检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Li, Jing Wu, Botong Sun, Xinwei Wang, Yizhou Dept. of Computer Science Peking University China Peking University Adv. Inst. of Info. Tech China Center on Frontiers of Computing Studies Peking University China Microsoft Research Asia China Deepwise Ai Lab China

We propose a causal hidden Markov model to achieve robust prediction of irreversible disease at an early stage, which is safety-critical and vital for medical treatment in early stages. Specifically, we introduce the hidden variables which propagate to generate medical data at each time step. To avoid learning spurious correlation (e.g., confounding bias), we explicitly separate these hidden variables into three parts: A) the disease (clinical)-related part;b) the disease (non-clinical)-related part;c) others, with only a),b) causally related to the disease however c) may contain spurious correlations (with the disease) inherited from the data provided. With personal attributes and disease label respectively provided as side information and supervision, we prove that these disease-related hidden variables can be disentangled from others, implying the avoidance of spurious correlation for generalization to medical data from other (out-of-) distributions. Guaranteed by this result, we propose a sequential variational auto-encoder with a reformulated objective function. We apply our model to the early prediction of peripapillary atrophy and achieve promising results on out-of-distribution test data. Further, the ablation study empirically shows the effectiveness of each component in our method. And the visualization shows the accurate identification of lesion regions from others. 1. Copyright © 2021, The Authors. All rights reserved.

关键词： hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Context modeling in 3D human pose estimation: A unified perspective

arXiv

引用

arXiv 2021年

作者： Ma, Xiaoxuan Su, Jiajun Wang, Chunyu Ci, Hai Wang, Yizhou Dept. of Computer Science Center on Frontiers of Computing Studies Peking University China Center for Data Science Adv. Inst. of Info. Tech. Peking University China Beijing Film Academy China Microsoft Research Asia Deepwise AI Lab

Estimating 3D human pose from a single image suffers from severe ambiguity since multiple 3D joint configurations may have the same 2D projection. The state-of-the-art methods often rely on context modeling methods such as pictorial structure model (PSM) or graph neural network (GNN) to reduce ambiguity. However, there is no study that rigorously compares them side by side. So we first present a general formula for context modeling in which both PSM and GNN are its special cases. By comparing the two methods, we found that the end-to-end training scheme in GNN and the limb length constraints in PSM are two complementary factors to improve results. To combine their advantages, we propose ContextPose based on attention mechanism that allows enforcing soft limb length constraints in a deep network. The approach effectively reduces the chance of getting absurd 3D pose estimates with incorrect limb lengths and achieves state-of-the-art results on two benchmark datasets. More importantly, the introduction of limb length constraints into deep networks enables the approach to achieve much better generalization performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Design of Polymorphic Real-Time Computational Model for UAV Application 17

Design of Polymorphic Real-Time Computational Model for UAV ...

引用

17th IEEE/ACIS International Conference on computer and Information Science, ICIS 2018

作者： Luo, Mei Li, Linlin Wang, Baocheng Zhang, Kailong Computer Lab Research Inst. of Hi-Tech Xi'an China School of Computer Northwest Polytechnical University Xi'an China

ISBN: (纸本)9781538658925

According to the requirements of informational weapon equipment, the application of polymorphic real-Time computation technology in complex embedded system is studied in this paper. The polymorphic real-Time computation model for unmanned aerial vehicle(UAV) is designed and verified from the aspects of modal management and polymorphic tasks. The simulation results show that the proposed model can provide modal decision and ensure its real-Time calculation to adapt the demand change of UAV system. © 2018 IEEE.

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

Forecasting irreversible disease via progression learning

arXiv

引用

arXiv 2020年

作者： Wu, Botong Ren, Sijie Li, Jing Sun, Xinwei Li, Shiming Wang, Yizhou Dept. of Computer Science Peking University Adv. Inst. of Info. Tech Peking University Center on Frontiers of Computing Studies Peking University Microsoft Research Asia Beijing Tongren Hospital Capital Medical University Deepwise AI Lab Beijing Stars Universal Technology Co. Ltd

