检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Shao, Helen De Santi, Natalí S.M. Villaescusa-Navarro, Francisco Teyssier, Romain Ni, Yueying Anglés-Alcázar, Daniel Genel, Shy Steinwandel, Ulrich P. Hernández-Martínez, Elena Dolag, Klaus Lovell, Christopher C. Garrison, Lehman H. Visbal, Eli Kulkarni, Mihir Hernquist, Lars Castro, Tiago Vogelsberger, Mark Department of Astrophysical Sciences Princeton University 4 Ivy Lane PrincetonNJ08544 United States Center for Computational Astrophysics Flatiron Institute 162 5th Avenue New YorkNY10010 United States Instituto de Física Universidade de São Paulo R. do Matão 1371 São Paulo05508-900 Brazil Harvard-Smithsonian Center for Astrophysics 60 Garden Street CambridgeMA02138 United States McWilliams Center for Cosmology Department of Physics Carnegie Mellon University PittsburghPA15213 United States Department of Physics University of Connecticut 196 Auditorium Road U-3046 StorrsCT06269 United States Columbia Astrophysics Laboratory Columbia University New YorkNY10027 United States Universitäts-Sternwarte Fakultät für Physik Ludwig-Maximilians-Universität München Scheinerstr. 1 München81679 Germany Max-Planck-Institut für Astrophysik Karl-Schwarzschild-Straße 1 Garching85741 Germany Institute of Cosmology and Gravitation University of Portsmouth Burnaby Road PortsmouthPO1 3FX United Kingdom Centre for Astrophysics Research School of Physics Engineering & Computer Science University of Hertfordshire HatfieldAL10 9AB United Kingdom Department of Physics and Astronomy Ritter Astrophysical Research Center University of Toledo 2801 W Bancroft Street ToledoOH43606 United States Harvard-Smithsonian Center for Astrophysics 60 Garden Street CambridgeMA02138 United States INAF-Osservatorio Astronomico di Trieste Via G. B. Tiepolo 11 TriesteI-34143 Italy INFN Sezione di Trieste Via Valerio 2 TS TriesteI-34127 Italy IFPU Institute for Fundamental Physics of the Universe via Beirut 2 Trieste34151 Italy Kavli Institute for Astrophysics and Space Research Department of Physics Mit CambridgeMA02139 United States The Nsf Ai Institute for Artificial Intelligence and Fundamental Interactions Massachusetts Institute of Technology CambridgeMA02139 United States

We discover analytic equations that can infer the value of Ωmfrom the positions and velocity moduli of halo and galaxy catalogues. The equations are derived by combining a tailored graph neural network (GNN) architecture with symbolic regression. We first train the GNN on dark matter halos from Gadget N-body simulations to perform field-level likelihood-free inference, and show that our model can infer Ωmwith ∼ 6% accuracy from halo catalogues of thousands of N-body simulations run with six different codes: Abacus, CUBEP3M, Gadget, Enzo, PKDGrav3, and Ramses. By applying symbolic regression to the different parts comprising the GNN, we derive equations that can predict Ωmfrom halo catalogues of simulations run with all of the above codes with accuracies similar to those of the GNN. We show that by tuning a single free parameter, our equations can also infer the value of Ωmfrom galaxy catalogues of thousands of state-of-the-art hydrodynamic simulations of the CAMELS project, each with a different astrophysics model, run with five distinct codes that employ different subgrid physics: IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE. Furthermore, the equations also perform well when tested on galaxy catalogues from simulations covering a vast region in parameter space that samples variations in 5 cosmological and 23 astrophysical parameters. We speculate that the equations may reflect the existence of a fundamental physics relation between the phase-space distribution of generic tracers and Ωm, one that is not affected by galaxy formation physics down to scales as small as 10 h-1kpc. © 2023, CC BY.

