检索结果-内蒙古大学图书馆

Promoting diversity for end-to-end conversation response generation

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Ruan, Yu-Ping Ling, Zhen-Hua Liu, Quan Gu, Jia-Chen Zhu, Xiaodan National Engineering Laboratory for Speech and Language Information Processing University of Science and Technology of China Hefei China iFLYTEK Research Hefei China Department of Electrical and Computer Engineering Queen's University Kingston Canada

We present our work on Track 2 in the Dialog System Technology Challenges 7 (DSTC7). The DSTC7-Track 2 aims to evaluate the response generation of fully data-driven conversation models in knowledge-grounded settings, which provides the contextual-relevant factual texts. The Sequence-to-Sequence models have been widely used for end-to-end generative conversation modelling and achieved impressive results. However, they tend to output dull and repeated responses in previous studies. Our work aims to promote the diversity for end-to-end conversation response generation, which follows a two-stage pipeline: 1) Generate multiple responses. At this stage, two different models are proposed, i.e., a variational generative (VariGen) model and a retrieval based (Retrieval) model. 2) Rank and return the most related response by training a topic coherence discrimination (TCD) model for the ranking process. According to the official evaluation results, our proposed Retrieval and VariGen systems ranked first and second respectively on objective diversity metrics, i.e., Entropy, among all participant systems. And the VariGen system ranked second on NIST and METEOR metrics. Copyright © 2019, The Authors. All rights reserved.

关键词：

Exploring unsupervised pretraining and sentence structure modelling for winograd schema challenge

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Ruan, Yu-Ping Zhu, Xiaodan Ling, Zhen-Hua Shi, Zhan Liu, Quan Wei, Si National Engineering Laboratory for Speech and Language Information Processing University of Science and Technology of China Hefei China Department of Electrical and Computer Engineering Queen's University KingstonON Canada iFLYTEK Research Hefei China

Winograd Schema Challenge (WSC) was proposed as an AI-hard problem in testing computers' intelligence on common sense representation and reasoning. This paper presents the new state-of-theart on WSC, achieving an accuracy of 71.1%. We demonstrate that the leading performance benefits from jointly modelling sentence structures, utilizing knowledge learned from cutting-edge pretraining models, and performing fine-tuning. We conduct detailed analyses, showing that fine-tuning is critical for achieving the performance, but it helps more on the simpler associative problems. Modelling sentence dependency structures, however, consistently helps on the harder non-associative subset of WSC. Analysis also shows that larger fine-tuning datasets yield better performances, suggesting the potential benefit of future work on annotating more Winograd schema sentences. Copyright © 2019, The Authors. All rights reserved.

关键词： Machine learning

Using ASR Methods for OCR

学校读者我要写书评

暂无评论

Using ASR Methods for OCR

International Conference on Document Analysis and Recognition

作者： Ashish Arora Chun Chieh Chang Babak Rekabdar Bagher BabaAli Daniel Povey David Etter Desh Raj Hossein Hadian Jan Trmal Paola Garcia Shinji Watanabe Vimal Manohar Yiwen Shao Sanjeev Khudanpur Center for Language and Speech Processing Johns Hopkins University Baltimore USA Human Language Technology Center of Excellence Johns Hopkins University Baltimore USA School of Mathematics Statistics and Computer Sciences University of Tehran Iran Department of Computer Engineering Sharif University of Technology Iran

Hybrid deep neural network hidden Markov models (DNN-HMM) have achieved impressive results on large vocabulary continuous speech recognition (LVCSR) tasks. However, the recent approaches using DNN-HMM models are not explored much for text recognition. Inspired by the current work in automatic speech recognition (ASR) and machine translation, we present an open vocabulary sub-word text recognition system. The sub-word lexicon and sub-word language model (LM) helps in overcoming the challenge of recognizing out of vocabulary (OOV) words, and a time delay neural network (TDNN) and convolution neural network (CNN) based DNN-HMM optical model (OM) efficiently models the sequence dependency in the line image. We present results on 12 datasets with training data varying from 6k lines to 600k lines. The system is built for 8 languages, i.e., English, French, Arabic, Chinese, Farsi, Tamil, Russian, and Korean. We report competitive results on several commonly used handwritten and printed text datasets.

关键词： Hidden Markov models Text recognition Vocabulary Training Optical imaging Training data Feature extraction

JOINT VERIFICATION-IDENTIFICATION IN END-TO-END MULTI-SCALE CNN FRAMEWORK FOR TOPIC IDENTIFICATION

学校读者我要写书评

暂无评论

JOINT VERIFICATION-IDENTIFICATION IN END-TO-END MULTI-SCALE ...