Forecasting Parapapillary atrophy (PPA), i.e., a symptom related to most irreversible eye diseases, provides an alarm for implementing an intervention to slow down the disease progression at early stage. A key question for this forecast is: how to fully utilize the historical data (e.g., retinal image) up to the current stage for future disease prediction? In this paper, we provide an answer with a novel framework, namely Disease Forecast via Progression Learning (DFPL), which exploits the irreversibility prior (i.e., cannot be reversed once diagnosed). Specifically, based on this prior, we decompose two factors that contribute to the prediction of the future disease: i) the current disease label given the data (retinal image, clinical attributes) at present and ii) the future disease label given the progression of the retinal images that from the current to the future. To model these two factors, we introduce the current and progression predictors in DFPL, respectively. In order to account for the degree of progression of the disease, we propose a temporal generative model to accurately generate the future image and compare it with the current one to get a residual image. The generative model is implemented by a recurrent neural network, in order to exploit the dependency of the historical data. To verify our approach, we apply it to a PPA in-house dataset and it yields a significant improvement (e.g., 4.48% of accuracy;3.45% of AUC) over others. Besides, our generative model can accurately localize the disease-related regions. Copyright © 2020, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

TCGM: An Information-Theoretic Framework for Semi-Supervised Multi-Modality Learning

arXiv

引用

arXiv 2020年

作者： Sun, Xinwei Xu, Yilun Cao, Peng Kong, Yuqing Hu, Lingjing Zhang, Shanghang Wang, Yizhou Microsoft Research-Asia Center on Frontiers of Computing Studies Peking University China Dept. of Computer Science Peking University China Yanjing Medical College Capital Medical University China UC Berkeley United States Adv. Inst. of Info. Tech Peking University China Deepwise AI Lab

Fusing data from multiple modalities provides more information to train machine learning systems. However, it is prohibitively expensive and time-consuming to label each modality with a large amount of data, which leads to a crucial problem of semi-supervised multi-modal learning. Existing methods suffer from either ineffective fusion across modalities or lack of theoretical guarantees under proper assumptions. In this paper, we propose a novel information-theoretic approach - namely, Total Correlation Gain Maximization (TCGM) – for semi-supervised multi-modal learning, which is endowed with promising properties: (i) it can utilize effectively the information across different modalities of unlabeled data points to facilitate training classifiers of each modality (ii) it has theoretical guarantee to identify Bayesian classifiers, i.e., the ground truth posteriors of all modalities. Specifically, by maximizing TC-induced loss (namely TC gain) over classifiers of all modalities, these classifiers can cooperatively discover the equivalent class of ground-truth classifiers;and identify the unique ones by leveraging limited percentage of labeled data. We apply our method to various tasks and achieve state-of-the-art results, including the news classification (Newsgroup dataset), emotion recognition (IEMOCAP and MOSI datasets), and disease prediction (Alzheimer’s Disease Neuroimaging Initiative dataset). Copyright © 2020, The Authors. All rights reserved.

关键词： Equivalence classes

来源：评论

学校读者我要写书评

暂无评论

Design of Polymorphic Real-Time Computational Model for UAV Application

Design of Polymorphic Real-Time Computational Model for UAV ...

引用

International Conference on computer and Information Science (ACIS)

作者： Mei Luo Linlin Li Baocheng Wang Kailong Zhang Computer Lab Research Inst. Of Hi-Tech Xi’an China School of Computer Northwest Polytechnical University Xi’an China

According to the requirements of informational weapon equipment, the application of polymorphic real-time computation technology in complex embedded system is studied in this paper. The polymorphic real-time computation model for unmanned aerial vehicle(UAV) is designed and verified from the aspects of modal management and polymorphic tasks. The simulation results show that the proposed model can provide modal decision and ensure its real-time calculation to adapt the demand change of UAV system.

关键词： Task analysis Real-time systems Computational modeling computer architecture Unmanned aerial vehicles Adaptation models Target tracking

来源：评论

学校读者我要写书评

暂无评论

引用

29th International Conference on Software Engineering and Knowledge Engineering, SEKE 2017

作者： Napoleão, Bianca M. Felizardo, Katia R. De Souza, Érica F. Vijaykumar, Nandamudi L. Depart. of Computer Federal Tech. Univ. of Paraná Cornélio Procópio PR Brazil Lab. of Comp. and Applied Math. Nat. Inst. for Space Research São José dos Campos SP Brazil

ISBN: (纸本)1891706411

Background: Several researchers have reported their experiences in applying secondary studies in Software Engineering (SE), however, there is a lack of studies discussing the distinction between Systematic Mappings (SMs) and Systematic Literature Reviews (SLRs). Aims: The objective of this paper is to present the results of a tertiary study conducted to collect and evaluate evidence to better understand similarities and differences between SLRs and SMs related to four aspects: research question, search string, search strategy and quality assessment. Method: We identified 170 secondary studies that were reviewed to answer a set of research Questions (RQ) related to the practical conduction of secondary studies in SE. Results: Results show that both SLRs and SMs have generic RQs, broad search strings, and adopt automatic search as search strategy. However, quality assessment has been more widely adopted in SLRs. Conclusions: In practice, only the quality assessment is conducted differently in SLRs and SMs.