关键词： Galaxies

来源：评论

学校读者我要写书评

暂无评论

Multi-Head Attention with Disagreement Regularization

arXiv

引用

arXiv 2018年

作者： Li, Jian Tu, Zhaopeng Yang, Baosong Lyu, Michael R. Zhang, Tong Department of Computer Science and Engineering Chinese University of Hong Kong Shenzhen Research Institute Chinese University of Hong Kong Tencent AI Lab University of Macau

Multi-head attention is appealing for the ability to jointly attend to information from different representation subspaces at different positions. In this work, we introduce a disagreement regularization to explicitly encourage the diversity among multiple attention heads. Specifically, we propose three types of disagreement regularization, which respectively encourage the subspace, the attended positions, and the output representation associated with each attention head to be different from other heads. Experimental results on widely-used WMT14 English⇒German and WMT17 Chinese⇒English translation tasks demonstrate the effectiveness and universality of the proposed approach. Copyright © 2018, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Scalable distributed DNN training using tensorflow and CUDA-Aware MPI: Characterization, designs, and performance evaluation

arXiv

引用

arXiv 2018年

作者： Awan, Ammar Ahmad Chu, Ching-Hsiang Subramoni, Hari Panda, Dhabaleswar K. Bédorf, Jeroen Department of Computer Science and Engineering Ohio State University Minds.ai Santa Cruz United States Leiden Observatory Leiden University Leiden Netherlands

The current wave of advances in Machine Learning (ml) and Deep Learning (DL) have been triggered by the availability of large-scale datasets, efficient CPU and GPU hardware, and development of easy-to-use software frameworks like TensorFlow (TF), Caffe and Torch. TensorFlow has been, by far, the most widely adopted ml/DL framework. However, little exists in the literature that provides a thorough understanding of the capabilities which TensorFlow offers for the distributed training of large ml/DL models that need computation and communication at scale. Most commonly used distributed training approaches for TF can be categorized as follows: 1) Google Remote Procedure Call (gRPC), 2) gRPC+'X': X = (InfiniBand Verbs, Message Passing Interface (MPI), and GPUDirect RDMA), and 3) No-gRPC: Baidu Allreduce with MPI, Horovod with MPI, and Horovod with NVIDIA NCCL. In this paper, we provide an in-depth performance characterization and analysis of these distributed training approaches on various GPU clusters including the Piz Daint system (#6 on Top500). We perform experiments to gain novel insights along the following vectors: 1) Application-level scalability of DNN training, 2) Effect of Batch Size on scaling efficiency, 3) Impact of the MPI library used for no-gRPC approaches, and 4) Type and size of DNN architectures (e.g ResNet vs. MobileNet). Based on these experiments, we present two key insights: 1) Overall, No-gRPC designs achieve better performance compared to gRPC-based approaches for most configurations, and 2) The performance of No-gRPC is heavily influenced by the gradient aggregation using the Allreduce communication pattern. Finally, we propose a truly CUDA-Aware MPI_Allreduce design that exploits 1) CUDA kernels to perform large reductions on the GPU and 2) A pointer cache to avoid overheads involved in queries to the CUDA driver. Our proposed designs have been implemented in MVAPICH2-GDR and offer 5-17× better performance than NCCL2 for small and medium messa

关键词： Message passing

来源：评论

学校读者我要写书评

暂无评论

Domain Adversarial Training for Accented Speech Recognition

Domain Adversarial Training for Accented Speech Recognition

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Sining Sun Ching-Feng Yeh Mei-Yuh Hwang Mari Ostendorf Lei Xie School of Computer Science Northwestern Polytechnical University Xi'an China Mobvoi AI Lab Seattle USA Department of Electrical Engineering University of Washington Seattle USA

ISBN: (纸本)9781538646595

In this paper, we propose a domain adversarial training (DAT) algorithm to alleviate the accented speech recognition problem. In order to reduce the mismatch between labeled source domain data ("standard" accent) and unlabeled target domain data (with heavy accents), we augment the learning objective for a Kaldi TDNN network with a domain adversarial training (DAT) objective to encourage the model to learn accent-invariant features. In experiments with three Mandarin accents, we show that DAT yields up to 7.45% relative character error rate reduction when we do not have transcriptions of the accented speech, compared with the baseline trained on standard accent data only. We also find a benefit from DAT when used in combination with training from automatic transcriptions on the accented data. Furthermore, we find that DAT is superior to multi-task learning for accented speech recognition.

