterative sampling policies with the low computational cost of a policy network. Our approach trains the Q-function using iterative sampling with the Cross-Entropy Method (CEM), while training a policy network to imitate CEM's sampling behavior. We demonstrate that our method is more stable to train than state of the art policy network methods, while preserving equivalent inference time compute costs, and achieving competitive total reward on standard benchmarks.

[ { "created": "Mon, 25 Mar 2019 21:46:58 GMT", "version": "v1" }, { "created": "Thu, 28 Mar 2019 21:52:02 GMT", "version": "v2" }, { "created": "Mon, 1 Jul 2019 21:03:09 GMT", "version": "v3" } ]

2019-07-03

[ [ "Simmons-Edler", "Riley", "" ], [ "Eisner", "Ben", "" ], [ "Mitchell", "Eric", "" ], [ "Seung", "Sebastian", "" ], [ "Lee", "Daniel", "" ] ]

Mon, 25 Mar 2019 21:46:58 GMT

2019-03-25T21:46:58Z

[ "cs" ]

arxiv

null

Editorial - Decision-making in acute medicine.

null

Acute physicians make patient-centred decisions at the start of the patient's hospital journey. Dozens more decisions are made by the individual members of the MDT (and, of course, by the patient) during the in-patient period. Decisions are made at every level of seniority and experience and range widely in scope and impact. The original articles in this issue are connected by a common thread: phenomena that inform and influence the decisions made by acute physicians.

null

pubmed

null

2,019

null

31,912,049

Acute medicine

1011.3750

Jun Xu

Jun Xu and Che Ming Ko

The effect of triangular flow on di-hadron azimuthal correlations in relativistic heavy ion collisions

5 pages, 5 figures, version accepted by Rapid Communication in Physical Review C

Phys.Rev.C83:021903,2011

10.1103/PhysRevC.83.021903

null

nucl-th hep-ex hep-ph nucl-ex

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

Using the AMPT model for relativistic heavy ion collisions, we have studied the di-hadron azimuthal angular correlations triggered by emitted jets in Au+Au collisions at center of mass energy $\sqrt{s_{NN}}=200$ GeV and impact parameter $b=8$ fm. A double-peak structure for the associated particles at the away side of trigger particles is obtained after subtracting background correlations due to the elliptic flow. Both the near-side peak and the away-side double peaks in the azimuthal angular correlations are, however, significantly suppressed (enhanced) in events of small (large) triangular flow, which are present as a result of fluctuations in the initial collision geometry. After subtraction of background correlations due to the triangular flow, the away-side double peaks change into a single peak with broad shoulders on both sides. The away side of the di-hadron correlations becomes essentially a single peak after further subtraction of higher-order flows.

[ { "created": "Tue, 16 Nov 2010 16:25:47 GMT", "version": "v1" }, { "created": "Fri, 24 Dec 2010 01:47:29 GMT", "version": "v2" }, { "created": "Tue, 25 Jan 2011 20:40:08 GMT", "version": "v3" } ]

2011-02-17

[ [ "Xu", "Jun", "" ], [ "Ko", "Che Ming", "" ] ]

Tue, 16 Nov 2010 16:25:47 GMT

2010-11-16T16:25:47Z

[ "physics" ]

arxiv

null

0807.4333

Berndt Muller

Rainer J. Fries, Berndt M\"uller, Andreas Sch\"afer

Stress Tensor and Bulk Viscosity in Relativistic Nuclear Collisions

null

Phys.Rev.C78:034913,2008

10.1103/PhysRevC.78.034913

null

nucl-th

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

We discuss the influence of different initial conditions for the stress tensor and the effect of bulk viscosity on the expansion and cooling of the fireball created in relativistic heavy-ion collisions. In particular, we explore the evolution of longitudinal and transverse components of the pressure and the extent of dissipative entropy production in the one-dimensional, boost-invariant hydrodynamic model. We find that a bulk viscosity consistent with recent estimates from lattice QCD further slows the equilibration of the system, however it does not significantly increase the entropy produced.

[ { "created": "Mon, 28 Jul 2008 02:19:29 GMT", "version": "v1" }, { "created": "Wed, 30 Jul 2008 13:18:00 GMT", "version": "v2" }, { "created": "Sat, 30 Aug 2008 11:24:52 GMT", "version": "v3" } ]

2008-11-26

[ [ "Fries", "Rainer J.", "" ], [ "Müller", "Berndt", "" ], [ "Schäfer", "Andreas", "" ] ]

Mon, 28 Jul 2008 02:19:29 GMT

2008-07-28T02:19:29Z

[ "physics" ]

arxiv

null

2010.08890

Giuseppe Brandi

Ioannis P. Antoniades, Giuseppe Brandi, L. G. Magafas, T. Di Matteo

The use of scaling properties to detect relevant changes in financial time series: a new visual warning tool

null

10.1016/j.physa.2020.125561

null

q-fin.ST q-fin.RM

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

The dynamical evolution of multiscaling in financial time series is investigated using time-dependent Generalized Hurst Exponents (GHE), $H_q$, for various values of the parameter $q$. Using $H_q$, we introduce a new visual methodology to algorithmically detect critical changes in the scaling of the underlying complex time-series. The methodology involves the degree of multiscaling at a particular time instance, the multiscaling trend which is calculated by the Change-Point Analysis method, and a rigorous evaluation of the statistical significance of the results. Using this algorithm, we have identified particular patterns in the temporal co-evolution of the different $H_q$ time-series. These GHE patterns, distinguish in a statistically robust way, not only between time periods of uniscaling and multiscaling, but also among different types of multiscaling: symmetric multiscaling (M) and asymmetric multiscaling (A). We apply the visual methodology to time-series comprising of daily close prices of four stock market indices: two major ones (S\&P~500 and NIKKEI) and two peripheral ones (Athens Stock Exchange general Index and Bombay-SENSEX). Results show that multiscaling varies greatly with time: time periods of strong multiscaling behavior and time periods of uniscaling behavior are interchanged while transitions from uniscaling to multiscaling behavior occur before critical market events, such as stock market bubbles. Moreover, particular asymmetric multiscaling patterns appear during critical stock market eras and provide useful information about market conditions. In particular, they can be used as 'fingerprints' of a turbulent market period as well as provide warning signals for an upcoming stock market 'bubble'. The applied visual methodology also appears to distinguish between exogenous and endogenous stock market crises, based on the observed patterns before the actual events.

[ { "created": "Sun, 18 Oct 2020 00:01:33 GMT", "version": "v1" }, { "created": "Tue, 20 Oct 2020 10:19:28 GMT", "version": "v2" }, { "created": "Tue, 3 Nov 2020 17:14:52 GMT", "version": "v3" }, { "created": "Wed, 9 Dec 2020 10:47:50 GMT", "version": "v4" } ]

2020-12-10

[ [ "Antoniades", "Ioannis P.", "" ], [ "Brandi", "Giuseppe", "" ], [ "Magafas", "L. G.", "" ], [ "Di Matteo", "T.", "" ] ]

Sun, 18 Oct 2020 00:01:33 GMT

2020-10-18T00:01:33Z

[ "q-fin" ]

arxiv

null

2102.01938

Prafulla Chandra Mr

Prafulla Chandra, Andrew Thangaraj and Nived Rajaraman

How good is Good-Turing for Markov samples?

null

cs.IT math.IT math.ST stat.ML stat.TH

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

The Good-Turing (GT) estimator for the missing mass (i.e., total probability of missing symbols) in $n$ samples is the number of symbols that appeared exactly once divided by $n$. For i.i.d. samples, the bias and squared-error risk of the GT estimator can be shown to fall as $1/n$ by bounding the expected error uniformly over all symbols. In this work, we study convergence of the GT estimator for missing stationary mass (i.e., total stationary probability of missing symbols) of Markov samples on an alphabet $\mathcal{X}$ with stationary distribution $[\pi_x:x \in \mathcal{X}]$ and transition probability matrix (t.p.m.) $P$. This is an important and interesting problem because GT is widely used in applications with temporal dependencies such as language models assigning probabilities to word sequences, which are modelled as Markov. We show that convergence of GT depends on convergence of $(P^{\sim x})^n$, where $P^{\sim x}$ is $P$ with the $x$-th column zeroed out. This, in turn, depends on the Perron eigenvalue $\lambda^{\sim x}$ of $P^{\sim x}$ and its relationship with $\pi_x$ uniformly over $x$. For randomly generated t.p.ms and t.p.ms derived from New York Times and Charles Dickens corpora, we numerically exhibit such uniform-over-$x$ relationships between $\lambda^{\sim x}$ and $\pi_x$. This supports the observed success of GT in language models and practical text data scenarios. For Markov chains with rank-2, diagonalizable t.p.ms having spectral gap $\beta$, we show minimax rate upper and lower bounds of $1/(n\beta^5)$ and $1/(n\beta)$, respectively, for the estimation of stationary missing mass. This theoretical result extends the $1/n$ minimax rate for i.i.d. or rank-1 t.p.ms to rank-2 Markov, and is a first such minimax rate result for missing mass of Markov samples.

[ { "created": "Wed, 3 Feb 2021 08:38:21 GMT", "version": "v1" }, { "created": "Sat, 6 Feb 2021 09:22:02 GMT", "version": "v2" }, { "created": "Sat, 27 May 2023 16:05:07 GMT", "version": "v3" } ]

2023-05-30

[ [ "Chandra", "Prafulla", "" ], [ "Thangaraj", "Andrew", "" ], [ "Rajaraman", "Nived", "" ] ]

Wed, 3 Feb 2021 08:38:21 GMT

2021-02-03T08:38:21Z

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

arxiv

null

1903.10063

Debarghya Mukherjee

Debarghya Mukherjee, Moulinath Banerjee and Ya'acov Ritov

Optimal Linear Discriminators For The Discrete Choice Model In Growing Dimensions

null

math.ST stat.TH

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

Manski's celebrated maximum score estimator for the discrete choice model, which is an optimal linear discriminator, has been the focus of much investigation in both the econometrics and statistics literatures, but its behavior under growing dimension scenarios largely remains unknown. This paper addresses that gap. Two different cases are considered: $p$ grows with $n$ but at a slow rate, i.e. $p/n \rightarrow 0$; and $p \gg n$ (fast growth). In the binary response model, we recast Manski's score estimation as empirical risk minimization for a classification problem, and derive the $\ell_2$ rate of convergence of the score estimator under a \emph{transition condition} in terms of our margin parameter that calibrates the level of difficulty of the estimation problem. We also establish upper and lower bounds for the minimax $\ell_2$ error in the binary choice model that differ by a logarithmic factor, and construct a minimax-optimal estimator in the slow growth regime. Some extensions to the general case -- the multinomial response model -- are also considered. Last but not least, we use a variety of learning algorithms to compute the maximum score estimator in growing dimensions.