关键词： Software engineering

来源：评论

学校读者我要写书评

暂无评论

Properties of Rubble-Pile Asteroid (101955) Bennu from OSIRIS-REx Imaging and Thermal Analysis

Properties of Rubble-Pile Asteroid (101955) Bennu from OSIRI...

引用

作者： DellaGiustina, D. N Emery, J. P Golish, D. R Rozitis, B Bennett, C. A Burke, K. N Ballouz, R Becker, K. J Christensen, P. R Drouet d’Aubigny, C. Y Hamilton, V. E Reuter, D. C Rizk, B Simon, A. A Asphaug, E Bandfield, J. L Barnouin, O. S Barucci, M. A Binzel, R. P Bottke, W. F Bowles, N. E Campins, H Clark, B. C Clark, B. E Connolly, H. C. J Daly, M. G De Leon, J Delbo, M Deshapriya, J. D. P Fornasier, S Hergenrother, C. W Jawin, E. R Howell, E. S Kaplan, H. H Kareta, T. R Corre, L. L Li, J Licandro, J Lim, L. F Michel, P Molaro, J Nolan, M. C Popescu, M Rizos Garcia, J. L Ryan, A Schwartz, S. R Shultz, N Siegler, M. A Smith, P. H Tatsumi, E Thomas, C. A Walsh, K. J Wolner, C. W. V Zou, X Lauretta, D. S highsmith, D. E Small, J Vokrouhlick, D Brown, E Hanna, K. L. D Warren, T Brunet, C Chicoine, R. A Desjardins, S Gaudreau, D Haltigin, T Millington-Veloza, S Rubi, A Aponte, J Gorius, N Lunsford, A Allen, B Grindlay, J Guevel, D Hoak, D Hong, J Schrader, D. L Bayron, J Golubov, O Sánchez, P Stromberg, J hirabayashi, M Hartzell, C. M Oliver, S Rascon, M Harch, A Joseph, J Squyres, S Richardson, D McGraw, L Ghent, R Al Asad, M. M Johnson, C. L Philpott, L Susorney, H. C. M Cloutis, E. A Hanna, R. D Ciceri, F hildebrand, A. R Ibrahim, E Breitenfeld, L Glotch, T Rogers, A. D Ferrone, S Fernandez, Y Chang, W Cheuvront, A Trang, D Tachibana, S Yurimoto, H Brucato, J. R Poggiali, G Pajola, M Dotto, E Mazzotta Epifani, E Crombie, M. K Lantz, C Izawa, M. R. M Leon, J. D Clemett, S Thomas-Keprta, K Van wal, S Yoshikawa, M Bellerose, J Bhaskaran, S Boyles, C Chesley, S. R Elder, C. M Farnocchia, D Harbison, A Kennedy, B Knight, A Martinez-Vlasoff, N Mastrodemos, N McElrath, T Owen, W Park, R Rush, B Swanson, L Takahashi, Y Velez, D Yetter, K Thayer, C Adam, C Antreasian, P Bauman, J Bryan, C Carcich, B Corvin, M Geeraert, J Hoffman, J Leonard, J. M Lessac-Chenen, E Levine, A McAdams, J McCarthy, L Nelson, D Page, B Pelgrift, B Sahr, E Stakkestad, K Stanbridge, D Wibben, D Williams, B Williams, K Wolff, P Hayne, P Kubitschek, NASA Ames Research Center Moffett Field CA United States NASA Johnson Space Center Houston TX United States NASA Marshall Space Flight Center Huntsville AL United States University of Tokyo Tokyo Japan University of Calgary Calgary Alberta Canada NASA Langley Research Center Hampton VA United States Royal Ontario Museum Toronto Ontario Canada University of Toronto Toronto Ontario Canada University of British Columbia Vancouver British Columbia Canada Instituto de Astrofisica de Canarias La Laguna Tenerife Spain Massachusetts Institute of Technology (MIT) Cambridge MA United States NASA Goddard Space Flight Center Greenbelt MD United States University of Central Florida Orlando FL United States Arizona Univ. Tucson AZ United States Maryland Univ. College Park MD United States Planetary Science Inst. Tucson AZ United States Jacobs Technology Inc. Houston TX United States Lockheed Martin Space Systems Co. Littleton CO United States Colorado Univ. Boulder CO United States Jet Propulsion Lab. California Inst. of Tech. Pasadena CA United States Catholic Univ. of America Washington DC United States Johns Hopkins Univ. Laurel MD United States Southwest Research Inst. Boulder CO United States Search for Extraterrestrial Intelligence Inst. Mountain View CA United States NASA Headquarters Washington DC United States Hawaii Univ. Manoa Honolulu HI United States Arizona State Univ. (ASU) Tempe AZ United States Harvard Univ. Cambridge MA United States Stony Brook Univ. Stony Brook NY United States Cornell Univ. Ithaca NY United States The University of Texas at Austin Austin TX United States Centre National de la Recherche Scientifique Annecy-le-Vieux France KinetX Aerospace Inc Simi Valley CA United States INAF–Osservatorio Astronomico di Rome Rome Italy Commonwealth Scientific and Industrial Research Organization (CSIRO) Canberra Australia Ithaca Coll. Ithaca NY United States Rowan Univ. Glassboro NJ United States Aerospace Corp. Chantilly VA United S