关键词： Domain adaptation accent robust speech recognition domain adversarial training Digital audio tape Speech recognition Sodium Glutamate Genetic Transcription Educational objectives Transcription Training Domain character error rate

来源：评论

学校读者我要写书评

暂无评论

An Incremental Reseeding Strategy for Clustering

An Incremental Reseeding Strategy for Clustering

引用

International conference on Imaging, Vision and Learning Based on Optimization and PDEs, IVLOPDE 2016

作者： Bresson, Xavier Hu, Huiyi Laurent, Thomas Szlam, Arthur von Brecht, James School of Computer Science and Engineering Nanyang Technological University Singapore Singapore Google Inc Mountain ViewCA United States Department of Mathematics Loyola Marymount University Los AngelesCA United States Facebook AI Research New YorkNY United States Department of Mathematics California State University Long BeachCA United States

ISBN: (纸本)9783642543005

We propose an easy-to-implement and highly parallelizable algorithm for multiway graph partitioning. The algorithm proceeds by alternating three simple routines in an iterative fashion: diffusion, thresholding, and random sampling. We demonstrate experimentally that the proper combination of these ingredients leads to an algorithm that achieves state-of-the-art performance in terms of cluster purity on standard benchmark data sets. We also describe a coarsen, cluster and refine approach similar to Dhillon et al. (IEEE Trans Pattern Anal Mach Intell 29(11):1944–1957, 2007) and Karypis and Kumar (SIAM J Sci Comput 20(1):359–392, 1998) that removes an order of magnitude from the runtime of our algorithm while still maintaining competitive accuracy. © 2018, Springer International Publishing AG, part of Springer Nature.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

RepMet: Representative-based metric learning for classification and few-shot object detection

arXiv

引用

arXiv 2018年

作者： Karlinsky, Leonid Shtok, Joseph Harary, Sivan Schwartz, Eli aides, Amit Feris, Rogerio Giryes, Raja Bronstein, Alex M. IBM Research AI Haifa Israel School of Electrical Engineering Tel-Aviv University Tel-Aviv Israel Department of Computer Science Technion Haifa Israel

Distance metric learning (Dml) has been successfully applied to object classification, both in the standard regime of rich training data and in the few-shot scenario, where each category is represented by only a few examples. In this work, we propose a new method for Dml that simultaneously learns the backbone network parameters, the embedding space, and the multi-modal distribution of each of the training categories in that space, in a single end-to-end training process. Our approach outperforms state-of-the-art methods for Dml-based object classification on a variety of standard fine-grained datasets. Furthermore, we demonstrate the effectiveness of our approach on the problem of few-shot object detection, by incorporating the proposed Dml architecture as a classification head into a standard object detection model. We achieve the best results on the ImageNet-LOC dataset compared to strong baselines, when only a few training examples are available. We also offer the community a new episodic benchmark based on the ImageNet dataset for the few-shot object detection task. Code will be released upon acceptance of the paper. Copyright © 2018, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

On the role of theory and modeling in neuroscience

arXiv

引用

arXiv 2020年

作者： Levenstein, Daniel Alvarez, Veronica A. Amarasingham, Asohan Azab, Habiba Chen, Zhe S. Gerkin, Richard C. Hasenstaub, Andrea Iyer, Ramakrishnan Jolivet, Renaud B. Marzen, Sarah Monaco, Joseph D. Prinz, Astrid A. Quraishi, Salma Santamaria, Fidel Shivkumar, Sabyasachi Singh, Matthew F. Traub, Roger Rotstein, Horacio G. Nadim, Farzan Redish, A. David Montreal Neurological Institute McGill University MontrealQC Canada Laboratory on Neurobiology of Compulsive Behaviors National Institute on Alcohol Abuse and Alcoholism NIH BethesdaMD United States Department of Neuroscience Center for Magnetic Resonance Research University of Minnesota MinneapolisMN United States School of Life Sciences Arizona State University TempeAZ United States Department of Otolaryngology - Head and Neck Surgery University of California San Francisco San FranciscoCA United States Allen Institute for Brain Science SeattleWA United States Maastricht University Maastricht Netherlands Department of Biomedical Engineering Johns Hopkins University School of Medicine BaltimoreMD United States Department of Biology University of Texas at San Antonio San AntonioTX United States Department of Psychological & Brain Sciences Department of Electrical & Systems Engineering Washington University in St. Louis St. LouisMO United States IBM T.J. Watson Research Center AI Foundations Yorktown HeightsNY United States W. M. Keck Science Department Pitzer Scripps and Claremont McKenna Colleges ClaremontCA United States Department of Biology Emory University AtlantaGA United States Department of Mathematics City College of New York United States Departments of Biology Computer Science and Psychology The Graduate Center City University of New York United States Brain and Cognitive Sciences University of Rochester RochesterNY United States Federated Department of Biological Sciences New Jersey Institute of Technology Rutgers University Institute for Brain and Neuroscience Research New Jersey Institute of Technology NewarkNJ United States Department of Neuroscience University of Minnesota MinneapolisMN United States Department of Psychiatry Neuroscience & Physiology New York University School of Medicine United States Mila - Quebec AI Institute MontrealQC Canada