IEEE International Conference on Acoustics, speech and Signal processing

作者： Raghavendra Pappagari Jesús Villalba Najim Dehak Center for Language and Speech Processing Department of Electrical and Computer Engineering The Johns Hopkins University USA

ISBN: (纸本)9781538646595

We present an end-to-end multi-scale Convolutional Neural Network (CNN) framework for topic identification (topic ID). In this work, we examined multi-scale CNN for classification using raw text input. Topical word embeddings are learnt at multiple scales using parallel convolutional layers. A technique to integrate verification and identification objectives is examined to improve topic ID performance. With this approach, we achieved significant improvement in identification task. We evaluated our framework on two contrasting datasets: 20 newsgroups and Fisher. We obtained 92.93% accuracy on Fisher and 86.12% on 20 newsgroups, which to our knowledge are the best published results on these datasets at the moment.

关键词： BOW raw text CNN verification identification topic id end-to-end BOW end-to-end Verification Martes pennanti newsgroups multi-scale Joints Dataset

End-to-end speech recognition using lattice-free MMI 19

学校读者我要写书评

暂无评论

End-to-end speech recognition using lattice-free MMI

19th Annual Conference of the International speech Communication, INTERspeech 2018

作者： Hadian, Hossein Sameti, Hossein Povey, Daniel Khudanpur, Sanjeev Department of Computer Engineering Sharif University of Technology Tehran Iran Center for Language and Speech Processing Johns Hopkins University BaltimoreMD United States Human Language Technology Center of Excellence Johns Hopkins University Baltimore United States

We present our work on end-to-end training of acoustic models using the lattice-free maximum mutual information (LF-MMI) objective function in the context of hidden Markov models. By end-to-end training, we mean flat-start training of a single DNN in one stage without using any previously trained models, forced alignments, or building state-tying decision trees. We use full biphones to enable context-dependent modeling without trees, and show that our end-to-end LF-MMI approach can achieve comparable results to regular LF-MMI on well-known large vocabulary tasks. We also compare with other end-to-end methods such as CTC in character-based and lexicon-free settings and show 5 to 25 percent relative reduction in word error rates on different large vocabulary tasks while using significantly smaller models. © 2018 International speech Communication Association. All rights reserved.

关键词： Hidden Markov models

Trends in Silicates in the β Pictoris Disk

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lu, Cicero X. Chen, Christine H. Sargent, B.A. Watson, Dan M. Lisse, Carey M. Green, Joel D. Sitko, Michael L. Mittal, Tushar Lebouteiller, V. Sloan, G.C. Rebollido, Isabel Hines, Dean C. Girard, Julien H. Werner, Michael W. Stapelfeldt, Karl R. Wu, Winston Worthen, Kadin Department of Physics and Astronomy The Johns Hopkins University 3400 N. Charles Street BaltimoreMD21218 United States Space Telescope Science Institute 3700 San Martin Dr. BaltimoreMD21218 United States Department of Physics and Astronomy University of Rochester 500 Wilson Blvd RochesterNY14627 United States Johns Hopkins University Applied Physics Laboratory 11100 Johns Hopkins Rd LaurelMD20723 United States Department of Physics University of Cincinnati CincinnatiOH45221 United States Space Science Institute BoulderCO80301 United States Mit Department of Earth Atmospheric and Planetary Sciences Green Bldg The 77 Massachusetts Ave CambridgeMA02139 United States Aim Cea Cnrs Université Paris-Saclay Université Paris Diderot Sorbonne Paris Cité Gif-sur-YvetteF-91191 France Department of Physics and Astronomy University of North Carolina Chapel HillNC27599-3255 United States Jet Propulsion Laboratory California Institute of Technology 4800 Oak Grove Drive PasadenaCA91107 United States Center for Language and Speech Processing Department of Computer Science Johns Hopkins University BaltimoreMD21218 United States Computer Science and Engineering University of Michigan Ann ArborMI48109 United States

While β Pic is known to host silicates in ring-like structures, whether the properties of these silicate dust vary with stellocentric distance remains an open question. We re-analyze the β Pictoris debris disk spectrum from the Spitzer Infrared Spectrograph (IRS) and a new IRTF/SpeX spectrum to investigate trends in Fe/Mg ratio, shape, and crystallinity in grains as a function of wavelength, a proxy for stellocentric distance. By analyzing a recalibrated and re-extracted spectrum, we identify a new 18 μm forsterite emission feature and recover a 23 μm forsterite emission feature with a substantially larger line-to-continuum ratio than previously reported. We find that these prominent spectral features are primarily produced by small submicron-sized grains, which are continuously generated and replenished from planetesimal collisions in the disk and can elucidate their parent bodies' composition. We discover three trends about these small grains: as stellocentric distance increases, (1) small silicate grains become more crystalline (less amorphous), (2) they become more irregular in shape, and (3) for crystalline silicate grains, the Fe/Mg ratio decreases. Applying these trends to β Pic's planetary architecture, we find that the dust population exterior to the orbits of β Pic b and c differs substantially in crystallinity and shape. We also find a tentative 3-5 μm dust excess due to spatially unresolved hot dust emission close to the star. From our findings, we infer that the surfaces of large planetesimals are more Fe-rich and collisionally-processed closer to the star but more Fe-poor and primordial farther from the star. © 2022, CC BY.

