检索结果-内蒙古大学图书馆

Online nonnegative matrix factorization with robust stochastic approximation

IEEE transactions on neural networks and learning systems 2012年第7期23卷 1087-99页

作者： Naiyang Guan Dacheng Tao Zhigang Luo Bo Yuan School of Computer Science National University of Defense Technology Changsha China Center for Quantum Computation & Intelligent Systems University of Technology Sydney NSW Australia Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China

Nonnegative matrix factorization (NMF) has become a popular dimension-reduction method and has been widely applied to image processing and pattern recognition problems. However, conventional NMF learning methods require the entire dataset to reside in the memory and thus cannot be applied to large-scale or streaming datasets. In this paper, we propose an efficient online RSA-NMF algorithm (OR-NMF) that learns NMF in an incremental fashion and thus solves this problem. In particular, OR-NMF receives one sample or a chunk of samples per step and updates the bases via robust stochastic approximation. Benefitting from the smartly chosen learning rate and averaging technique, OR-NMF converges at the rate of in each update of the bases. Furthermore, we prove that OR-NMF almost surely converges to a local optimal solution by using the quasi-martingale. By using a buffering strategy, we keep both the time and space complexities of one step of the OR-NMF constant and make OR-NMF suitable for large-scale or streaming datasets. Preliminary experimental results on real-world datasets show that OR-NMF outperforms the existing online NMF (ONMF) algorithms in terms of efficiency. Experimental results of face recognition and image annotation on public datasets confirm the effectiveness of OR-NMF compared with the existing ONMF algorithms.

关键词： Complexity theory Convergence Approximation methods Algorithm design and analysis Robustness Approximation algorithms Streaming media

来源：评论

学校读者我要写书评

暂无评论

Optimal limit-cycle control recast as Bayesian inference

Optimal limit-cycle control recast as Bayesian inference

引用

作者： Tassa, Yuval Erez, Tom Todorov, Emanuel Interdisciplinary Center for Neural Computation Hebrew University Jerusalem Israel Department of Computer Science Washington University St. Louis MO United States Department of Computer Science and Engineering University of Washington Seattle WA United States Department of Applied Mathematics University of Washington Seattle WA United States

ISBN: (纸本)9783902661937

We introduce an algorithm that generates an optimal controller for stochastic nonlinear problems with a periodic solution, e.g. locomotion. Uniquely, the quantity we approximate is neither the Value nor Policy functions, but rather the stationary state- distribution of the optimally-controlled process. We recast the control problem as Bayesian inference over a graphical model with a ring topology. The posterior approximates the controlled stationary distribution with local gaussians along the optimal limit-cycle. Linear-feedback gains and open-loop controls are extracted from the covariances and the means, respectively. Complexity scales linearly or quadratically with the state dimension, depending on the dynamics approximation. We demonstrate our algorithm on a toy 2-dimensional problem and then on a challenging 23-dimensional simulated walking robot. © 2011 IFAC.

关键词： Inference engines

来源：评论

学校读者我要写书评

暂无评论

A survey of topology denoise technologies

A survey of topology denoise technologies

引用

6th International Conference on computer science and Education, ICCSE 2011

作者： Ning, Zhang Tengfei, Ye Yu, Wuyi Li, Maoqing Li, Xin School of Information Science and Technology Xiamen University Xiamen Fujian China Electrical and Computer Engineering Center for Computation and Technology Louisiana State University Baton Rouge LA 70803 United States

ISBN: (纸本)9781424497188

Polygonal meshes are the most popular representation scheme for 3D surfaces in computer graphics. Polygonal meshes usually have topological and/or geometric noises during their scanning and collection steps. Many geometric computing and manipulation operations need to be performed on clean meshes. Therefore, topological noises on polygonal meshes need to be removed in a preprocessing step, which is called topology denoise. We classify a few popular topology denoise technologies and analyze their properties. © 2011 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

High-order local dynamic programming

High-order local dynamic programming

引用

IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, (ADPRL)

作者： Yuval Tassa Emanuel Todorov Interdisciplinary Center of Neural Computation Hebrew University Jerusalem Israel Applied Mathematics and Computer Science & Engineering University of Washington Seattle USA

We describe a new local dynamic programming algorithm for solving stochastic continuous Optimal Control problems. We use cubature integration to both propagate the state distribution and perform the Bellman backup. The algorithm can approximate the local policy and cost-to-go with arbitrary function bases. We compare the classic quadratic cost-to-go/linear-feedback controller to a cubic cost-to-go/quadratic policy controller on a 10-dimensional simulated swimming robot, and find that the higher order approximation yields a more general policy with a larger basin of attraction.