Establishing the abundance and physical properties of regolith and boulders on asteroids is crucial for understanding the formation and degradation mechanisms at work on their surfaces. Using images and thermal data from NASA's Origins, Spectral Interpretation, Resource Identification, and Security-Regolith Explorer (OSIRIS-REx) spacecraft, we show that asteroid (101955) Bennu's surface is globally rough, dense with boulders, and low in albedo. The number of boulders is surprising given Bennu's moderate thermal inertia, suggesting that simple models linking thermal inertia to particle size do not adequately capture the complexity relating these properties. At the same time, we find evidence for a wide range of particle sizes with distinct albedo characteristics. Our findings imply that ages of Bennu's surface particles span from the disruption of the asteroid's parent body (boulders) to recent in situ production (micrometre-scale particles).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Perspectives on Chemical Oceanography in the 21st century: Participants of the COME ABOARD Meeting examine aspects of the field in the context of 40 years of DISCO

Perspectives on Chemical Oceanography in the 21st century: P...

引用

作者： Fassbender, Andrea J. Palevsky, hilary I. Martz, Todd R. Ingalls, Anitra E. Gledhill, Martha Fawcett, Sarah E. Brandes, Jay A. Aluwihare, Lihini I. Aluwihare, Lihini Anderson, Robert Bender, Sara Boyle, Ed Brandes, Jay Bronk, Debbie Buesseler, Ken Burdige, David Casciotti, Karen Close, hilary Conte, Maureen Cutter, Greg Estapa, Meg Fassbender, Andrea Fennel, Katja Ferron, Sara Glazer, Brian Goni, Miguel Grand, Maxime Guay, Chris Hatta, Mariko Hayes, Chris Horner, Tristan Ingall, Ellery Ingalls, Anitra Johnson, Ken Juranek, Laurie Knapp, Angela N. Lam, Phoebe Luther, George Matrai, Patricia Measures, Chris Nicholson, David Paytan, Adina Pellenbarg, Robert Popendorf, Kim Reddy, Chris Ruttenberg, Kathleen Sabine, Chris Sansone, Frank Shaltout, Nayrah Sikes, Elisabeth Sundquist, Eric Valentine, David Wang, Zhaohui Aleck Wilson, Sam Barrett, Pamela Behrens, Melanie Belcher, Anna Biermann, Lauren Boiteau, Rene Clarke, Jennifer Collins, James Coppola, Alysha Ebling, Alina Garcia-Tigreros, Fenix Goldman, Johanna Guallart, Elisa F. Haskell, William Hurley, Sarah Janssen, David Johnson, Winn Lennartz, Sinikka Liu, Shuting Palevsky, hilary Rahman, Shaily Ray, Daisy Sarkar, Amit Steiner, Zvika Widner, Brittany Yang, Bo Monterey Bay Aquarium Research Institute Moss LandingCA95039 United States Marine Chemistry and Geochemistry Department Woods Hole Oceanographic Institution Woods HoleMA02543 United States Scripps Institution of Oceanography University of California San Diego San DiegoCA92037 United States School of Oceanography University of Washington SeattleWA98195 United States GEOMAR Helmholtz Centre for Ocean Research Kiel24148 Germany Department of Oceanography University of Cape Town Cape Town7700 South Africa Department of Marine Sciences University of Georgia AthensGA30602 United States Scripps Institution of Oceanography UCSD United States LDEO Columbia University United States Gordon & Betty Moore Foundation United States Massachusetts Institute of Technology United States University of Georgia United States Virginia Institute of Marine Science United States Woods Hole Oceanographic Institution United States Old Dominion University United States Stanford University United States RSMAS University of Miami United States Bermuda Institute of Ocean Sciences United States Skidmore College United States Monterey Bay Aquarium Research Inst. United States University of Cape Town South Africa Dalhousie University Canada University of Hawaii United States GEOMAR Germany Oregon State University United States University of Southampton NOC United Kingdom Windward Community College HI United States University of Southern Mississippi United States Georgia Tech United States University of Washington United States Florida State University United States University of California Santa Cruz United States University of Delaware United States Bigelow Laboratory for Ocean Sciences United States College of the Desert United States Pacific Marine Environ. Lab. NOAA United States National Inst. of Oceanography & Fisheries Egypt Rutgers University United States USGS Woods Hole Science Center United States University of California Santa Barbara United States Max Planck Insti