关键词： Silicates

SDN controllers: Benchmarking & performance evaluation

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Zhu, Liehuang Karim, Md Monjurul Sharif, Kashif Li, Fan Du, Xiaojiang Guizani, Mohsen Beijing Engineering Research Center for Massive Language Information Processing & Cloud Computing Application School of Computer Science and Technology Beijing Institute of Technology Beijing China Department of Computer and Information Sciences Temple University Philadelphia United States Department of Electrical and Computer Engineering University of Idaho MoscowID United States

Software Defined Networks offer flexible and intelligent network operations by splitting a traditional network into a centralized control plane and a programmable data plane. The intelligent control plane is responsible for providing flow paths to switches and optimizes network performance. The controller in the control plane is the fundamental element used for all operations of data plane management. Hence, the performance and capabilities of the controller itself are extremely important. Furthermore, the tools used to benchmark their performance must be accurate and effective in measuring different evaluation parameters. There are dozens of controller proposals available in existing literature. However, there is no quantitative comparative analysis for them. In this article, we present a comprehensive qualitative comparison of different SDN controllers, along with a quantitative analysis of their performance in different network scenarios. More specifically, we categorize and classify 34 controllers based on their capabilities, and present a qualitative comparison of their properties. We also discuss in-depth capabilities of benchmarking tools used for SDN controllers, along with best practices for quantitative controller evaluation. This work uses three benchmarking tools to compare nine controllers against multiple criteria. Finally, we discuss detailed research findings on the performance, benchmarking criteria, and evaluation testbeds for SDN controllers. Copyright © 2019, The Authors. All rights reserved.

关键词： Benchmarking

Understanding metric-related pitfalls in image analysis validation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Reinke, Annika Tizabi, Minu D. Baumgartner, Michael Eisenmann, Matthias Heckmann-Nötzel, Doreen Kavur, A. Emre Rädsch, Tim Sudre, Carole H. Acion, Laura Antonelli, Michela Arbel, Tal Bakas, Spyridon Benis, Arriel Blaschko, Matthew B. Buettner, Florian Cardoso, M. Jorge Cheplygina, Veronika Chen, Jianxu Christodoulou, Evangelia Cimini, Beth A. Collins, Gary S. Farahani, Keyvan Ferrer, Luciana Galdran, Adrian van Ginneken, Bram Glocker, Ben Godau, Patrick Haase, Robert Hashimoto, Daniel A. Hoffman, Michael M. Huisman, Merel Isensee, Fabian Jannin, Pierre Kahn, Charles E. Kainmueller, Dagmar Kainz, Bernhard Karargyris, Alexandros Karthikesalingam, Alan Kenngott, Hannes Kleesiek, Jens Kofler, Florian Kooi, Thijs Kopp-Schneider, Annette Kozubek, Michal Kreshuk, Anna Kurc, Tahsin Landman, Bennett A. Litjens, Geert Madani, Amin Maier-Hein, Klaus Martel, Anne L. Mattson, Peter Meijering, Erik Menze, Bjoern Moons, Karel G.M. Müller, Henning Nichyporuk, Brennan Nickel, Felix Petersen, Jens Rafelski, Susanne M. Rajpoot, Nasir Reyes, Mauricio Riegler, Michael A. Rieke, Nicola Saez-Rodriguez, Julio Sánchez, Clara I. Shetty, Shravya Summers, Ronald M. Taha, Abdel A. Tiulpin, Aleksei Tsaftaris, Sotirios A. van Calster, Ben Varoquaux, Gaël Yaniv, Ziv R. Jäger, Paul F. Maier-Hein, Lena Faculty of Mathematics and Computer Science Heidelberg University Heidelberg Germany Heidelberg Division of Intelligent Medical Systems Germany NCT Heidelberg A Partnership Between DKFZ University Medical Center Heidelberg Germany Heidelberg Division of Medical Image Computing Germany Heidelberg Division of Intelligent Medical Systems Germany MRC Unit for Lifelong Health and Ageing UCL Centre for Medical Image Computing Department of Computer Science University College London London United Kingdom School of Biomedical Engineering and Imaging Science King’s College London London United Kingdom Instituto de Cálculo CONICET – Universidad de Buenos Aires Buenos Aires Argentina Centre for Medical Image Computing University College London London United Kingdom McGill University Montreal Canada Division of Computational Pathology Dept of Pathology & Laboratory Medicine Indiana University School of Medicine IU Health Information and Translational Sciences Building Indianapolis United States University of Pennsylvania Richards Medical Research Laboratories FL7 PhiladelphiaPA United States Department of Digital Medical Technologies Holon Institute of Technology Holon Israel European Federation for Medical Informatics Le Mont-sur-Lausanne Switzerland Center for Processing Speech and Images Department of Electrical Engineering KU Leuven Leuven Belgium partner site Frankfurt/Mainz a partnership between DKFZ and UCT Frankfurt Marburg Germany Heidelberg Germany Goethe University Frankfurt Department of Medicine Germany Goethe University Frankfurt Department of Informatics Germany and Frankfurt Cancer Insititute Germany Department of Computer Science IT University of Copenhagen Copenhagen Denmark Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V. Dortmund Germany Imaging Platform Broad Institute of MIT and Harvard CambridgeMA United States Centre for Statistics in Medicine University of Oxford Oxford United Kingdom Center for Biomedical In

