检索结果-内蒙古大学图书馆

Cross-participant modelling based on joint or disjoint feature selection: an fMRI conceptual decoding study

Applied Informatics 2014年第1期1卷 1-21页

作者： Hiroyuki Akama Brian Murphy Miao Mei Lei Massimo Poesio Graduate School of Decision Science and Technology Tokyo Institute of Technology Meguro-ku Tokyo Japan Knowledge & Data Engineering EEECS Queen's University Belfast Belfast Northern Ireland Machine Learning Department Carnegie Mellon University Pittsburgh USA Centre for Mind/Brain Sciences University of Trento Rovereto Italy School of Computer Science and Electronic Engineering University of Essex Colchester UK

Multivariate classification techniques have proven to be powerful tools for distinguishing experimental conditions in single sessions of functional magnetic resonance imaging (fMRI) data. But they are vulnerable to a considerable penalty in classification accuracy when applied across sessions or participants, calling into question the degree to which fine-grained encodings are shared across subjects. Here, we introduce joint learning techniques, where feature selection is carried out using a held-out subset of a target dataset, before training a linear classifier on a source dataset. Single trials of functional MRI data from a covert property generation task are classified with regularized regression techniques to predict the semantic class of stimuli. With our selection techniques (joint ranking feature selection (JRFS) and disjoint feature selection (DJFS)), classification performance during cross-session prediction improved greatly, relative to feature selection on the source session data only. Compared with JRFS, DJFS showed significant improvements for cross-participant classification. And when using a groupwise training, DJFS approached the accuracies seen for prediction across different sessions from the same participant. Comparing several feature selection strategies, we found that a simple univariate ANOVA selection technique or a minimal searchlight (one voxel in size) is appropriate, compared with larger searchlights.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Diverse subgroup set discovery

Diverse subgroup set discovery

引用

作者： Van Leeuwen, Matthijs Knobbe, Arno Machine Learning Department of Computer Science Katholieke Universiteit Leuven Leuven Belgium Algorithmic Data Analysis Department of Information and Computer Sciences Universiteit Utrecht Utrecht Netherlands Leiden Institute of Advanced Computer Science Universiteit Leiden Leiden Netherlands

Large data is challenging for most existing discovery algorithms, for several reasons. First of all, such data leads to enormous hypothesis spaces, making exhaustive search infeasible. Second, many variants of essentially the same pattern exist, due to (numeric) attributes of high cardinality, correlated attributes, and so on. This causes top-k mining algorithms to return highly redundant result sets, while ignoring many potentially interesting results. These problems are particularly apparent with subgroup discovery (SD) and its generalisation, exceptional model mining. To address this, we introduce subgroup set discovery: one should not consider individual subgroups, but sets of subgroups. We consider three degrees of redundancy, and propose corresponding heuristic selection strategies in order to eliminate redundancy. By incorporating these (generic) subgroup selection methods in a beam search, the aim is to improve the balance between exploration and exploitation. The proposed algorithm, dubbed DSSD for diverse subgroup set discovery, is experimentally evaluated and compared to existing approaches. For this, a variety of target types with corresponding datasets and quality measures is used. The subgroup sets that are discovered by the competing methods are evaluated primarily on the following three criteria: (1) diversity in the subgroup covers (exploration), (2) the maximum quality found (exploitation), and (3) runtime. The results show that DSSD outperforms each traditional SD method on all or a (non-empty) subset of these criteria, depending on the specific setting. The more complex the task, the larger the benefit of using our diverse heuristic search turns out to be. © The Author(s) 2012.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

Trace chemical and major constituents measurements of the international space station atmosphere by the vehicle cabin atmosphere monitor

Trace chemical and major constituents measurements of the in...