[ { "created": "Sun, 24 Mar 2019 21:39:34 GMT", "version": "v1" }, { "created": "Sun, 14 Apr 2019 23:42:10 GMT", "version": "v2" }, { "created": "Sun, 9 Aug 2020 21:44:25 GMT", "version": "v3" } ]

2020-08-11

[ [ "Mukherjee", "Debarghya", "" ], [ "Banerjee", "Moulinath", "" ], [ "Ritov", "Ya'acov", "" ] ]

Sun, 24 Mar 2019 21:39:34 GMT

2019-03-24T21:39:34Z

[ "math", "stat" ]

arxiv

null

1706.04562

Davide Girolami

Davide Girolami, Tommaso Tufarelli and Cristian E. Susa

Quantifying genuine multipartite correlations and their pattern complexity

Close to published version

Phys. Rev. Lett. 119, 140505 (2017)

10.1103/PhysRevLett.119.140505

null

quant-ph cond-mat.dis-nn cs.IT math.IT nlin.AO physics.data-an

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

We propose an information-theoretic framework to quantify multipartite correlations in classical and quantum systems, answering questions such as: what is the amount of seven-partite correlations in a given state of ten particles? We identify measures of genuine multipartite correlations, i.e. statistical dependencies which cannot be ascribed to bipartite correlations, satisfying a set of desirable properties. Inspired by ideas developed in complexity science, we then introduce the concept of weaving to classify states which display different correlation patterns, but cannot be distinguished by correlation measures. The weaving of a state is defined as the weighted sum of correlations of every order. Weaving measures are good descriptors of the complexity of correlation structures in multipartite systems.

[ { "created": "Wed, 14 Jun 2017 16:09:26 GMT", "version": "v1" }, { "created": "Sat, 16 Sep 2017 15:29:12 GMT", "version": "v2" }, { "created": "Mon, 9 Oct 2017 15:19:13 GMT", "version": "v3" } ]

2017-10-10

[ [ "Girolami", "Davide", "" ], [ "Tufarelli", "Tommaso", "" ], [ "Susa", "Cristian E.", "" ] ]

Wed, 14 Jun 2017 16:09:26 GMT

2017-06-14T16:09:26Z

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

arxiv

null

2108.04165

Baoliang Chen

Baoliang Chen, Lingyu Zhu, Chenqi Kong, Hanwei Zhu, Shiqi Wang and Zhu Li

No-Reference Image Quality Assessment by Hallucinating Pristine Features

null

10.1109/TIP.2022.3205770

null

eess.IV cs.CV

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

In this paper, we propose a no-reference (NR) image quality assessment (IQA) method via feature level pseudo-reference (PR) hallucination. The proposed quality assessment framework is grounded on the prior models of natural image statistical behaviors and rooted in the view that the perceptually meaningful features could be well exploited to characterize the visual quality. Herein, the PR features from the distorted images are learned by a mutual learning scheme with the pristine reference as the supervision, and the discriminative characteristics of PR features are further ensured with the triplet constraints. Given a distorted image for quality inference, the feature level disentanglement is performed with an invertible neural layer for final quality prediction, leading to the PR and the corresponding distortion features for comparison. The effectiveness of our proposed method is demonstrated on four popular IQA databases, and superior performance on cross-database evaluation also reveals the high generalization capability of our method. The implementation of our method is publicly available on https://github.com/Baoliang93/FPR.

[ { "created": "Mon, 9 Aug 2021 16:48:34 GMT", "version": "v1" }, { "created": "Tue, 10 Aug 2021 04:24:03 GMT", "version": "v2" }, { "created": "Fri, 2 Sep 2022 04:02:02 GMT", "version": "v3" } ]

2022-10-26

[ [ "Chen", "Baoliang", "" ], [ "Zhu", "Lingyu", "" ], [ "Kong", "Chenqi", "" ], [ "Zhu", "Hanwei", "" ], [ "Wang", "Shiqi", "" ], [ "Li", "Zhu", "" ] ]

Mon, 9 Aug 2021 16:48:34 GMT

2021-08-09T16:48:34Z

[ "cs", "eess" ]

arxiv

null

1411.4724

Min-Seok Seo

Sang Hui Im and Min-Seok Seo

Peccei-Quinn invariant singlet extended SUSY with anomalous U(1) gauge symmetry

30 pages, 2 tables, references added, version published in JHEP

null

10.1007/JHEP05(2015)063

CTPU-14-12

hep-ph

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

Recent discovery of the SM-like Higgs boson with $m_h\simeq 125$ GeV motivates an extension of the minimal supersymmetric standard model (MSSM), which involves a singlet Higgs superfield with a sizable Yukawa coupling to the doublet Higgs superfields. We examine such singlet-extended SUSY models with a Peccei-Quinn (PQ) symmetry that originates from an anomalous $U(1)_A$ gauge symmetry. We focus on the specific scheme that the PQ symmetry is spontaneously broken at an intermediate scale $v_{\rm PQ}\sim \sqrt{m_{\rm SUSY}M_{\rm Pl}}$ by an interplay between Planck scale suppressed operators and tachyonic soft scalar mass $m_{\rm SUSY}\sim \sqrt{D_A}$ induced dominantly by the $U(1)_A$ $D$-term, $D_A$. This scheme also results in spontaneous SUSY breaking in the PQ sector, generating the gaugino masses $M_{1/2}\sim \sqrt{D_A}$ when it is transmitted to the MSSM sector by the conventional gauge mediation mechanism. As a result, the MSSM soft parameters in this scheme are induced mostly by the $U(1)_A$ $D$-term and the gauge mediated SUSY breaking from the PQ sector, so that the sparticle masses can be near the present experimental bounds without causing the SUSY flavor problem. The scheme is severely constrained by the condition that a phenomenologically viable form of the low energy operators of the singlet and doublet Higgs superfields is generated by the PQ breaking sector in a way similar to the Kim-Nilles solution of the $\mu$ problem, and the resulting Higgs mass parameters allow the electroweak symmetry breaking with small $\tan\beta$. We find two minimal models with two singlet Higgs superfields, satisfying this condition with a relatively simple form of the PQ breaking sector, and briefly discuss some phenomenological aspects of the model.

[ { "created": "Tue, 18 Nov 2014 03:22:52 GMT", "version": "v1" }, { "created": "Mon, 8 Dec 2014 04:10:42 GMT", "version": "v2" }, { "created": "Thu, 14 May 2015 01:32:07 GMT", "version": "v3" } ]

2015-06-11

[ [ "Im", "Sang Hui", "" ], [ "Seo", "Min-Seok", "" ] ]

Tue, 18 Nov 2014 03:22:52 GMT

2014-11-18T03:22:52Z

[ "physics" ]

arxiv

null

1906.08871

Gautam Krishna

Gautam Krishna, Co Tran, Mason Carnahan, Ahmed H Tewfik

Advancing Speech Recognition With No Speech Or With Noisy Speech

Extended version of our accepted IEEE EUSIPCO 2019 paper with additional results for CTC model based recognition. arXiv admin note: substantial text overlap with arXiv:1906.08045, arXiv:1906.08044

null

eess.AS cs.CL cs.LG cs.SD stat.ML

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

In this paper we demonstrate end-to-end continuous speech recognition (CSR) using electroencephalography (EEG) signals with no speech signal as input. An attention model based automatic speech recognition (ASR) and connectionist temporal classification (CTC) based ASR systems were implemented for performing recognition. We further demonstrate CSR for noisy speech by fusing with EEG features.

[ { "created": "Mon, 17 Jun 2019 23:06:51 GMT", "version": "v1" }, { "created": "Wed, 17 Jul 2019 03:13:24 GMT", "version": "v2" }, { "created": "Sat, 27 Jul 2019 03:46:44 GMT", "version": "v3" }, { "created": "Sat, 17 Aug 2019 02:44:48 GMT", "version": "v4" }, { "created": "Wed, 21 Aug 2019 19:00:45 GMT", "version": "v5" }, { "created": "Sun, 10 Nov 2019 02:47:45 GMT", "version": "v6" }, { "created": "Wed, 13 Nov 2019 03:53:06 GMT", "version": "v7" }, { "created": "Wed, 4 Mar 2020 22:28:26 GMT", "version": "v8" }, { "created": "Sun, 15 Mar 2020 02:51:46 GMT", "version": "v9" } ]

2020-03-17

[ [ "Krishna", "Gautam", "" ], [ "Tran", "Co", "" ], [ "Carnahan", "Mason", "" ], [ "Tewfik", "Ahmed H", "" ] ]

Mon, 17 Jun 2019 23:06:51 GMT

2019-06-17T23:06:51Z

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

arxiv

null

2002.02081

Nan Jiang

Nan Jiang, Jiawei Huang

Minimax Value Interval for Off-Policy Evaluation and Policy Optimization

null

cs.LG math.OC stat.ML

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

We study minimax methods for off-policy evaluation (OPE) using value functions and marginalized importance weights. Despite that they hold promises of overcoming the exponential variance in traditional importance sampling, several key problems remain: (1) They require function approximation and are generally biased. For the sake of trustworthy OPE, is there anyway to quantify the biases? (2) They are split into two styles ("weight-learning" vs "value-learning"). Can we unify them? In this paper we answer both questions positively. By slightly altering the derivation of previous methods (one from each style; Uehara et al., 2020), we unify them into a single value interval that comes with a special type of double robustness: when either the value-function or the importance-weight class is well specified, the interval is valid and its length quantifies the misspecification of the other class. Our interval also provides a unified view of and new insights to some recent methods, and we further explore the implications of our results on exploration and exploitation in off-policy policy optimization with insufficient data coverage.