关键词： Approximation methods Heuristic algorithms Mathematical model Equations Trajectory Dynamic programming Noise

来源：评论

学校读者我要写书评

暂无评论

Optimal Limit-Cycle Control recast as Bayesian Inference

引用

IFAC Proceedings Volumes 2011年第1期44卷 4707-4713页

作者： Yuval Tassa Tom Erez Emanuel Todorov Interdisciplinary Center for Neural Computation Hebrew University Jerusalem Israel Department of Computer Science Washington University St. Louis MO USA Department of Computer Science and Engineering and the Department of Applied Mathematics University of Washington Seattle WA USA

Abstract We introduce an algorithm that generates an optimal controller for stochastic nonlinear problems with a periodic solution, e.g. locomotion. Uniquely, the quantity we approximate is neither the Value nor Policy functions, but rather the stationary state-distribution of the optimally-controlled process. We recast the control problem as Bayesian inference over a graphical model with a ring topology. The posterior approximates the controlled stationary distribution with local gaussians along the optimal limit-cycle. Linear-feedback gains and open-loop controls are extracted from the covariances and the means, respectively. Complexity scales linearly or quadratically with the state dimension, depending on the dynamics approximation. We demonstrate our algorithm on a toy 2-dimensional problem and then on a challenging 23-dimensional simulated walking robot.

关键词： optimal control limit cycles feedback control control-estimation dualities

来源：评论

学校读者我要写书评

暂无评论

Report on the AAAI 2010 robot exhibition

Report on the AAAI 2010 robot exhibition

引用

作者： Anderson, Monica D. Chernova, Sonia Dodds, Zachary Thomaz, Andrea L. Touretzky, David S. Computer Science Department University of Alabama United States Robotics Engineering Program Worcester Polytechnic Institute United States Department of Computer Science Harvey Mudd College United States School of Interactive Computing Georgia Institute of Technology United States Center for the Neural Basis of Cognition Carnegie Mellon University United States

The 19th robotics program at the annual AAAI conference was held in Atlanta, Georgia, in July 2010. In this article we give a summary of three components of the exhibition: the Small-Scale Manipulation Challenge: Robotic Chess;the Learning by Demonstration Challenge;and the Education Track. In each section we detail the challenge task. We also describe the participating teams, highlight the research questions they tackled, and briefly describe the systems they demonstrated. Copyright © 2011, Association for the Advancement of Artificial Intelligence. All rights reserved.

关键词： Exhibitions

来源：评论

学校读者我要写书评

暂无评论

A Survey of Topology Denoise Technologies

A Survey of Topology Denoise Technologies

引用

The 6th International Conference on computer science & Education(第六届国际计算机新科技与教育学术会议 ICCSE 2011)

作者： Maoqing Li Zhang Ning Ye Tengfei Wuyi Yu Xin Li School of Information Science and Technology Xiamen University Xiamen FujianChina Electrical & Computer Engineering Center for Computation & Technology Louisiana State University Bat

关键词： Topology Denoise Data Completion

来源：评论

学校读者我要写书评

暂无评论

An adaptive fuzzy ant colony optimization for feature selection

引用

Journal of computational Information Systems 2011年第4期7卷 1206-1213页

作者： Liu, Yuanning Wang, Gang Chen, Huiling Zhao, Zhengdong Zhu, Xiaodong Liu, Zhen College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun China Network Center Changchun University of Science and Technology Changchun 130022 China Graduate School of Engineering Nagasaki Institute of Applied Science Japan

Standard pattern classifiers perform on all data features. Whereas, some of the features are redundant or irrelevant, which reduce prediction accuracy, and increase running time of classifier. The purpose of this study is to search optimal feature subset, in order to increase the classification performance. The feature selection problem differs from traditional optimization problem on the problem size, namely the searching space or the problem dimension. Thus it is easy to converge at a local optimum by means of the ant colony optimization (ACO) based method with static parameter settings, but the adaptive parameter control strategy can balance between exploration and exploitation in the search space to intelligently improve solutions and avoid premature convergence. Only little research has been reported to adjust parameters of ACO adaptively. This paper proposes a novel adaptive fuzzy ant colony optimization (AFACO) based method for feature selection. We design two fuzzy controllers to adjust parameters of ACO, namely pheromone evaporation rate and the number of ants. The performance of the proposed method is compared with that of the Standard ACO based, Standard particle swarm optimization (PSO) based and genetic algorithm (GA) based methods on 10 benchmark data sets, taken from UCI machine learning and StatLog databases. The experimental results show that the proposed method outperforms the other methods in prediction accuracy, and is efficient and effective for feature selection problem. Copyright © 2011 Binary Information Press.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Development of planar microelectrode array for extracellular characterization of action potentials in sinoatrial node cells

Development of planar microelectrode array for extracellular...