引用

42nd International Conference on Environmental Systems 2012, ICES 2012

作者： Darrach, M.R. Chutjian, A. Bornstein, B.J. Croonquist, A.P. Garkanian, V. Hofman, J. Karmon, D. Kenny, J. Kidd, R.D. Lee, S. Macaskill, J.A. Madzunkov, S.M. Mandrake, L. Schaefer, R.T. Toomarian, N. Jet Propulsion Laboratory California Institute of Technology Pasadena CA 91109 United States Atomic and Molecular Physics Group JPL/Caltech Pasadena CA 91109 United States Instrument Autonomy Group JPL/Caltech Pasadena CA 91109 United States MicroDevices Group JPL/Caltech Pasadena CA 91109 United States Optical Communications Group JPL/Caltech Pasadena CA 91109 United States Instruments and Science Data Systems JPL/Caltech Pasadena CA 91109 United States Instrument Integration and Test JPL/Caltech Pasadena CA 91109 United States Planetary Chemistry and Astrobiology JPL/Caltech Pasadena CA 91109 United States High Capacity Computing and Modeling JPL/Caltech Pasadena CA 91109 United States Machine Learning and Instrument Autonomy Group JPL/Caltech Pasadena CA 91109 United States Advanced Computer Systems and Technology JPL/Caltech Pasadena CA 91109 United States Advanced Instrument Concepts JPL/Caltech Pasadena CA 91109 United States

ISBN: (纸本)9781600869341

We report on trace gas and major atmospheric constituents results obtained by the Vehicle Cabin Atmosphere Monitor (VCAM) following almost two years of operation aboard the International Space Station (ISS). VCAM isan autonomous environmental monitor based on a highly compact gas chromatograph/quadrupole ion trap mass spectrometer. It was flown to the International Space Station (ISS) on shuttle mission STS-131 and commenced operations on June 2010. VCAM is capable of providing measurements of both parts-per-billion (ppb) levels of volatile trace-gas constituents, and of the atmospheric major constituents (nitrogen, oxygen, argon, and carbon dioxide) in a space vehicle or station. It is designed to operate autonomously and maintenance-free, approximately once per day, with a self-contained gas supply sufficient for a one-year lifetime. VCAM's performance is sufficient to detect and identify 90% of the target compounds at their 180-day Spacecraft Maximum Allowable Concentration levels. © 2012 by the American institute of Aeronautics and Astronautics, Inc.

关键词： Gases

来源：评论

学校读者我要写书评

暂无评论

On-orbit measurements of the ISS atmosphere by the vehicle cabin atmosphere monitor

On-orbit measurements of the ISS atmosphere by the vehicle c...

引用

41st International Conference on Environmental Systems 2011, ICES 2011

作者： Darrach, M.R. Chutjian, A. Bornstein, B.J. Croonquist, A.P. Garkanian, V. Haemmerle, V.R. Hofman, J. Heinrichs, W.M. Karmon, D. Kenny, J. Kidd, R.D. Lee, S. MacAskill, J.A. Madzunkov, S.M. Mandrake, L. Rust, T.M. Schaefer, R.T. Thomas, J.L. Toomarian, N. Jet Propulsion Laboratory California Institute of Technology Pasadena CA 91109 United States Atomic and Molecular Physics Group JPL/Caltech United States Instrument Autonomy Group JPL/Caltech United States MicroDevices Group JPL/CalTech United States Optical Communications Group JPL/Caltech United States Processing Algorithms and Calibration Engineering JPL/Caltech United States Inst System Engineering JPL/Caltech United States Instruments and Science Data Systems JPL/Caltech United States Instrument Integration and Test JPL/Caltech United States Planetary Chemistry and Astrobiology JPL/Caltech United States High Capacity Computing and Modeling JPL/Caltech United States Machine Learning and Instrument Autonomy Group JPL/Caltech United States SpaceX 1 Rocket Rd Hawthorne CA 90250 United States Advanced Computer Systems and Technology JPL/Caltech United States Advanced Instrument Concepts JPL/Caltech United States