[ { "created": "Thu, 6 Feb 2020 02:54:11 GMT", "version": "v1" }, { "created": "Wed, 26 Feb 2020 14:56:55 GMT", "version": "v2" }, { "created": "Sat, 4 Jul 2020 13:55:25 GMT", "version": "v3" }, { "created": "Sun, 11 Oct 2020 00:41:22 GMT", "version": "v4" }, { "created": "Mon, 26 Oct 2020 05:12:58 GMT", "version": "v5" }, { "created": "Wed, 4 Nov 2020 23:43:32 GMT", "version": "v6" } ]

2020-11-06

[ [ "Jiang", "Nan", "" ], [ "Huang", "Jiawei", "" ] ]

Thu, 6 Feb 2020 02:54:11 GMT

2020-02-06T02:54:11Z

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

arxiv

null

2005.03675

Ines Chami

Ines Chami, Sami Abu-El-Haija, Bryan Perozzi, Christopher R\'e, Kevin Murphy

Machine Learning on Graphs: A Model and Comprehensive Taxonomy

null

cs.LG cs.NE cs.SI stat.ML

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

There has been a surge of recent interest in learning representations for graph-structured data. Graph representation learning methods have generally fallen into three main categories, based on the availability of labeled data. The first, network embedding (such as shallow graph embedding or graph auto-encoders), focuses on learning unsupervised representations of relational structure. The second, graph regularized neural networks, leverages graphs to augment neural network losses with a regularization objective for semi-supervised learning. The third, graph neural networks, aims to learn differentiable functions over discrete topologies with arbitrary structure. However, despite the popularity of these areas there has been surprisingly little work on unifying the three paradigms. Here, we aim to bridge the gap between graph neural networks, network embedding and graph regularization models. We propose a comprehensive taxonomy of representation learning methods for graph-structured data, aiming to unify several disparate bodies of work. Specifically, we propose a Graph Encoder Decoder Model (GRAPHEDM), which generalizes popular algorithms for semi-supervised learning on graphs (e.g. GraphSage, Graph Convolutional Networks, Graph Attention Networks), and unsupervised learning of graph representations (e.g. DeepWalk, node2vec, etc) into a single consistent approach. To illustrate the generality of this approach, we fit over thirty existing methods into this framework. We believe that this unifying view both provides a solid foundation for understanding the intuition behind these methods, and enables future research in the area.

[ { "created": "Thu, 7 May 2020 18:00:02 GMT", "version": "v1" }, { "created": "Mon, 18 Jan 2021 15:41:03 GMT", "version": "v2" }, { "created": "Tue, 12 Apr 2022 03:24:00 GMT", "version": "v3" } ]

2022-04-13

[ [ "Chami", "Ines", "" ], [ "Abu-El-Haija", "Sami", "" ], [ "Perozzi", "Bryan", "" ], [ "Ré", "Christopher", "" ], [ "Murphy", "Kevin", "" ] ]

Thu, 7 May 2020 18:00:02 GMT

2020-05-07T18:00:02Z

[ "cs", "stat" ]

arxiv

null

1406.2536

Arutyunov Gleb E

Gleb Arutyunov, Daniel Medina-Rincon

Deformed Neumann model from spinning strings on (AdS_5 x S^5)_\eta

17 pages, no figures; v3: published version

null

10.1007/JHEP10(2014)050

ITP-UU-14/17, SPIN-14/19

hep-th nlin.SI

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

We show that bosonic spinning strings on the \eta-deformed AdS_5 x S^5 background are naturally described as periodic solutions of a novel finite-dimensional integrable system which can be viewed as a deformation of the celebrated Neumann model. For this deformed model we find the Lax representation and the analogue of the Uhlenbeck integrals.

[ { "created": "Tue, 10 Jun 2014 13:12:12 GMT", "version": "v1" }, { "created": "Thu, 12 Jun 2014 16:18:34 GMT", "version": "v2" }, { "created": "Thu, 8 Sep 2016 14:58:48 GMT", "version": "v3" } ]

2016-09-09

[ [ "Arutyunov", "Gleb", "" ], [ "Medina-Rincon", "Daniel", "" ] ]

Tue, 10 Jun 2014 13:12:12 GMT

2014-06-10T13:12:12Z

[ "physics", "nlin" ]

arxiv

null

1607.03884

Kyle Niemeyer

Nicholas J. Curtis, Kyle E. Niemeyer, and Chih-Jen Sung

An investigation of GPU-based stiff chemical kinetics integration methods

34 pages, 6 figures; pdfLaTeX

Combust. Flame 179 (2017) 312-324

10.1016/j.combustflame.2017.02.005

null

physics.comp-ph cs.DC physics.chem-ph

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

A fifth-order implicit Runge-Kutta method and two fourth-order exponential integration methods equipped with Krylov subspace approximations were implemented for the GPU and paired with the analytical chemical kinetic Jacobian software pyJac. The performance of each algorithm was evaluated by integrating thermochemical state data sampled from stochastic partially stirred reactor simulations and compared with the commonly used CPU-based implicit integrator CVODE. We estimated that the implicit Runge-Kutta method running on a single GPU is equivalent to CVODE running on 12-38 CPU cores for integration of a single global integration time step of 1e-6 s with hydrogen and methane models. In the stiffest case studied---the methane model with a global integration time step of 1e-4 s---thread divergence and higher memory traffic significantly decreased GPU performance to the equivalent of CVODE running on approximately three CPU cores. The exponential integration algorithms performed more slowly than the implicit integrators on both the CPU and GPU. Thread divergence and memory traffic were identified as the main limiters of GPU integrator performance, and techniques to mitigate these issues were discussed. Use of a finite-difference Jacobian on the GPU---in place of the analytical Jacobian provided by pyJac---greatly decreased integrator performance due to thread divergence, resulting in maximum slowdowns of 7.11-240.96 times; in comparison, the corresponding slowdowns on the CPU were just 1.39-2.61 times, underscoring the importance of use of an analytical Jacobian for efficient GPU integration. Finally, future research directions for working towards enabling realistic chemistry in reactive-flow simulations via GPU\slash SIMD accelerated stiff chemical kinetic integration were identified.

[ { "created": "Wed, 13 Jul 2016 19:47:57 GMT", "version": "v1" }, { "created": "Mon, 8 Aug 2016 19:47:54 GMT", "version": "v2" }, { "created": "Tue, 14 Feb 2017 17:37:42 GMT", "version": "v3" } ]

2017-03-30

[ [ "Curtis", "Nicholas J.", "" ], [ "Niemeyer", "Kyle E.", "" ], [ "Sung", "Chih-Jen", "" ] ]

Wed, 13 Jul 2016 19:47:57 GMT

2016-07-13T19:47:57Z

[ "physics", "cs" ]

arxiv

null

1810.08985

Abhishek Dubey

Sanchita Basak, Saptarshi Sengupta, Abhishek Dubey

Mechanisms for Integrated Feature Normalization and Remaining Useful Life Estimation Using LSTMs Applied to Hard-Disks

9 pages, 13 figures, 2 tables

Proceedings of IEEE Smartcomp 2019

null

cs.LG stat.ML

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

With emerging smart communities, improving overall system availability is becoming a major concern. In order to improve the reliability of the components in a system we propose an inference model to predict Remaining Useful Life (RUL) of those components. In this paper we work with components of backend data servers such as hard disks, that are subject to degradation. A Deep Long-Short Term Memory (LSTM) Network is used as the backbone of this fast, data-driven decision framework and dynamically captures the pattern of the incoming data. In the article, we discuss the architecture of the neural network and describe the mechanisms to choose the various hyper-parameters. Further, we describe the challenges faced in extracting effective training sets from highly unorganized and class-imbalanced big data and establish methods for online predictions with extensive data pre-processing, feature extraction and validation through online simulation sets with unknown remaining useful lives of the hard disks. Our algorithm performs especially well in predicting RUL near the critical zone of a device approaching failure. With the proposed approach we are able to predict whether a disk is going to fail in next ten days with an average precision of 0.8435. We also show that the architecture trained on a particular model can be used to predict RUL for devices in different models from same manufacturer through transfer learning.

[ { "created": "Sun, 21 Oct 2018 16:24:46 GMT", "version": "v1" }, { "created": "Tue, 12 Feb 2019 08:57:58 GMT", "version": "v2" }, { "created": "Mon, 17 Jun 2019 00:41:38 GMT", "version": "v3" } ]

2019-06-18

[ [ "Basak", "Sanchita", "" ], [ "Sengupta", "Saptarshi", "" ], [ "Dubey", "Abhishek", "" ] ]

Sun, 21 Oct 2018 16:24:46 GMT

2018-10-21T16:24:46Z

[ "cs", "stat" ]

arxiv

null

BML-111, a lipoxin receptor agonist, ameliorates 'two-hit'-induced acute pancreatitis-associated lung injury in mice by the upregulation of heme oxygenase-1.

null

The objective of this study is to investigate the effects of BML-111 on acute pancreatitis-associated lung injury (APALI) induced by cerulein with subsequent an LPS administration in mice and its possible mechanisms. One hundred and twenty-eight mice were randomly allocated to four groups, namely the APALI group, the BML-111 pretreatment group, the BM-111 control group, and the control group. The 'two-hit' mice APALI model was established by intraperitoneal injection of cerulein 7 times at hourly intervals and Escherichia coli lipopolysaccharide (LPS) once after the last dose of cerulein immediately. The samples were taken at 3, 6, 12, and 24 h after the last injection. Serum levels of amylase, TNF-a, IL-1β and IL-10, were determined. Histological score of the pancreas and lung, the wet/dry weight ratio, and heme oxygenase-1 (HO-1) expression in the lung were also evaluated. BML-111 pretreatment significantly reduced the serum levels of amylase, TNF-α, IL-1β, the wet/dry weight ratio of lung, and the pathology injury scores of pancreas and lung, and the serum levels of IL-10 were markedly increased. The severity of pancreatic and lung histology were also significantly improved by the administration of BML-111, and the expressions of HO-1 in lung tissues also increased in the BML-111 group compared with those in the APALI group. In conclusion, BML-111 exerts protective effects on APALI induced by cerulein and LPS. In addition to its anti-inflammatory effects, the beneficial effects may also be due to the upregulation of HO-1 expression in the lung tissues.

null

pubmed

null

2,014

null

23,802,775

Artificial cells, nanomedicine, and biotechnology

1802.06458

Jayaraman J. Thiagarajan

Deepta Rajan and Jayaraman J. Thiagarajan

A Generative Modeling Approach to Limited Channel ECG Classification

null

stat.ML cs.LG

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

Processing temporal sequences is central to a variety of applications in health care, and in particular multi-channel Electrocardiogram (ECG) is a highly prevalent diagnostic modality that relies on robust sequence modeling. While Recurrent Neural Networks (RNNs) have led to significant advances in automated diagnosis with time-series data, they perform poorly when models are trained using a limited set of channels. A crucial limitation of existing solutions is that they rely solely on discriminative models, which tend to generalize poorly in such scenarios. In order to combat this limitation, we develop a generative modeling approach to limited channel ECG classification. This approach first uses a Seq2Seq model to implicitly generate the missing channel information, and then uses the latent representation to perform the actual supervisory task. This decoupling enables the use of unsupervised data and also provides highly robust metric spaces for subsequent discriminative learning. Our experiments with the Physionet dataset clearly evidence the effectiveness of our approach over standard RNNs in disease prediction.