In recent years, the field of neuroscience has gone through rapid experimental advances and a significant increase in the use of quantitative and computational methods. This growth has created a need for clearer analyses of the theory and modeling approaches used in the field. This issue is particularly complex in neuroscience because the field studies phenomena that cross a wide range of scales and often require consideration at varying degrees of abstraction, from precise biophysical interactions to the computations they implement. We argue that a pragmatic perspective of science, in which descriptive, mechanistic, and normative models and theories each play a distinct role in defining and bridging levels of abstraction will facilitate neuroscientific practice. This analysis leads to methodological suggestions, including selecting a level of abstraction that is appropriate for a given problem, identifying transfer functions to connect models and data, and the use of models themselves as a form of experiment. © 2020, CC BY.

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

A Provably Convergent Scheme for Compressive Sensing under Random Generative Priors

arXiv

引用

arXiv 2018年

作者： Huang, Wen Hand, Paul Heckel, Reinhard Voroninski, Vladislav Department of Mathematical Sciences Xiamen University China Department of Mathematics and College of Computer and Information Science Northeastern University United States Department of Electrical and Computer Engineering Rice University United States Helm.ai United States

Deep generative modeling has led to new and state of the art approaches for enforcing structural priors in a variety of inverse problems. In contrast to priors given by sparsity, deep models can provide direct low-dimensional parameterizations of the manifold of images or signals belonging to a particular natural class, allowing for recovery algorithms to be posed in a low-dimensional space. This dimensionality may even be lower than the sparsity level of the same signals when viewed in a fixed basis. What is not known about these methods is whether there are computationally efficient algorithms whose sample complexity is optimal in the dimensionality of the representation given by the generative model. In this paper, we present such an algorithm and analysis. Under the assumption that the generative model is a neural network that is sufficiently expansive at each layer and has Gaussian weights, we provide a gradient descent scheme and prove that for noisy compressive measurements of a signal in the range of the model, the algorithm converges to that signal, up to the noise level. The scaling of the sample complexity with respect to the input dimensionality of the generative prior is linear, and thus can not be improved except for constants and factors of other variables. To the best of the authors’ knowledge, this is the first recovery guarantee for compressive sensing under generative priors by a computationally efficient algorithm. Copyright © 2018, The Authors. All rights reserved.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Publisher Correction: Brain imaging with portable low-field MRI