ISBN: (纸本)9781600869488

We report on trace gas and major atmospheric constituents results obtained by the Vehicle Cabin Atmosphere Monitor (VCAM) during operations aboard the International Space Station (ISS). VCAM is an autonomous environmental monitor based on a miniature gas chromatograph/mass spectrometer. It was flown to the ISS on shuttle mission STS-131 and commenced operations on 6/10/10. VCAM provides measurements of ppb-to-ppm levels of volatile trace-gas constituents, and of the atmospheric major constituents (nitrogen, oxygen, argon, and carbon dioxide) in a space vehicle or station. It is designed to operate autonomously and maintenance-free, approximately once per day, with a self-contained gas supply sufficient for a one-year lifetime. VCAM is designed to detect and identify 90% of the target compounds at their 180-day Spacecraft Maximum Allowable Concentration levels. © 2011 by the American institute of Aeronautics and Astronautics, Inc.

关键词： Gases

来源：评论

学校读者我要写书评

暂无评论

CONTROL SYSTEM DESIGN CONSIDERING A TRADEOFF BETWEEN EVALUATED UNCERTAINTY RANGES AND CONTROL PERFORMANCE

引用

Asian Journal of Control 2008年第1期1卷

作者： Y. Wakasa Y. Yamamoto Dept. of Applied Analysis and Complex Dynamical Systems Graduate School of Informatics Kyoto University Kyoto Japan. Yuji Wakasa was born in Okayama Japan in 1968. He received the B.S. and M.S. degrees in engineering from Kyoto university Japan in 1992 and 1994 respectively. From 1994 to 1998 he was a Research Associate in the Department of Information Technology Okayama University. Since April 1998 he has been a Research Associate in the Graduate School of Informatics Kyoto University. His current research interests include robust control and control system design via mathematical programming. Yutaka Yamamoto received his B.S. and M.S. degrees in engineering from Kyoto University Kyoto Japan in 1972 and 1974 respectively and the M.S. and Ph.D. degree in mathematics from the University of Florida in 1976 and 1978 respectively. From 1978 to 1987 he was with Department of Applied Mathematics and Physics Kyoto University and from 1987 to 1997 with Department of Applied System Science. Since 1998 he is a professor at the current position. His current research interests include realization and robust control of distributed parameter systems learning control sampled-data systems and digital signal processing. Dr. Yamamoto is a receipient of the Sawaragi memorial paper award (1985) the Outstanding Paper Award of SICE (1987) Best Author Award of SICE (1990) the George Axelby Outstanding Paper Award of IEEE CSS in 1996 Takeda Paper Prize of SICE in 1997. He is a Fellow of IEEE. He was an associate editor of Automatica. He is currently an associate editor of IEEE Transactions on Automatic Control Systems and Control Letters and Mathematics of Control Signals and Systems. He is a member of the IEEE the Society of Instrument and Control Engineers (SICE) and the Institute of Systems Control and Information Engineers.

This paper presents a design method of control systems such that a designer can flexibly take account of tradeoffs between evaluated uncertainty ranges and the level of control performance. The problem is reduced to a BMI problem and approximately solved by LMIs. The convergence of the proposed approximation is proved in a modified sense. A numerical example shows the effectiveness of the proposed method in comparison with the standard robust control.

关键词： Control system design LMI BMI tradeoff curves

来源：评论

学校读者我要写书评

暂无评论

data mining as a tool for environmental scientists

Data mining as a tool for environmental scientists

引用

3rd Biennial Meeting of the International Environmental Modelling and Software Society: Summit on Environmental Modelling and Software, iEMSs 2006

作者： Spate, Jessica Gibert, Karina Sànchez-Marrè, Miquel Frank, Eibe Comas, Joaquim Athanasiadis, Ioannis Letcher, Rebecca Mathematical Sciences Institute Australian National University Canberra Australia Department of Statistics and Operation Research Technical University of Catalonia Barcelona Catalonia Spain Knowledge Engineering and Machine Learning Group Technical University of Catalonia Barcelona Catalonia Spain Department of Computer Science University of Waikato Waikato New Zealand University of Girona Girona Catalonia Spain Istituto Dalle Molle di Studi Sull'Intelligenza Artificiale Lugano Switzerland Integrated Catchment Assessment and Management Centre Australian National University Canberra Australia