[ { "created": "Sun, 18 Feb 2018 22:29:31 GMT", "version": "v1" }, { "created": "Tue, 20 Feb 2018 01:53:25 GMT", "version": "v2" }, { "created": "Thu, 14 Jun 2018 01:32:04 GMT", "version": "v3" } ]

2018-07-17

[ [ "Rajan", "Deepta", "" ], [ "Thiagarajan", "Jayaraman J.", "" ] ]

Sun, 18 Feb 2018 22:29:31 GMT

2018-02-18T22:29:31Z

[ "cs", "stat" ]

arxiv

null

2205.15049

Bahar Taskesen

Yves Rychener, Bahar Taskesen, Daniel Kuhn

Metrizing Fairness

null

cs.LG math.OC stat.ML

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

We study supervised learning problems that have significant effects on individuals from two demographic groups, and we seek predictors that are fair with respect to a group fairness criterion such as statistical parity (SP). A predictor is SP-fair if the distributions of predictions within the two groups are close in Kolmogorov distance, and fairness is achieved by penalizing the dissimilarity of these two distributions in the objective function of the learning problem. In this paper, we identify conditions under which hard SP constraints are guaranteed to improve predictive accuracy. We also showcase conceptual and computational benefits of measuring unfairness with integral probability metrics (IPMs) other than the Kolmogorov distance. Conceptually, we show that the generator of any IPM can be interpreted as a family of utility functions and that unfairness with respect to this IPM arises if individuals in the two demographic groups have diverging expected utilities. We also prove that the unfairness-regularized prediction loss admits unbiased gradient estimators, which are constructed from random mini-batches of training samples, if unfairness is measured by the squared $\mathcal L^2$-distance or by a squared maximum mean discrepancy. In this case, the fair learning problem is susceptible to efficient stochastic gradient descent (SGD) algorithms. Numerical experiments on synthetic and real data show that these SGD algorithms outperform state-of-the-art methods for fair learning in that they achieve superior accuracy-unfairness trade-offs -- sometimes orders of magnitude faster.

[ { "created": "Mon, 30 May 2022 12:28:10 GMT", "version": "v1" }, { "created": "Thu, 2 Jun 2022 07:52:03 GMT", "version": "v2" }, { "created": "Fri, 10 Mar 2023 16:31:08 GMT", "version": "v3" }, { "created": "Wed, 22 May 2024 15:33:23 GMT", "version": "v4" }, { "created": "Tue, 11 Jun 2024 09:34:06 GMT", "version": "v5" } ]

2024-06-12

[ [ "Rychener", "Yves", "" ], [ "Taskesen", "Bahar", "" ], [ "Kuhn", "Daniel", "" ] ]

Mon, 30 May 2022 12:28:10 GMT

2022-05-30T12:28:10Z

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

arxiv

null

1705.09024

Jiangbin Lyu Dr.

Jiangbin Lyu, Yong Zeng and Rui Zhang

UAV-Aided Offloading for Cellular Hotspot

31 pages, 8 figures, accepted for publication in IEEE Transactions on Wireless Communications https://ieeexplore.ieee.org/document/8329013/

null

10.1109/TWC.2018.2818734

null

cs.IT math.IT

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

In conventional terrestrial cellular networks, mobile terminals (MTs) at the cell edge often pose a performance bottleneck due to their long distances from the serving ground base station (GBS), especially in hotspot period when the GBS is heavily loaded. This paper proposes a new hybrid network architecture by leveraging the use of unmanned aerial vehicle (UAV) as an aerial mobile base station, which flies cyclically along the cell edge to offload data traffic for cell-edge MTs. We aim to maximize the minimum throughput of all MTs by jointly optimizing the UAV's trajectory, bandwidth allocation and user partitioning. We first consider orthogonal spectrum sharing between the UAV and GBS, and then extend to spectrum reuse where the total bandwidth is shared by both the GBS and UAV with their mutual interference effectively avoided. Numerical results show that the proposed hybrid network with optimized spectrum sharing and cyclical multiple access design significantly improves the spatial throughput over the conventional GBS-only network; while the spectrum reuse scheme provides further throughput gains at the cost of slightly higher complexity for interference control. Moreover, compared to the conventional small-cell offloading scheme, the proposed UAV offloading scheme is shown to outperform in terms of throughput, besides saving the infrastructure cost.

[ { "created": "Thu, 25 May 2017 02:14:58 GMT", "version": "v1" }, { "created": "Thu, 24 Aug 2017 11:35:55 GMT", "version": "v2" }, { "created": "Mon, 29 Jan 2018 06:52:04 GMT", "version": "v3" }, { "created": "Sun, 20 May 2018 04:26:55 GMT", "version": "v4" } ]

2018-05-22

[ [ "Lyu", "Jiangbin", "" ], [ "Zeng", "Yong", "" ], [ "Zhang", "Rui", "" ] ]

Thu, 25 May 2017 02:14:58 GMT

2017-05-25T02:14:58Z

[ "cs", "math" ]

arxiv

null

1806.10586

Yu Bai

Yu Bai, Tengyu Ma, Andrej Risteski

Approximability of Discriminators Implies Diversity in GANs

null

cs.LG cs.DS stat.ML

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

While Generative Adversarial Networks (GANs) have empirically produced impressive results on learning complex real-world distributions, recent works have shown that they suffer from lack of diversity or mode collapse. The theoretical work of Arora et al. suggests a dilemma about GANs' statistical properties: powerful discriminators cause overfitting, whereas weak discriminators cannot detect mode collapse. By contrast, we show in this paper that GANs can in principle learn distributions in Wasserstein distance (or KL-divergence in many cases) with polynomial sample complexity, if the discriminator class has strong distinguishing power against the particular generator class (instead of against all possible generators). For various generator classes such as mixture of Gaussians, exponential families, and invertible and injective neural networks generators, we design corresponding discriminators (which are often neural nets of specific architectures) such that the Integral Probability Metric (IPM) induced by the discriminators can provably approximate the Wasserstein distance and/or KL-divergence. This implies that if the training is successful, then the learned distribution is close to the true distribution in Wasserstein distance or KL divergence, and thus cannot drop modes. Our preliminary experiments show that on synthetic datasets the test IPM is well correlated with KL divergence or the Wasserstein distance, indicating that the lack of diversity in GANs may be caused by the sub-optimality in optimization instead of statistical inefficiency.

[ { "created": "Wed, 27 Jun 2018 17:33:52 GMT", "version": "v1" }, { "created": "Thu, 5 Jul 2018 06:15:38 GMT", "version": "v2" }, { "created": "Tue, 24 Jul 2018 06:19:39 GMT", "version": "v3" }, { "created": "Mon, 1 Jul 2019 05:59:03 GMT", "version": "v4" } ]

2019-07-02

[ [ "Bai", "Yu", "" ], [ "Ma", "Tengyu", "" ], [ "Risteski", "Andrej", "" ] ]

Wed, 27 Jun 2018 17:33:52 GMT

2018-06-27T17:33:52Z

[ "cs", "stat" ]

arxiv

null

2008.08970

Monika Csikos

M\'onika Csik\'os and Nabil H. Mustafa

Optimal Approximations Made Easy

null

Published in Information Processing Letters, Volume 176, June 2022, 106250

null

cs.LG cs.CG math.CO stat.ML

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

The fundamental result of Li, Long, and Srinivasan on approximations of set systems has become a key tool across several communities such as learning theory, algorithms, computational geometry, combinatorics and data analysis. The goal of this paper is to give a modular, self-contained, intuitive proof of this result for finite set systems. The only ingredient we assume is the standard Chernoff's concentration bound. This makes the proof accessible to a wider audience, readers not familiar with techniques from statistical learning theory, and makes it possible to be covered in a single self-contained lecture in a geometry, algorithms or combinatorics course.

[ { "created": "Thu, 20 Aug 2020 13:58:14 GMT", "version": "v1" }, { "created": "Fri, 26 Mar 2021 16:16:31 GMT", "version": "v2" }, { "created": "Thu, 1 Sep 2022 13:24:04 GMT", "version": "v3" } ]

2022-09-02

[ [ "Csikós", "Mónika", "" ], [ "Mustafa", "Nabil H.", "" ] ]

Thu, 20 Aug 2020 13:58:14 GMT

2020-08-20T13:58:14Z

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

arxiv

null

1807.03368

Jos\'e Bento

Sam Safavi and Jos\'e Bento

Tractable $n$-Metrics for Multiple Graphs

null

cs.DM cs.LG eess.SP math.CO math.OC

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

Graphs are used in almost every scientific discipline to express relations among a set of objects. Algorithms that compare graphs, and output a closeness score, or a correspondence among their nodes, are thus extremely important. Despite the large amount of work done, many of the scalable algorithms to compare graphs do not produce closeness scores that satisfy the intuitive properties of metrics. This is problematic since non-metrics are known to degrade the performance of algorithms such as distance-based clustering of graphs (Stratis and Bento 2018). On the other hand, the use of metrics increases the performance of several machine learning tasks (Indyk et al. 1999, Clarkson et al. 1999, Angiulli et al. 2002, Ackermann et al. 2010). In this paper, we introduce a new family of multi-distances (a distance between more than two elements) that satisfies a generalization of the properties of metrics to multiple elements. In the context of comparing graphs, we are the first to show the existence of multi-distances that simultaneously incorporate the useful property of alignment consistency (Nguyen et al. 2011), and a generalized metric property. Furthermore, we show that these multi-distances can be relaxed to convex optimization problems, without losing the generalized metric property.

[ { "created": "Mon, 9 Jul 2018 20:13:05 GMT", "version": "v1" }, { "created": "Wed, 11 Jul 2018 16:41:43 GMT", "version": "v2" }, { "created": "Mon, 8 Oct 2018 18:35:30 GMT", "version": "v3" }, { "created": "Sat, 14 Nov 2020 21:08:21 GMT", "version": "v4" } ]

2020-11-17

[ [ "Safavi", "Sam", "" ], [ "Bento", "José", "" ] ]

Mon, 9 Jul 2018 20:13:05 GMT

2018-07-09T20:13:05Z

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

arxiv

null

2107.12719

Alessio Xompero

Alessio Xompero, Santiago Donaher, Vladimir Iashin, Francesca Palermo, G\"okhan Solak, Claudio Coppola, Reina Ishikawa, Yuichi Nagao, Ryo Hachiuma, Qi Liu, Fan Feng, Chuanlin Lan, Rosa H. M. Chan, Guilherme Christmann, Jyun-Ting Song, Gonuguntla Neeharika, Chinnakotla Krishna Teja Reddy, Dinesh Jain, Bakhtawar Ur Rehman, Andrea Cavallaro

The CORSMAL benchmark for the prediction of the properties of containers

Authors' post-print accepted for publication in IEEE Access, see https://doi.org/10.1109/ACCESS.2022.3166906 . 14 pages, 6 tables, 7 figures

IEEE Access, vol. 10, 2022, 1-15

10.1109/ACCESS.2022.3166906

null

cs.MM cs.CV cs.SD eess.AS

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

The contactless estimation of the weight of a container and the amount of its content manipulated by a person are key pre-requisites for safe human-to-robot handovers. However, opaqueness and transparencies of the container and the content, and variability of materials, shapes, and sizes, make this estimation difficult. In this paper, we present a range of methods and an open framework to benchmark acoustic and visual perception for the estimation of the capacity of a container, and the type, mass, and amount of its content. The framework includes a dataset, specific tasks and performance measures. We conduct an in-depth comparative analysis of methods that used this framework and audio-only or vision-only baselines designed from related works. Based on this analysis, we can conclude that audio-only and audio-visual classifiers are suitable for the estimation of the type and amount of the content using different types of convolutional neural networks, combined with either recurrent neural networks or a majority voting strategy, whereas computer vision methods are suitable to determine the capacity of the container using regression and geometric approaches. Classifying the content type and level using only audio achieves a weighted average F1-score up to 81% and 97%, respectively. Estimating the container capacity with vision-only approaches and estimating the filling mass with audio-visual multi-stage approaches reach up to 65% weighted average capacity and mass scores. These results show that there is still room for improvement on the design of new methods. These new methods can be ranked and compared on the individual leaderboards provided by our open framework.