引用

2nd International Symposium on Bioelectronics and Bioinformatics, ISBB 2011

作者： Zhang, Xin Zhang, Yulong Li, Yingxin Nelson, Richard D. LaRue, John C. Neural Engineering Research Lab. Z and L Creative Corporation Lewes DE 19958 United States Department of Electrical Engineering and Computer Science University of California Irvine CA 92697 United States Pen-Tung Sah Micro-Nano Technology Research Center Xiamen University Fujian 361005 China Department of Physiology Temple University School of Medicine Philadelphia PA 19140 United States Mechanical and Aerospace Engineering Department University of California Irvine CA 92697 United States

ISBN: (纸本)9781457700774

In this research work we prototyped a planar microelectrode arrays that utilize SU-8 as the insulating material for non-invasive chronically characterization of action potentials from SA node cells. The microelectrode arrays are fabricated to form a microchannel based extracellular action potential characterization system which comprises a 2 by 50 linear microelectrode array for acquiring the action potentials with the principle of capacitive coupling and a full customized CMOS analog bio-signal readout ASIC for bio-signal conditioning and processing. The prototyped multi-channel mixed signal fully-customized CMOS biosensor analog front-end serves as the low magnitude action potential signal readout circuitry, which is comprised of analog signal buffers, high gain amplifier and control/interface. The concept of using epoxy-based photoresist SU-8 as the insulator instead of the conventional silicon nitride or silicon dioxide is due to the high insulation property and bio-compatibility of the SU-8 material. This biocompatible planar microelectrode arrays with microchannels provides the biological scientists powerful tools for the long term non-invasive study of action potentials from SA node cells to screen libraries of chemical compounds for new drug candidates. © 2011 IEEE.

关键词： Microchannels

来源：评论

学校读者我要写书评

暂无评论

Data Mining in Earth System science (DMESS 2011)

引用

Procedia computer science 2011年 4卷 1450-1455页

作者： Forrest M. Hoffman J. Walter Larson Richard Tran Mills Bjørn-Gustaf J. Brooks Auroop R. Ganguly William W. Hargrove Jian Huang Jitendra Kumar Ranga R. Vatsavai Computational Earth Sciences Group Oak Ridge National Laboratory Oak Ridge TN USA Department of Earth System Science University of California Irvine CA USA Mathematics and Computer Science Division Argonne National Laboratory Argonne IL USA Computation Institute University of Chicago/Argonne National Laboratory Chicago IL USA School of Computer Science The Australian National University Canberra ACT AUSTRALIA Department of Electrical Engineering & Computer Science University of Tennessee Knoxville TN USA Center for Climatic Research University of Wisconsin Madison WI USA Geographic Information Science and Technology Group Oak Ridge National Laboratory Oak Ridge TN USA Eastern Forest Environmental Threat Assessment Center (EFETAC) USDA Forest Service Asheville NC USA

From field-scale measurements to global climate simulations and remote sensing, the growing body of very large and long time series Earth science data are increasingly difficult to analyze, visualize, and interpret. Data mining, information theoretic, and machine learning techniques—such as cluster analysis, singular value decomposition, block entropy, Fourier and wavelet analysis, phase-space reconstruction, and artificial neural networks—are being applied to problems of segmentation, feature extraction, change detection, model-data comparison, and model validation. The size and complexity of Earth science data exceed the limits of most analysis tools and the capacities of desktop computers. New scalable analysis and visualization tools, running on parallel cluster computers and supercomputers, are required to analyze data of this magnitude. This workshop will demonstrate how data mining techniques are applied in the Earth sciences and describe innovative computer science methods that support analysis and discovery in the Earth sciences.

关键词： Data mining Remote sensing High performance computing Segmentation Change detection Synthesis visualization

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：