Validation metrics are key for tracking scientific progress and bridging the current chasm between artificial intelligence (AI) research and its translation into practice. However, increasing evidence shows that particularly in image analysis, metrics are often chosen inadequately. While taking into account the individual strengths, weaknesses, and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multi-stage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides the first reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. While focused on biomedical image analysis, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. The work serves to enhance global comprehension of a key topic in image analysis validation. © 2023, CC BY.

关键词： computer vision

Benchmarking quality-dependent and cost-sensitive score-level multimodal biometric fusion algorithms

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Poh, Norman Bourlai, Thirimachos Kittler, Josef Allano, Lorene Alonso-Fernandez, Fernando Ambekar, Onkar Baker, John Dorizzi, Bernadette Fatukasi, Omolara Fierrez, Julian Ganster, Harald Ortega-Garcia, Javier Maurer, Donald Salah, Albert Ali Scheidat, Tobias Vielhauer, Claus Centre for Vision Speech and Signal Processing School of Electronics and Physical Sciences University of Surrey Guildford SurreyGU2 7XH United Kingdom Biometrics Center Lane Department of Computer Science and Electrical Engineering College of Engineering and Mineral Resources West Virginia University MorgantownWV26506-6109 United States Institut Telecom Telecom and Management SudParis Evry91011 France Biometric Recognition Group – ATVS Escuela Politecnica Superior Universidad Autonoma de Madrid Madrid28049 Spain Amsterdam1098 XG Netherlands Applied Physics Laboratory Johns Hopkins University LaurelMD20723 United States Electronics and Physics Department Institut Telecom Telecom and Management SudParis Evry91011 France Institute of Digital Image Processing Joanneum Research Graz8010 Austria ISLA-ISIS University of Amsterdam Amsterdam1098 XG Netherlands Department of Informatics and Media Brandenburg University of Applied Sciences Brandenburg an der HavelD-14770 Germany Research Group Multimedia and Security Department of Technical and Business Information Systems Faculty of Computer Science Otto-von-Guericke-University of Magdeburg MagdeburgD-39106 Germany

Automatically verifying the identity of a person by means of biometrics (e.g., face and fingerprint) is an important application in our day-to-day activities such as accessing banking services and security control in airports. To increase the system reliability, several biometric devices are often used. Such a combined system is known as a multimodal biometric system. This paper reports a benchmarking study carried out within the framework of the BioSecure DS2 (Access Control) evaluation campaign organized by the University of Surrey, involving face, fingerprint, and iris biometrics for person authentication, targeting the application of physical access control in a medium-size establishment with some 500 persons. While multimodal biometrics is a well-investigated subject in the literature, there exists no benchmark for a fusion algorithm comparison. Working towards this goal, we designed two sets of experiments: quality-dependent and cost-sensitive evaluation. The quality-dependent evaluation aims at assessing how well fusion algorithms can perform under changing quality of raw biometric images principally due to change of devices. The cost-sensitive evaluation, on the other hand, investigates how well a fusion algorithm can perform given restricted computation and in the presence of software and hardware failures, resulting in errors such as failure-to-acquire and failure-to-match. Since multiple capturing devices are available, a fusion algorithm should be able to handle this nonideal but nevertheless realistic scenario. In both evaluations, each fusion algorithm is provided with scores from each biometric comparison subsystem as well as the quality measures of both the template and the query data. The response to the call of the evaluation campaign proved very encouraging, with the submission of 22 fusion systems. To the best of our knowledge, this campaign is the first attempt to benchmark quality-based multimodal fusion algorithms. In the presence of changing

关键词： Benchmarking