[ { "created": "Tue, 27 Jul 2021 10:36:19 GMT", "version": "v1" }, { "created": "Tue, 30 Nov 2021 20:21:48 GMT", "version": "v2" }, { "created": "Thu, 21 Apr 2022 11:17:22 GMT", "version": "v3" } ]

2022-04-22

[ [ "Xompero", "Alessio", "" ], [ "Donaher", "Santiago", "" ], [ "Iashin", "Vladimir", "" ], [ "Palermo", "Francesca", "" ], [ "Solak", "Gökhan", "" ], [ "Coppola", "Claudio", "" ], [ "Ishikawa", "Reina", "" ], [ "Nagao", "Yuichi", "" ], [ "Hachiuma", "Ryo", "" ], [ "Liu", "Qi", "" ], [ "Feng", "Fan", "" ], [ "Lan", "Chuanlin", "" ], [ "Chan", "Rosa H. M.", "" ], [ "Christmann", "Guilherme", "" ], [ "Song", "Jyun-Ting", "" ], [ "Neeharika", "Gonuguntla", "" ], [ "Reddy", "Chinnakotla Krishna Teja", "" ], [ "Jain", "Dinesh", "" ], [ "Rehman", "Bakhtawar Ur", "" ], [ "Cavallaro", "Andrea", "" ] ]

Tue, 27 Jul 2021 10:36:19 GMT

2021-07-27T10:36:19Z

[ "cs", "eess" ]

arxiv

null

The {C}ambridge language survey

null

The Cambridge Language Survey is a research project whose activities centre around the use of an Integrated Language Database, whereby a computerised dictionary is used for intelligent cross-reference during corpus analysis - searching for example for all the inflections of a verb rather than just the base form. Types of grammatical coding and semantic categorisation appropriate to such a computerised dictionary are discussed, as are software tools for parsing, finding collocations, and performing sense-tagging. The weighted evaluation of semantic, grammatical, and collocational information to discriminate between word senses is described in some detail. Mention is made of several branches of research including the development of parallel corpora, semantic interpretation by sense-tagging, and the use of a Learner Corpus for the analysis of errors made by non-native-speakers. Sense-tagging is identified as an under-exploited approach to language analysis and one for which great opportunities for product development exist.

null

acl

inproceedings

procter-1993-cambridge

Frederking, Robert E. and Taylor, Kathryn B.

apr # " 26–28"

1,993

Heidelberg, Germany

Springer Berlin Heidelberg

https://aclanthology.org/1993.eamt-1.5/

Procter, Paul

Third International EAMT Workshop: Machine Translation and the Lexicon

77--84

null

1911.11041

Denis Boyer

Gabriel Mercado-V\'asquez, Denis Boyer

First hitting times to intermittent targets

3 Figures

Physical Review Letters 123, 250603 (2019)

10.1103/PhysRevLett.123.250603

null

cond-mat.stat-mech q-bio.SC

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

In noisy environments such as the cell, many processes involve target sites that are often hidden or inactive, and thus not always available for reaction with diffusing entities. To understand reaction kinetics in these situations, we study the first hitting time statistics of a Brownian particle searching for a target site that switches stochastically between visible and hidden phases. At high crypticity, an unexpected rate limited power-law regime emerges for the first hitting time density, which markedly differs from the classic $t^{-3/2}$ scaling for steady targets. Our problem admits an asymptotic mapping onto a mixed, or Robin, boundary condition. Similar results are obtained with non-Markov targets and particles diffusing anomalously.

[ { "created": "Mon, 25 Nov 2019 16:53:32 GMT", "version": "v1" }, { "created": "Tue, 7 Jan 2020 18:56:50 GMT", "version": "v2" } ]

2020-01-29

[ [ "Mercado-Vásquez", "Gabriel", "" ], [ "Boyer", "Denis", "" ] ]

Mon, 25 Nov 2019 16:53:32 GMT

2019-11-25T16:53:32Z

[ "q-bio", "physics" ]

arxiv

null

1910.05067

Sergio P. P\'erez

Antonio Russo, Sergio P. Perez, Miguel A. Dur\'an-Olivencia, Peter Yatsyshin, Jos\'e A. Carrillo, Serafim Kalliadasis

A Finite-Volume Method for Fluctuating Dynamical Density Functional Theory

null

physics.comp-ph cond-mat.stat-mech cs.NA math.NA physics.flu-dyn

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

We introduce a finite-volume numerical scheme for solving stochastic gradient-flow equations. Such equations are of crucial importance within the framework of fluctuating hydrodynamics and dynamic density functional theory. Our proposed scheme deals with general free-energy functionals, including, for instance, external fields or interaction potentials. This allows us to simulate a range of physical phenomena where thermal fluctuations play a crucial role, such as nucleation and other energy-barrier crossing transitions. A positivity-preserving algorithm for the density is derived based on a hybrid space discretization of the deterministic and the stochastic terms and different implicit and explicit time integrators. We show through numerous applications that not only our scheme is able to accurately reproduce the statistical properties (structure factor and correlations) of the physical system, but, because of the multiplicative noise, it allows us to simulate energy barrier crossing dynamics, which cannot be captured by mean-field approaches.

[ { "created": "Fri, 11 Oct 2019 10:27:15 GMT", "version": "v1" }, { "created": "Tue, 31 Dec 2019 18:03:34 GMT", "version": "v2" }, { "created": "Thu, 31 Dec 2020 00:35:52 GMT", "version": "v3" }, { "created": "Fri, 1 Jan 2021 01:51:10 GMT", "version": "v4" } ]

2021-01-05

[ [ "Russo", "Antonio", "" ], [ "Perez", "Sergio P.", "" ], [ "Durán-Olivencia", "Miguel A.", "" ], [ "Yatsyshin", "Peter", "" ], [ "Carrillo", "José A.", "" ], [ "Kalliadasis", "Serafim", "" ] ]

Fri, 11 Oct 2019 10:27:15 GMT

2019-10-11T10:27:15Z

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

arxiv

null

2104.07099

Francesco Ferrante

Francesco Ferrante, Giacomo Casadei, and Christophe Prieur

Synchronization of Identical Boundary-Actuated Semilinear Infinite-Dimensional Systems

V1 is the extended version of the original submission before peer review. V2 is the extended version revised based on reviewers' feedback. V2 fixes a number of mistakes and typos in V1. In particular, an oversight in the statement of Proposition 2 has been fixed and title has been updated. V3 includes the link to the published version. V4 fixes a missing symbol in the published version

IEEE Control Systems Letters, Volume: 6, pages 1322-1327, 2022

10.1109/LCSYS.2021.3081481

null

eess.SY cs.SY

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

This paper deals with synchronization of a class of infinite-dimensional systems. The considered network is described by a collection of semilinear Lipschitz boundary-actuated infinite-dimensional dynamics. For undirected connected graphs, sufficient conditions for asymptotic synchronization are established. We show that the proposed conditions when applied to systems of hyperbolic semilinear conservation laws can be recast into a set of matrix inequalities. For this class of systems, sufficient conditions in the form of linear matrix inequalities for the design of synchronizing policies are provided.

[ { "created": "Wed, 14 Apr 2021 20:01:56 GMT", "version": "v1" }, { "created": "Sat, 15 May 2021 10:48:41 GMT", "version": "v2" }, { "created": "Wed, 1 Sep 2021 10:28:49 GMT", "version": "v3" }, { "created": "Wed, 3 Nov 2021 15:11:30 GMT", "version": "v4" }, { "created": "Thu, 25 Nov 2021 08:21:45 GMT", "version": "v5" } ]

2021-11-29

[ [ "Ferrante", "Francesco", "" ], [ "Casadei", "Giacomo", "" ], [ "Prieur", "Christophe", "" ] ]

Wed, 14 Apr 2021 20:01:56 GMT

2021-04-14T20:01:56Z

[ "cs", "eess" ]

arxiv

null

2006.12679

Tianhong Wang

Tianhong Wang, Vladimir Khudik, and Gennady Shvets

Laser-Pulse and Electron-Bunch Plasma Wakefield Accelerator

null

10.1103/PhysRevAccelBeams.23.111304

null

physics.plasm-ph

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

Propagation distances of intense laser pulses and high-charge electron beams through the plasma are, respectively, limited by diffraction and self-deceleration. This imposes severe constraints on the performance of the two major advanced accelerator concepts: laser and plasma wakefield accelerators. Using numerical simulations, we demonstrate that when the two beams co-propagate in the plasma, they can interact synergistically and extend each other's travel distances. The key interactions responsible for the synergy are found to be laser channeling by the electron bunch, and direct laser acceleration of the bunch electrons by the laser pulse. Remarkably, the amount of energy transferred from the laser pulse to the plasma can be increased by several times by the guiding electron bunch despite its small energy content. Implications of such synergistic interactions for the high-gradient acceleration of externally injected witness charges are discussed, and a new concept of a Laser-pulse and Electron-bunch Plasma Accelerator (LEPA) is formulated.