Nature Reviews Bioengineering

引用

Nature Reviews Bioengineering 2023年 680页

作者： W. Taylor Kimberly Annabel J. Sorby-Adams Andrew G. Webb Ed X. Wu Rachel Beekman Kevin N. Sheth Ritvij Bowry Steven J. Schiff Adam de Havenon Francis X. Shen Gordon Sze Pamela Schaefer Juan Eugenio Iglesias Matthew S. Rosen Department of Neurology and the Center for Genomic Medicine Massachusetts General Hospital and Harvard Medical School Boston MA USA Department of Radiology Leiden University Medical Center Leiden The Netherlands Laboratory of Biomedical Imaging and Signal Processing Department of Electrical and Electronic Engineering The University of Hong Kong Hong Kong SAR China Division of Neurocritical Care and Emergency Neurology Department of Neurology Yale New Haven Hospital and Yale School of Medicine Yale Center for Brain & Mind Health New Haven CT USA Departments of Neurosurgery and Neurology McGovern Medical School University of Texas Health Neurosciences Houston TX USA Department of Neurosurgery Yale School of Medicine New Haven CT USA Division of Vascular Neurology Department of Neurology Yale New Haven Hospital and Yale School of Medicine New Haven CT USA Harvard Medical School Center for Bioethics Harvard law School Boston MA USA Department of Psychiatry Massachusetts General Hospital Boston MA USA Department of Radiology Yale New Haven Hospital and Yale School of Medicine New Haven CT USA Division of Neuroradiology Department of Radiology Massachusetts General Hospital and Harvard Medical School Boston MA USA Athinoula A. Martinos Center for Biomedical Imaging Department of Radiology Massachusetts General Hospital and Harvard Medical School Boston MA USA Centre for Medical Image Computing University College London London UK Computer Science and AI Laboratory Massachusetts Institute of Technology Boston MA USA

来源：评论

学校读者我要写书评

暂无评论

A deep learning system for detecting systemic lupus erythematosus from retinal images

Cell Reports Medicine

引用

Cell Reports Medicine 2025年

作者： Tingyao Li Shiqun Lin Zhouyu Guan Yukun Zhou Dian Zeng Zheyuan Wang Yan Zhou Pinqi Fang Shujie Yu Ruhan Liu Xiang Chen Yan-Ran (Joyce) Wang Yuwei Lu Jia Shu Yiming Qin Yiting Wu Yilan Wu Chan Wu Shangzhu Zhang Jie Shen Bin Sheng Shanghai International Joint Laboratory of Intelligent Prevention and Treatment for Metabolic Diseases Department of Computer Science and Engineering School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Department of Endocrinology and Metabolism Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine Shanghai Diabetes Institute Shanghai Clinical Center for Diabetes Shanghai China MOE Key Laboratory of AI School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Furong Laboratory Central South University Changsha Hunan China Institute for Proactive Healthcare Shanghai Jiao Tong University Shanghai China Department of Ophthalmology Peking Union Medical College Hospital Peking Union Medical College Chinese Academy of Medical Sciences Beijing China Beijing Key Laboratory of Fundus Diseases Intelligent Diagnosis & Drug/Device Development and Translation Beijing China Centre for Medical Image Computing University College London London UK NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust London UK Institute of Ophthalmology University College London London UK School of Clinical Medicine Beijing Tsinghua Changgung Hospital Tsinghua Medicine Tsinghua University Beijing China Medical Records and Statistics Office Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine Shanghai China School of Medicine Stanford University Stanford CA USA Department of Rheumatology and Clinical Immunology Chinese Academy of Medical Sciences & Peking Union Medical College National Clinical Research Center for Dermatologic and Immunologic Diseases (NCRC-DID) Ministry of Science & Technology State Key Laboratory of Complex Severe and Rare Diseases Peking Union Medical College Hospital (PUMCH) Key Laboratory of Rheumatology and Clinical Immunology Ministry of Education Beijing China

Systemic lupus erythematosus (SLE) is a serious autoimmune disorder predominantly affecting women. However, screening for SLE and related complications poses significant challenges globally, due to complex diagnostic criteria and public unawareness. Since SLE-related retinal involvement could provide insights into disease activity and severity, we develop a deep learning system (DeepSLE) to detect SLE and its retinal and kidney complications from retinal images. In multi-ethnic validation datasets comprising 247,718 images from China and UK, DeepSLE achieves areas under the receiver operating characteristic curve of 0.822–0.969 for SLE. Additionally, DeepSLE demonstrates robust performance across subgroups stratified by gender, age, ethnicity, and socioeconomic status. To ensure DeepSLE’s explainability, we conduct both qualitative and quantitative analyses. Furthermore, in a prospective reader study, DeepSLE demonstrates higher sensitivities compared with primary care physicians. Altogether, DeepSLE offers digital solutions for detecting SLE and related complications from retinal images, holding potential for future clinical deployment.

关键词： systemic lupus erythematosus retinal imaging deep learning lupus retinopathy lupus nephritis color fundus photography health equity multi-ethnic validation opportunistic screening reader study

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：