ISBN: (纸本)1424308526

Over recent years a huge library of data mining algorithms has been developed to tackle a variety of problems in fields such as medical imaging and network traffic analysis. Many of these techniques are far more flexible than more classical modelling approaches and could be usefully applied to data-rich environmental problems. Certain techniques such as Artificial Neural Networks, Clustering, Case-Based Reasoning and more recently Bayesian Decision Networks have found application in environmental modelling while other methods, for example classification and association rule extraction, have not yet been taken up on any wide scale. We propose that these and other data mining techniques could be usefully applied to difficult problems in the field. This paper introduces several data mining concepts and briefly discusses their application to environmental modelling, where data may be sparse, incomplete, or heterogenous.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

CLEER - AN AI-SYSTEM DEVELOPED TO ASSIST EQUIPMENT ARRANGEMENTS ON WARSHIPS

引用

NAVAL ENGINEERS JOURNAL 1989年第3期101卷 127-137页

作者： ZHOU, HH SILVERMAN, BG SIMKOL, J Dr. H. Harry Zhou:is a research professor at the Institute for Artificial Intelligence of The George Washington University. Dr. Zhou received his master's degree and Ph.D. in computer science from Vanderbilt University in 1984 and 1987 respectively. He did his dissertation in the fields of artificial intelligence analogical reasoning and machine learning. His research interests include: classifier systems genetic algorithms learning by analogy inductive learning adaptive expert systems automated knowledge acquisition and adaption. He is also interested in data base design programming languages mental modeling and software engineering. Dr. Barry G. Silverman:is director of the Institute for Artificial Intelligence and a professor at the Engineering Administration Department of The George Washington University. He is also president of IntelliTek Inc. an AI consulting firm. Dr. Silverman received the B.S.E. M.S.E. and Ph.D. degrees from the University of Pennsylvania. He has been a principal developer of four generic AI products as well as eight AI applications. Since 1979 he has written over 100 papers and reports on these AI efforts. Joel Simkol:is a research scientist currently engaged in designing expert system architectures to support electronic warfare vulnerability analyses threat assessments shipboard topside antenna arrangements and C3countermeasures employment. Mr. Simkol's work in applying expert systems technology to electromagnetic interference and to spectrum management has generated increased interest and participation from all branches of government agencies.

This paper describes a modularized AI system being built to help improve electromagnetic compatibility (EMC) among shipboard topside equipment and their associated systems. CLEER is intended to act as an easy to use integrator of existing expert knowledge and pre-existing data bases and large scale analytical models. Due to these interfaces; to the need for portability of the software; and to artificial intelligence related design requirements (such as the need for spatial reasoning, expert data base management, model base management, track-based reasoning, and analogical (similar ship) reasoning) it was realized that traditional expert system shells would be inappropriate, although relatively off-the-shelf AI technology could be incorporated. In the same vein, the rapid prototyping approach to expert system design and knowledge engineering was not pursued in favor of a rigorous systems engineering methodology. The critical design decisions affecting CLEER's development are summarized in this paper along with lessons learned to date all in terms of “how,” “why,” and “when” specific features are being developed.