[ { "created": "Tue, 23 Jun 2020 00:38:36 GMT", "version": "v1" }, { "created": "Wed, 15 Jul 2020 17:51:47 GMT", "version": "v2" }, { "created": "Sun, 26 Jul 2020 16:14:45 GMT", "version": "v3" }, { "created": "Thu, 5 Nov 2020 15:20:29 GMT", "version": "v4" } ]

2020-12-30

[ [ "Wang", "Tianhong", "" ], [ "Khudik", "Vladimir", "" ], [ "Shvets", "Gennady", "" ] ]

Tue, 23 Jun 2020 00:38:36 GMT

2020-06-23T00:38:36Z

[ "physics" ]

arxiv

null

{CUNI} Systems for the Unsupervised News Translation Task in {WMT} 2019

null

In this paper we describe the CUNI translation system used for the unsupervised news shared task of the ACL 2019 Fourth Conference on Machine Translation (WMT19). We follow the strategy of Artetxe ae at. (2018b), creating a seed phrase-based system where the phrase table is initialized from cross-lingual embedding mappings trained on monolingual data, followed by a neural machine translation system trained on synthetic parallel data. The synthetic corpus was produced from a monolingual corpus by a tuned PBMT model refined through iterative back-translation. We further focus on the handling of named entities, i.e. the part of vocabulary where the cross-lingual embedding mapping suffers most. Our system reaches a BLEU score of 15.3 on the German-Czech WMT19 shared task.

null

acl

inproceedings

kvapilikova-etal-2019-cuni

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 Martins, Andr{\'e} and Monz, Christof and Negri, Matteo and N{\'e}v{\'e}ol, Aur{\'e}lie and Neves, Mariana and Post, Matt and Turchi, Marco and Verspoor, Karin

aug

2,019

Florence, Italy

Association for Computational Linguistics

https://aclanthology.org/W19-5323/

Kvapil{\'i}kov{\'a}, Ivana and Mach{\'a}{\v{c}}ek, Dominik and Bojar, Ond{\v{r}}ej

Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1)

241--248

null

2209.08288

Aydogan Ozcan

Luzhe Huang, Hanlong Chen, Tairan Liu, Aydogan Ozcan

Self-supervised learning of hologram reconstruction using physics consistency

45 pages, 6 Figures

Nature Machine Intelligence (2023)

10.1038/s42256-023-00704-7

null

cs.CV cs.LG eess.IV physics.optics

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

The past decade has witnessed transformative applications of deep learning in various computational imaging, sensing and microscopy tasks. Due to the supervised learning schemes employed, these methods mostly depend on large-scale, diverse, and labeled training data. The acquisition and preparation of such training image datasets are often laborious and costly, also leading to biased estimation and limited generalization to new sample types. Here, we report a self-supervised learning model, termed GedankenNet, that eliminates the need for labeled or experimental training data, and demonstrate its effectiveness and superior generalization on hologram reconstruction tasks. Without prior knowledge about the sample types to be imaged, the self-supervised learning model was trained using a physics-consistency loss and artificial random images that are synthetically generated without any experiments or resemblance to real-world samples. After its self-supervised training, GedankenNet successfully generalized to experimental holograms of various unseen biological samples, reconstructing the phase and amplitude images of different types of objects using experimentally acquired test holograms. Without access to experimental data or knowledge of real samples of interest or their spatial features, GedankenNet's self-supervised learning achieved complex-valued image reconstructions that are consistent with the Maxwell's equations, and its output inference and object solutions accurately represent the wave propagation in free-space. GedankenNet framework also exhibits resilience to random, unknown perturbations in the physical forward model, including changes in the hologram distances, pixel size and illumination wavelength. This self-supervised learning of image reconstruction tasks creates new opportunities for various inverse problems in holography, microscopy and computational imaging fields.

[ { "created": "Sat, 17 Sep 2022 09:02:10 GMT", "version": "v1" }, { "created": "Sat, 17 Jun 2023 08:53:11 GMT", "version": "v2" } ]

2023-08-08

[ [ "Huang", "Luzhe", "" ], [ "Chen", "Hanlong", "" ], [ "Liu", "Tairan", "" ], [ "Ozcan", "Aydogan", "" ] ]

Sat, 17 Sep 2022 09:02:10 GMT

2022-09-17T09:02:10Z

[ "physics", "cs", "eess" ]

arxiv

null

0911.2849

Rongli Huang jane

RongLi Huang, ZhiZhang Wang

On the entire self-shrinking solutions to Lagrangian mean curvature flow

Title changed, introduction rewritten and Theorem 1.3 added

Calc. Var. (2011) 41:321--339

10.1007/s00526-010-0364-9

null

math.AP math.DG

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

The authors prove that the logarithmic Monge-Amp\`{e}re flow with uniformly bound and convex initial data satisfies uniform decay estimates away from time $t=0$. Then applying the decay estimates, we conclude that every entire classical strictly convex solution of the equation {equation*} \det D^{2}u=\exp\{n(-u+1/2\sum_{i=1}^{n}x_{i}\frac{\partial u}{\partial x_{i}})\}, {equation*} should be a quadratic polynomial if the inferior limit of the smallest eigenvalue of the function $|x|^{2}D^{2}u$ at infinity has an uniform positive lower bound larger than $2(1-1/n)$. Using a similar method, we can prove that every classical convex or concave solution of the equation {equation*} \sum_{i=1}^{n}\arctan\lambda_{i}=-u+1/2\sum_{i=1}^{n}x_{i}\frac{\partial u}{\partial x_{i}}. {equation*} must be a quadratic polynomial, where $\lambda_{i}$ are the eigenvalues of the Hessian $D^{2}u$.

[ { "created": "Sun, 15 Nov 2009 10:52:14 GMT", "version": "v1" }, { "created": "Fri, 27 Nov 2009 02:37:45 GMT", "version": "v2" }, { "created": "Mon, 25 Jan 2010 02:19:45 GMT", "version": "v3" }, { "created": "Tue, 16 Feb 2010 08:29:05 GMT", "version": "v4" }, { "created": "Sat, 20 Feb 2010 03:58:47 GMT", "version": "v5" }, { "created": "Fri, 20 May 2011 15:07:18 GMT", "version": "v6" }, { "created": "Thu, 15 Mar 2012 01:25:15 GMT", "version": "v7" } ]

2012-03-16

[ [ "Huang", "RongLi", "" ], [ "Wang", "ZhiZhang", "" ] ]

Sun, 15 Nov 2009 10:52:14 GMT

2009-11-15T10:52:14Z

[ "math" ]

arxiv

null

2205.00615

Manfred von Willich

Hoi-Kwong Lo, Mattia Montagna and Manfred von Willich

Distributed Symmetric Key Establishment: A scalable, quantum-proof key distribution system

Our protocol has been renamed Distributed Symmetric Key Establishment (DSKE). 11 pages, 6 figures

null

quant-ph cs.CR

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

We propose and implement a protocol for a scalable, cost-effective, information-theoretically secure key distribution and management system. The system, called Distributed Symmetric Key Establishment (DSKE), relies on pre-shared random numbers between DSKE clients and a group of Security Hubs. Any group of DSKE clients can use the DSKE protocol to distill from the pre-shared numbers a secret key. The clients are protected from Security Hub compromise via a secret sharing scheme that allows the creation of the final key without the need to trust individual Security Hubs. Precisely, if the number of compromised Security Hubs does not exceed a certain threshold, confidentiality is guaranteed to DSKE clients and, at the same time, robustness against denial-of-service (DoS) attacks. The DSKE system can be used for quantum-secure communication, can be easily integrated into existing network infrastructures, and can support arbitrary groups of communication parties that have access to a key. We discuss the high-level protocol, analyze its security, including its robustness against disruption. A proof-of-principle demonstration of secure communication between two distant clients with a DSKE-based VPN using Security Hubs on Amazon Web Server (AWS) nodes thousands of kilometres away from them was performed, demonstrating the feasibility of DSKE-enabled secret sharing one-time-pad encryption with a data rate above 50 Mbit/s and a latency below 70 ms.

[ { "created": "Mon, 2 May 2022 01:46:11 GMT", "version": "v1" }, { "created": "Tue, 3 May 2022 02:26:44 GMT", "version": "v2" }, { "created": "Fri, 25 Nov 2022 03:16:39 GMT", "version": "v3" }, { "created": "Tue, 30 Jul 2024 01:21:58 GMT", "version": "v4" } ]

2024-07-31

[ [ "Lo", "Hoi-Kwong", "" ], [ "Montagna", "Mattia", "" ], [ "von Willich", "Manfred", "" ] ]

Mon, 2 May 2022 01:46:11 GMT

2022-05-02T01:46:11Z

[ "physics", "cs" ]

arxiv

null

1007.1828

Yuri Shtemler

V.M. Chernyavski, Y. M. Shtemler, E. Golbraikh, M. Mond

Generation of intermediately-long sea waves by weakly sheared winds

This paper has been withdrawn by the author due to misprinting errors

null

physics.ao-ph

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

The present study concerns the numerical modeling of sea-wave instability under the effect of logarithmic-wind profile in hurricane conditions. The central point of the study is the calculation of the wave growth rate, which is proportional to the fractional input energy from the weakly-sheared (logarithmic) wind to the wave exponentially varying with time. It is shown for hurricane conditions that the Miles-type stability model based on the Charnock's formula with the standard constant coefficient underestimates the growth rate ~5 to 50 times as compared with the model employing the roughness adopted from experimental data for hurricane winds. The drag reduction with wind speed at hurricane conditions coupled with the similar behavior of the dimensionless gravity acceleration, leads to the minimum in the maximal growth rate and the maximum in the most unstable wavelength.

[ { "created": "Mon, 12 Jul 2010 06:35:51 GMT", "version": "v1" }, { "created": "Tue, 13 Jul 2010 03:32:45 GMT", "version": "v2" }, { "created": "Sun, 19 Dec 2010 18:37:55 GMT", "version": "v3" }, { "created": "Fri, 11 Mar 2011 20:15:40 GMT", "version": "v4" } ]

2011-03-14

[ [ "Chernyavski", "V. M.", "" ], [ "Shtemler", "Y. M.", "" ], [ "Golbraikh", "E.", "" ], [ "Mond", "M.", "" ] ]

Mon, 12 Jul 2010 06:35:51 GMT

2010-07-12T06:35:51Z

[ "physics" ]

arxiv

null

1906.12053

Louxin Zhang

Generating Normal Networks via Leaf Insertion and Nearest Neighbor Interchange

4 figures and 13 pages

RECOMBCG, 2019

null

q-bio.PE cs.DS math.CO

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

Galled trees are studied as a recombination model in theoretic population genetics. This class of phylogenetic networks has been generalized to tree-child networks, normal networks and tree-based networks by relaxing a structural condition. Although these networks are simple, their topological structures have yet to be fully understood. It is well-known that all phylogenetic trees on $n$ taxa can be generated by the insertion of the $n$-th taxa to each edge of all the phylogenetic trees on $n-1$ taxa. We prove that all tree-child networks with $k$ reticulate nodes on $n$ taxa can be uniquely generated via three operations from all the tree-child networks with $k-1$ or $k$ reticulate nodes on $n-1$ taxa . An application of this result is found in counting tree-child networks and normal networks. In particular, a simple formula is given for the number of rooted phylogenetic networks with one reticulate node.