The questions that chemical oceanographers prioritize over the coming decades, and the methods we use to address these questions, will define our field's contribution to 21st century science. In recognition of this, the U.S. National Science Foundation and National Oceanic and Atmospheric Administration galvanized a community effort (the Chemical Oceanography MEeting: A BOttom-up Approach to research Directions, or COME ABOARD) to synthesize bottom-up perspectives on selected areas of research in Chemical Oceanography. Representing only a small subset of the community, COME ABOARD participants did not attempt to identify targeted research directions for the field. inst.ad, we focused on how best to foster diverse research in Chemical Oceanography, placing emphasis on the following themes: strengthening our core chemical skillset;expanding our tools through collaboration with chemists, engineers, and computer scientists;considering new roles for large programs;enhancing interface research through interdisciplinary collaboration;and expanding ocean literacy by engaging with the public. For each theme, COME ABOARD participants reflected on the present state of Chemical Oceanography, where the community hopes to go and why, and actionable pathways to get there. A unifying concept among the discussions was that dissimilar funding structures and metrics of success may be required to accommodate the various levels of readiness and stages of knowledge development found throughout our community. In addition to the science, participants of the concurrent Dissertations Symposium in Chemical Oceanography (DISCO) XXV, a meeting of recent and forthcoming Ph.D. graduates in Chemical Oceanography, provided perspectives on how our field could show leadership in addressing long-standing diversity and early-career challenges that are pervasive throughout science. Here we summarize the COME ABOARD Meeting discussions, providing a synthesis of reflections and perspectives on the field.

关键词： Oceanography

来源：评论

学校读者我要写书评

暂无评论

26th Annual Computational Neuroscience Meeting (CNS*2017): Part 3 Antwerp, Belgium. 15-20 July 2017 Abstracts

引用

BMC NEUROSCIENCE 2017年第SUPPL 1期18卷 95-176页

作者： [Anonymous] Department of Neuroscience Yale University New Haven CT 06520 USA Department Physiology & Pharmacology SUNY Downstate Brooklyn NY 11203 USA NYU School of Engineering 6 MetroTech Center Brooklyn NY 11201 USA Departament de Matemàtica Aplicada Universitat Politècnica de Catalunya Barcelona 08028 Spain Institut de Neurobiologie de la Méditerrannée (INMED) INSERM UMR901 Aix-Marseille Univ Marseille France Center of Neural Science New York University New York NY USA Aix-Marseille Univ INSERM INS Inst Neurosci Syst Marseille France Laboratoire de Physique Théorique et Modélisation CNRS UMR 8089 Université de Cergy-Pontoise 95300 Cergy-Pontoise Cedex France Department of Mathematics and Computer Science ENSAT Abdelmalek Essaadi’s University Tangier Morocco Laboratory of Natural Computation Department of Information and Electrical Engineering and Applied Mathematics University of Salerno 84084 Fisciano SA Italy Department of Medicine University of Salerno 84083 Lancusi SA Italy Dipartimento di Fisica Università degli Studi Aldo Moro Bari and INFN Sezione Di Bari Italy Data Analysis Department Ghent University Ghent Belgium Coma Science Group University of Liège Liège Belgium Cruces Hospital and Ikerbasque Research Center Bilbao Spain BIOtech Department of Industrial Engineering University of Trento and IRCS-PAT FBK 38010 Trento Italy Department of Data Analysis Ghent University Ghent 9000 Belgium The Wellcome Trust Centre for Neuroimaging University College London London WC1N 3BG UK Department of Electronic Engineering NED University of Engineering and Technology Karachi Pakistan Blue Brain Project École Polytechnique Fédérale de Lausanne Lausanne Switzerland Departement of Mathematics Swansea University Swansea Wales UK Laboratory for Topology and Neuroscience at the Brain Mind Institute École polytechnique fédérale de Lausanne Lausanne Switzerland Institute of Mathematics University of Aberdeen Aberdeen Scotland UK Department of Integrativ

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：