关键词： Artificial Intelligence

来源：评论

学校读者我要写书评

暂无评论

Computational Neuroscience: mathematical and Statistical Perspectives

引用

Statistics and Its Application 1000年第1期5卷 183-214页

作者： Robert E. Kass Shun-Ichi Amari Kensuke Arai Emery N. Brown Casey O. Diekman Markus Diesmann Brent Doiron Uri T. Eden Adrienne L. Fairhall Grant M. Fiddyment Tomoki Fukai Sonja Grün Matthew T. Harrison Moritz Helias Hiroyuki Nakahara Jun-nosuke Teramae Peter J. Thomas Mark Reimers Jordan Rodu Horacio G. Rotstein Eric Shea-Brown Hideaki Shimazaki Shigeru Shinomoto Byron M. Yu Mark A. Kramer 1Department of Statistics Machine Learning Department and Center for the Neural Basis of Cognition Carnegie Mellon University Pittsburgh Pennsylvania 15213 USA email: kass@stat.cmu.edu 2Mathematical Neuroscience Laboratory RIKEN Brain Science Institute Wako Saitama Prefecture 351-0198 Japan 3Department of Mathematics and Statistics Boston University Boston Massachusetts 02215 USA 5Department of Anesthesia Harvard Medical School Boston Massachusetts 02115 USA 4Department of Brain and Cognitive Sciences Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA 6Department of Mathematical Sciences New Jersey Institute of Technology Newark New Jersey 07102 USA 8Department of Theoretical Systems Neurobiology Institute of Biology RWTH Aachen University 52062 Aachen Germany 7Institute of Neuroscience and Medicine Jülich Research Centre 52428 Jülich Germany 9Department of Mathematics University of Pittsburgh Pittsburgh Pennsylvania 15260 USA 10Department of Physiology and Biophysics University of Washington Seattle Washington 98105 USA 11Division of Applied Mathematics Brown University Providence Rhode Island 02912 USA 12Department of Integrated Theoretical Neuroscience Osaka University Suita Osaka Prefecture 565-0871 Japan 13Department of Mathematics Applied Mathematics and Statistics Case Western Reserve University Cleveland Ohio 44106 USA 14Department of Neuroscience Michigan State University East Lansing Michigan 48824 USA 15Department of Statistics University of Virginia Charlottesville Virginia 22904 USA 17Institute for Brain and Neuroscience Research New Jersey Institute of Technology Newark New Jersey 07102 USA 16Federated Department of Biological Sciences Rutgers University/New Jersey Institute of Technology Newark New Jersey 07102 USA 19Department of Physics Kyoto University Kyoto Kyoto Prefecture 606-8502 Japan 18Honda Research Institute Japan Wako Saitama Prefecture 351-0188 Japan 20Department of Electrical and Compute

mathematical and statistical models have played important roles in neuroscience, especially by describing the electrical activity of neurons recorded individually, or collectively across large networks. As the field moves forward rapidly, new challenges are emerging. For maximal effectiveness, those working to advance computational neuroscience will need to appreciate and exploit the complementary strengths of mechanistic theory and the statistical paradigm.

关键词： neural data analysis neural modeling neural networks theoretical neuroscience

来源：评论

学校读者我要写书评

暂无评论

Structured Representation learning 1

Structured Representation Learning

引用

丛书名： Synthesis Lectures on Computer Vision

1000年

作者： Yue Song Thomas Anderson Keller Nicu Sebe Max Welling

ISBN: (数字)9783031881114

ISBN: (纸本)9783031881107;9783031881138

This book introduces approaches to generalize the benefits of equivariant deep learning to a broader set of learned structures through learned homomorphisms. In the field of machine learning, the idea of incorporating knowledge of data symmetries into artificial neural networks is known as equivariant deep learning and has led to the development of cutting edge architectures for image and physical data processing. The power of these models originates from data-specific structures ingrained in them through careful engineering. To-date however, the ability for practitioners to build such a structure into models is limited to situations where the data must exactly obey specific mathematical symmetries. The authors discuss naturally inspired inductive biases, specifically those which may provide types of efficiency and generalization benefits through what are known as homomorphic representations, a new general type of structured representation inspired from techniques in physics and neuroscience. A review of some of the first attempts at building models with learned homomorphic representations are introduced. The authors demonstrate that these inductive biases improve the ability of models to represent natural transformations and ultimately pave the way to the future of efficient and effective artificial neural networks.

关键词： machine learning Artificial Intelligence Image Processing and Computer Vision Computer Imaging, Vision, Pattern Recognition and Graphics mathematical Models of Cognitive Processes and Neural Networks Computer science, general

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：