[ { "created": "Fri, 28 Jun 2019 06:13:56 GMT", "version": "v1" }, { "created": "Mon, 1 Jul 2019 01:20:53 GMT", "version": "v2" }, { "created": "Fri, 2 Aug 2019 12:00:28 GMT", "version": "v3" } ]

2019-08-05

[ [ "Zhang", "Louxin", "" ] ]

Fri, 28 Jun 2019 06:13:56 GMT

2019-06-28T06:13:56Z

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

arxiv

null

1503.00890

C\'ecile Proust-Lima

C\'ecile Proust-Lima, Viviane Philipps, Benoit Liquet

Estimation of extended mixed models using latent classes and latent processes: the R package lcmm

null

Journal of Statistical Software (2017), 78(2), 1-56

10.18637/jss.v078.i02

null

stat.CO stat.ME

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

The R package lcmm provides a series of functions to estimate statistical models based on linear mixed model theory. It includes the estimation of mixed models and latent class mixed models for Gaussian longitudinal outcomes (hlme), curvilinear and ordinal univariate longitudinal outcomes (lcmm) and curvilinear multivariate outcomes (multlcmm), as well as joint latent class mixed models (Jointlcmm) for a (Gaussian or curvilinear) longitudinal outcome and a time-to-event that can be possibly left-truncated right-censored and defined in a competing setting. Maximum likelihood esimators are obtained using a modified Marquardt algorithm with strict convergence criteria based on the parameters and likelihood stability, and on the negativity of the second derivatives. The package also provides various post-fit functions including goodness-of-fit analyses, classification, plots, predicted trajectories, individual dynamic prediction of the event and predictive accuracy assessment. This paper constitutes a companion paper to the package by introducing each family of models, the estimation technique, some implementation details and giving examples through a dataset on cognitive aging.

[ { "created": "Tue, 3 Mar 2015 10:47:08 GMT", "version": "v1" }, { "created": "Sun, 24 Jan 2016 13:01:19 GMT", "version": "v2" } ]

2017-08-24

[ [ "Proust-Lima", "Cécile", "" ], [ "Philipps", "Viviane", "" ], [ "Liquet", "Benoit", "" ] ]

Tue, 3 Mar 2015 10:47:08 GMT

2015-03-03T10:47:08Z

[ "stat" ]

arxiv

null

{S}ci{REL} at {S}em{E}val-2018 Task 7: A System for Semantic Relation Extraction and Classification

null

This paper describes our system, SciREL (Scientific abstract RELation extraction system), developed for the SemEval 2018 Task 7: Semantic Relation Extraction and Classification in Scientific Papers. We present a feature-vector based system to extract explicit semantic relation and classify them. Our system is trained in the ACL corpus (BIrd et al., 2008) that contains annotated abstracts given by the task organizers. When an abstract with annotated entities is given as the input into our system, it extracts the semantic relations through a set of defined features and classifies them into one of the given six categories of relations through feature engineering and a learned model. For the best combination of features, our system SciREL obtained an F-measure of 20.03 on the official test corpus which includes 150 abstracts in the relation classification Subtask 1.1. In this paper, we provide an in-depth error analysis of our results to prevent duplication of research efforts in the development of future systems

null

acl

inproceedings

mahendran-etal-2018-scirel

Apidianaki, Marianna and Mohammad, Saif M. and May, Jonathan and Shutova, Ekaterina and Bethard, Steven and Carpuat, Marine

jun

2,018

New Orleans, Louisiana

Association for Computational Linguistics

https://aclanthology.org/S18-1137/

Mahendran, Darshini and Brahmana, Chathurika and McInnes, Bridget

Proceedings of the 12th International Workshop on Semantic Evaluation

853--857

null

1603.01821

Kristian Uldall Kristiansen

Blowup for flat slow manifolds

Shortened title and updated references. To appear in Nonlinearity

null

math.DS

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

In this paper we present a method for extending the blowup method, in the formulation of Krupa and Szmolyan, to flat slow manifolds that lose hyperbolicity beyond any algebraic order. Although these manifolds have infinite co-dimension, they do appear naturally in certain settings. For example in (a) the regularization of piecewise smooth systems by $\tanh$, (b) a model of aircraft landing dynamics, and finally (c) in a model of earthquake faulting. We demonstrate the approach on a simple model system and the examples (a) and (b).

[ { "created": "Sun, 6 Mar 2016 12:47:48 GMT", "version": "v1" }, { "created": "Tue, 8 Mar 2016 05:30:28 GMT", "version": "v2" }, { "created": "Tue, 15 Mar 2016 20:23:15 GMT", "version": "v3" }, { "created": "Wed, 16 Nov 2016 09:43:33 GMT", "version": "v4" }, { "created": "Mon, 27 Mar 2017 11:24:36 GMT", "version": "v5" } ]

2017-03-28

[ [ "Kristiansen", "Kristian Uldall", "" ] ]

Sun, 6 Mar 2016 12:47:48 GMT

2016-03-06T12:47:48Z

[ "math" ]

arxiv

null

Challenges in adapting an interlingua for bidirectional {E}nglish-{I}talian translation

null

We describe our experience in adapting an existing high- quality, interlingual, unidirectional machine translation system to a new domain and bidirectional translation for a new language pair (English and Italian). We focus on the interlingua design changes which were necessary to achieve high quality output in view of the language mismatches between English and Italian. The representation we propose contains features that are interpreted differently, depending on the translation direction. This decision simplified the process of creating the interlingua for individual sentences, and allows the system to defer mapping of language-specific features (such as tense and aspect), which are realized when the target syntactic feature structure is created. We also describe a set of problems we encountered in translating modal verbs, and discuss the representation of modality in our interlingua.

null

acl

inproceedings

cavalli-sforza-etal-2000-challenges

White, John S.

oct # " 10-14"

2,000

Cuernavaca, Mexico

Springer

https://aclanthology.org/2000.amta-papers.17/

Cavalli-Sforza, Violetta and Czuba, Krzysztof and Mitamura, Teruko and Nyberg, Eric

Proceedings of the Fourth Conference of the Association for Machine Translation in the Americas: Technical Papers

169--178

null

1407.0665

J\'er\'emie Bouttier

J\'er\'emie Bouttier, Guillaume Chapuy and Sylvie Corteel

From Aztec diamonds to pyramids: steep tilings

36 pages, 22 figures (v3: final accepted version with new Figure 6, new improved proof of Proposition 11)

Trans. Amer. Math. Soc. 369 (2017), 5921-5959

10.1090/tran/7169

IPhT-t14/095

math.CO cond-mat.stat-mech math.PR

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

We introduce a family of domino tilings that includes tilings of the Aztec diamond and pyramid partitions as special cases. These tilings live in a strip of $\mathbb{Z}^2$ of the form $1 \leq x-y \leq 2\ell$ for some integer $\ell \geq 1$, and are parametrized by a binary word $w\in\{+,-\}^{2\ell}$ that encodes some periodicity conditions at infinity. Aztec diamond and pyramid partitions correspond respectively to $w=(+-)^\ell$ and to the limit case $w=+^\infty-^\infty$. For each word $w$ and for different types of boundary conditions, we obtain a nice product formula for the generating function of the associated tilings with respect to the number of flips, that admits a natural multivariate generalization. The main tools are a bijective correspondence with sequences of interlaced partitions and the vertex operator formalism (which we slightly extend in order to handle Littlewood-type identities). In probabilistic terms our tilings map to Schur processes of different types (standard, Pfaffian and periodic). We also introduce a more general model that interpolates between domino tilings and plane partitions.

[ { "created": "Wed, 2 Jul 2014 18:05:29 GMT", "version": "v1" }, { "created": "Tue, 15 Jul 2014 12:30:48 GMT", "version": "v2" }, { "created": "Thu, 20 Apr 2017 08:13:46 GMT", "version": "v3" } ]

2017-09-11

[ [ "Bouttier", "Jérémie", "" ], [ "Chapuy", "Guillaume", "" ], [ "Corteel", "Sylvie", "" ] ]

Wed, 2 Jul 2014 18:05:29 GMT

2014-07-02T18:05:29Z

[ "physics", "math" ]

arxiv

null

Transfer-Based Learning-to-Rank Assessment of Medical Term Technicality

null

While measuring the readability of texts has been a long-standing research topic, assessing the technicality of terms has only been addressed more recently and mostly for the English language. In this paper, we train a learning-to-rank model to determine a specialization degree for each term found in a given list. Since no training data for this task exist for French, we train our system with non-lexical features on English data, namely, the Consumer Health Vocabulary, then apply it to French. The features include the likelihood ratio of the term based on specialized and lay language models, and tests for containing morphologically complex words. The evaluation of this approach is conducted on 134 terms from the UMLS Metathesaurus and 868 terms from the Eugloss thesaurus. The Normalized Discounted Cumulative Gain obtained by our system is over 0.8 on both test sets. Besides, thanks to the learning-to-rank approach, adding morphological features to the language model features improves the results on the Eugloss thesaurus.

null

acl

inproceedings

bouamor-etal-2016-transfer

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/L16-1366/

Bouamor, Dhouha and Llanos, Leonardo Campillos and Ligozat, Anne-Laure and Rosset, Sophie and Zweigenbaum, Pierre

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

2312--2316

null

math/0104105

Nguyen Anh Ky

Irreducible representations of Upq[gl(2/2)]

Latex, 8 pages. A reference added in v.2

J.Phys.A34:7881-7888,2001

10.1088/0305-4470/34/38/312

null

math.QA hep-th math-ph math.MP math.RT nlin.SI quant-ph

null

The two-parametric quantum superalgebra $U_{pq}[gl(2/2)]$ and its representations are considered. All finite-dimensional irreducible representations of this quantum superalgebra can be constructed and classified into typical and nontypical ones according to a proposition proved in the present paper. This proposition is a nontrivial deformation from the one for the classical superalgebra gl(2/2), unlike the case of one-parametric deformations.

[ { "created": "Tue, 10 Apr 2001 03:17:04 GMT", "version": "v1" }, { "created": "Thu, 12 Apr 2001 11:50:01 GMT", "version": "v2" } ]

2008-11-26

[ [ "Ky", "Nguyen Anh", "" ] ]

Tue, 10 Apr 2001 03:17:04 GMT

2001-04-10T03:17:04Z

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

arxiv

null

2001.05651

Dezhao Wang

Dezhao Wang, Sifeng Xia, Wenhan Yang, and Jiaying Liu

Combining Progressive Rethinking and Collaborative Learning: A Deep Framework for In-Loop Filtering

Accepted for publication in IEEE Transactions on Image Processing (TIP). Website available at https://dezhao-wang.github.io/PRN-v2/

null

10.1109/TIP.2021.3068638

null

eess.IV cs.CV

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

In this paper, we aim to address issues of (1) joint spatial-temporal modeling and (2) side information injection for deep-learning based in-loop filter. For (1), we design a deep network with both progressive rethinking and collaborative learning mechanisms to improve quality of the reconstructed intra-frames and inter-frames, respectively. For intra coding, a Progressive Rethinking Network (PRN) is designed to simulate the human decision mechanism for effective spatial modeling. Our designed block introduces an additional inter-block connection to bypass a high-dimensional informative feature before the bottleneck module across blocks to review the complete past memorized experiences and rethinks progressively. For inter coding, the current reconstructed frame interacts with reference frames (peak quality frame and the nearest adjacent frame) collaboratively at the feature level. For (2), we extract both intra-frame and inter-frame side information for better context modeling. A coarse-to-fine partition map based on HEVC partition trees is built as the intra-frame side information. Furthermore, the warped features of the reference frames are offered as the inter-frame side information. Our PRN with intra-frame side information provides 9.0% BD-rate reduction on average compared to HEVC baseline under All-intra (AI) configuration. While under Low-Delay B (LDB), Low-Delay P (LDP) and Random Access (RA) configuration, our PRN with inter-frame side information provides 9.0%, 10.6% and 8.0% BD-rate reduction on average respectively. Our project webpage is https://dezhao-wang.github.io/PRN-v2/.

[ { "created": "Thu, 16 Jan 2020 05:14:34 GMT", "version": "v1" }, { "created": "Tue, 21 Jan 2020 06:47:36 GMT", "version": "v2" }, { "created": "Wed, 31 Mar 2021 09:33:27 GMT", "version": "v3" } ]

2021-04-28

[ [ "Wang", "Dezhao", "" ], [ "Xia", "Sifeng", "" ], [ "Yang", "Wenhan", "" ], [ "Liu", "Jiaying", "" ] ]

Thu, 16 Jan 2020 05:14:34 GMT

2020-01-16T05:14:34Z

[ "cs", "eess" ]

arxiv

null

cond-mat/9905076

Johannes Berg

Johannes Berg and Martin Weigt

Entropy and typical properties of Nash equilibria in two-player games

6 pages, was "Self averaging of Nash equilibria in two player games", main section rewritten, some new results, for additional information see http://itp.nat.uni-magdeburg.de/~jberg/games.html

Europhys. Lett.,48 (2), 129-135 (1999)

10.1209/epl/i1999-00456-2

null

cond-mat.dis-nn adap-org cond-mat.stat-mech nlin.AO

null

We use techniques from the statistical mechanics of disordered systems to analyse the properties of Nash equilibria of bimatrix games with large random payoff matrices. By means of an annealed bound, we calculate their number and analyse the properties of typical Nash equilibria, which are exponentially dominant in number. We find that a randomly chosen equilibrium realizes almost always equal payoffs to either player. This value and the fraction of strategies played at an equilibrium point are calculated as a function of the correlation between the two payoff matrices. The picture is complemented by the calculation of the properties of Nash equilibria in pure strategies.

[ { "created": "Thu, 6 May 1999 13:42:40 GMT", "version": "v1" }, { "created": "Wed, 21 Jul 1999 15:36:12 GMT", "version": "v2" } ]

2009-10-31

[ [ "Berg", "Johannes", "" ], [ "Weigt", "Martin", "" ] ]

Thu, 6 May 1999 13:42:40 GMT

1999-05-06T13:42:40Z

[ "physics", "nlin" ]

arxiv

null

2207.09971

Mathias Schreiner

Mathias Schreiner, Arghya Bhowmik, Tejs Vegge, Peter Bj{\o}rn J{\o}rgensen and Ole Winther

NeuralNEB -- Neural Networks can find Reaction Paths Fast

null

physics.comp-ph cs.LG physics.chem-ph

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

Quantum mechanical methods like Density Functional Theory (DFT) are used with great success alongside efficient search algorithms for studying kinetics of reactive systems. However, DFT is prohibitively expensive for large scale exploration. Machine Learning (ML) models have turned out to be excellent emulators of small molecule DFT calculations and could possibly replace DFT in such tasks. For kinetics, success relies primarily on the models capability to accurately predict the Potential Energy Surface (PES) around transition-states and Minimal Energy Paths (MEPs). Previously this has not been possible due to scarcity of relevant data in the literature. In this paper we train state of the art equivariant Graph Neural Network (GNN)-based models on around 10.000 elementary reactions from the Transition1x dataset. We apply the models as potentials for the Nudged Elastic Band (NEB) algorithm and achieve a Mean Average Error (MAE) of 0.13+/-0.03 eV on barrier energies on unseen reactions. We compare the results against equivalent models trained on QM9 and ANI1x. We also compare with and outperform Density Functional based Tight Binding (DFTB) on both accuracy and computational resource. The implication is that ML models, given relevant data, are now at a level where they can be applied for downstream tasks in quantum chemistry transcending prediction of simple molecular features.

[ { "created": "Wed, 20 Jul 2022 15:29:45 GMT", "version": "v1" }, { "created": "Thu, 21 Jul 2022 08:47:54 GMT", "version": "v2" }, { "created": "Thu, 1 Sep 2022 15:08:14 GMT", "version": "v3" } ]

2022-09-02

[ [ "Schreiner", "Mathias", "" ], [ "Bhowmik", "Arghya", "" ], [ "Vegge", "Tejs", "" ], [ "Jørgensen", "Peter Bjørn", "" ], [ "Winther", "Ole", "" ] ]

Wed, 20 Jul 2022 15:29:45 GMT

2022-07-20T15:29:45Z

[ "physics", "cs" ]

arxiv

null

Field Assessment of Critical Stroke by Emergency Services for Acute Delivery to a Comprehensive Stroke Center: FACE

null

Patients with acute ischemic stroke (AIS) due to large vessel occlusion (LVO) should be triaged to an endovascular-capable hospital by the emergency medical service (EMS). We designed a prehospital LVO prediction scale based on EMS assessments. In the derivation cohort, 1157 patients transferred to our hospital by the EMS because of suspected stroke within 24 h of onset were retrospectively examined. Factors associated with AIS due to LVO were identified based on the EMS assessment, and a prehospital scale identifying LVO was developed. The accuracy of this scale was validated in 502 consecutive patients who were transferred to 4 stroke centers, and its accuracy was compared with those of 4 previously reported scales. AIS due to LVO was diagnosed in 149 of 1157 patients (13%) in the derivation cohort. One point each was assigned for facial palsy, arm weakness, consciousness impairment (cannot say his/her name), atrial fibrillation, and diastolic blood pressure ≤ 85 mmHg, with two points for conjugate eye deviation (FACE

null

pubmed

null

2,020

null

31,833,034

Translational stroke research

1905.00594

Hans Hallen

Ziad Ali, Alexandra Duel-Hallen and Hans Hallen

Early Warning of mmWave Signal Blockage and AoA Transition Using sub-6 GHz Observations

in IEEE Communications Letters

null

10.1109/LCOMM.2019.2952602

null

eess.SP

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

The susceptibility of millimeter-wave (mmWave) signals to physical blockage and abrupt signal strength variations presents a challenge to reliable 5G communication. This work proposes and examines the feasibility of utilizing lower-frequency signals as early-warning indicators of mobile mmWave signal blockage or recovery. A physics-based channel simulation tool incorporating Fresnel diffraction and image sources is employed to demonstrate that sub-6 GHz signals "lead" mmWave signals in reaching a specific signal-strength threshold by several to tens of milliseconds at mobile speeds, suggesting early-warning systems are viable. This predictive approach stems from frequency-dependent properties of diffraction and does not assume a specific topology or mobile and obstacle speeds. Realistic simulations that include transitions from line of sight (LoS) to non-line of sight (NLoS) and reflection scenarios are employed to verify the proposed prediction capabilities. Moreover, prediction of the strongest multipath component and its angle of arrival (AoA) using sub-6 GHz observations is investigated.

[ { "created": "Thu, 2 May 2019 07:07:27 GMT", "version": "v1" }, { "created": "Wed, 2 Oct 2019 19:54:26 GMT", "version": "v2" }, { "created": "Wed, 13 Nov 2019 02:57:30 GMT", "version": "v3" } ]

2019-11-14

[ [ "Ali", "Ziad", "" ], [ "Duel-Hallen", "Alexandra", "" ], [ "Hallen", "Hans", "" ] ]

Thu, 2 May 2019 07:07:27 GMT

2019-05-02T07:07:27Z

[ "eess" ]

arxiv

null

1707.01838

Vincent Briane

Vincent Briane, Charles Kervrann and Myriam Vimond

A statistical analysis of particle trajectories in living cells

Revised introduction. A clearer and shorter description of the model (section 2)

Phys. Rev. E 97, 062121 (2018)

10.1103/PhysRevE.97.062121

null

stat.AP physics.bio-ph

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

Recent advances in molecular biology and fluorescence microscopy imaging have made possible the inference of the dynamics of single molecules in living cells. Such inference allows to determine the organization and function of the cell. The trajectories of particles in the cells, computed with tracking algorithms, can be modelled with diffusion processes. Three types of diffusion are considered : (i) free diffusion; (ii) subdiffusion or (iii) superdiffusion. The Mean Square Displacement (MSD) is generally used to determine the different types of dynamics of the particles in living cells (Qian, Sheetz and Elson 1991). We propose here a non-parametric three-decision test as an alternative to the MSD method. The rejection of the null hypothesis -- free diffusion -- is accompanied by claims of the direction of the alternative (subdiffusion or a superdiffusion). We study the asymptotic behaviour of the test statistic under the null hypothesis, and under parametric alternatives which are currently considered in the biophysics literature, (Monnier et al,2012) for example. In addition, we adapt the procedure of Benjamini and Hochberg (2000) to fit with the three-decision test setting, in order to apply the test procedure to a collection of independent trajectories. The performance of our procedure is much better than the MSD method as confirmed by Monte Carlo experiments. The method is demonstrated on real data sets corresponding to protein dynamics observed in fluorescence microscopy.

[ { "created": "Thu, 6 Jul 2017 15:44:26 GMT", "version": "v1" }, { "created": "Wed, 13 Sep 2017 17:57:01 GMT", "version": "v2" } ]

2018-06-20

[ [ "Briane", "Vincent", "" ], [ "Kervrann", "Charles", "" ], [ "Vimond", "Myriam", "" ] ]

Thu, 6 Jul 2017 15:44:26 GMT

2017-07-06T15:44:26Z

[ "physics", "stat" ]

arxiv

null