检索结果-内蒙古大学图书馆

Multiagent reinforcement learning based on negotiation in general-sum games

Shanghai Jiaotong Daxue Xuebao/Journal of Shanghai Jiaotong University 2005年第SUPPL.期39卷 108-112页

作者： Zhang, Hua-Xiang Zhao, Tong Huang, Shang-Teng Information and Management School Shandong Normal Univ. Ji'nan 250014 China Dept. of Automatic Control Qingdao Univ. of Science and Technology Qingdao 266061 China Dept. of Computer Science and Eng. Shanghai Jiaotong Univ. Shanghai 200030 China

In general-sum games, multiagent cooperation has no global objective, and only individual rationality is concerned. Agent's learning is based on the assumption of opponents' policies, and this assumption may be wrong. By defining the global objective of agents, a novel multiagent reinforcement learning algorithm was proposed. All agents selected negotiated policies during learning, and punished those agents deviating from negotiated policies to ensure the execution of these policies. It was proved that the learned Q values on each stage games converge under certain restrictions. An example was given to analyze the proven result.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Weighted vector directional filters optimized by genetic algorithms

引用

5th International Conference on Parallel Processing and Applied Mathematics, PPAM 2003

作者： Lukac, Rastislav Smolka, Bogdan Swierniak, Andrzej Plataniotis, Konstantinos N. Venetsanopoulos, Anastasios N. Slovak Image Processing Center Jarkova 343 Dobsina049 25 Slovakia Department of Automatic Control Silesian University of Technology Akademicka 16 Str Gliwice44-101 Poland The Edward S. Rogers Sr. Dept. of Electrical and Computer Engineering University of Toronto 10 King’s College Road Toronto Canada

ISBN: (纸本)3540219463

This paper focuses on genetic optimization and filtering efficiency ofa recently developed class ofWeighted Vector Directional Filters (WVDFs), which minimize the aggregated weighted angular distances between the samples contained in a sliding filtering window. As shown in the paper, the proposed genetically optimized WVDFs hold excellent detail preserving characteristics and are able to remove impulsive noise affecting color images. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Impulse noise

来源：评论

学校读者我要写书评

暂无评论

Adaptive noise reduction in microarray images based on the center-weighted vector medians

引用

5th International Conference on Parallel Processing and Applied Mathematics, PPAM 2003

ISBN: (纸本)3540219463

In this paper, we provide a new noise reduction method for the enhancement of the images of gene chips. We demonstrate that the new technique is capable of reducing outliers present in microarray images while preserving the spot edges. This paper contains the comparisons of the proposed technique with the standard schemes used in color image filtering, as well as examples of their efficiency when these methods are applied to typical microarray images. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Noise abatement

来源：评论

学校读者我要写书评

暂无评论

Output impedance performance for parallel operation of UPS inverters using wireless and average current-sharing controllers

Output impedance performance for parallel operation of UPS i...

引用

Annual IEEE Conference on Power Electronics Specialists (PESC)

作者： J.M. Guerrero L.G. de Vicuna J. Miret J. Matas J. Cruz Dept. of Automatic Control Systems and Computer Eng. E. U. d'Eng. Tecnica Industrial de Barcelona Barcelona Spain Dept. of Electronics Eng. Universitat Politecnica de Catalunya Barcelona Spain Dept. of Electronics Eng. Universitat Politecnica de Catalunya

This paper encompasses the study of the output impedance impact of parallel-connected UPS inverters. Two novel nonlinear control strategies are proposed. The first one is based on the single-wire current-sharing scheme, which is well known in parallel dc-to-dc converter systems. The second one is a wireless control technique derived from the droop method. The output impedance of the inverters is investigated in both cases. Results of two parallel-connected 1-kVA UPS inverters show the feasibility of the proposed approach. Finally, the two proposed controllers are compared between them and those with the existing solutions.

关键词： Impedance measurement Uninterruptible power systems Inverters Voltage control automatic control Frequency control systems Wire Distributed control Communication system control

来源：评论

学校读者我要写书评

暂无评论

Hybrid heuristic and mathematical programming in oil pipelines networks

Hybrid heuristic and mathematical programming in oil pipelin...

引用

Congress on Evolutionary Computation

作者： J.M. de la Crnz García J.L. Risco Martín A. Herrán González P. Fernández Blanco Dept. of Computer Architecture and Automatic Control Universidad Complutense Madrid Spain C.E.S. Felipe II Ing. Téc. en Informática de Sistemas Aranjuez Spain Universidad Complutense Madrid Spain C.E.S. Felipe II Ing. Téc. en lnformática de Sistemas Aranjuez Spain

We solve the problem of the distribution of petroleum products through oil pipelines networks. This problem is modeled and solved using two techniques: A heuristic method, a multiobjective evolutionary algorithm and mathematical programming. In the multiobjective evolutionary algorithm, several objective functions are defined to express the goals of the solutions as well as the preferences among them. Some constraints are included as hard objective functions and some are evaluated through a repairing function to avoid infeasible solutions. In the mathematical programming approach the multiobjective optimization is solved using the constraint method in mixed integer linear programming. Some constraints of the mathematical model are nonlinear, so they are linearized. The results obtained with both methods for three concrete networks are presented. They are compared with a hybrid solution, where we use the results obtained by mathematical programming as the seed of the evolutionary algorithm.

关键词： Mathematical programming Intelligent networks Petroleum Pipelines Packaging Refining computer architecture automatic control Concrete Evolutionary computation

来源：评论

学校读者我要写书评

暂无评论

Modeling Demand for Inventory Management of Slow-Moving Items in Case of Reporting Errors

引用

IFAC Proceedings Volumes 2004年第4期37卷 27-32页

作者： Alexandre Dolgui Anatoly Pashkevich Maxim Pashkevich Scientific Methods for Industrial Management Dept. (MSGI) Industrial Engineering and Computer Sciences Center (G2I) Ecole de Mines de Saint Etienne 158 Cours Fauriel 42023 Saint Etienne France Robotic Laboratory Department of Automatic Control Belarusian State University of Informatics and Radioelectronics 6 P.Brovka St. Minsk 220027 Belarus Department of Mathematical Modeling and Data Analysis Faculty of Applied Mathematics and Computer Science Belarusian State University Skarina Av. 4 Minsk 220050 Belarus

The paper is devoted to the problem of modeling demand for inventory management of slow-moving items in the case of reporting errors. It is proposed a generalization of the beta-binomial demand model that takes into account possible reporting errors in the learning sample. For the new model, there are developed identification and forecasting algorithms that provide consistent estimators of the model parameters and mean square optimal forecasts. The efficiency of the proposed approach is illustrated by an application example for slow-moving car parts.

关键词： inventory control error correction statistical inference parameter estimation prediction methods

来源：评论

学校读者我要写书评

暂无评论

Mixed reality in traffic scenes: A view synthesis approach

Mixed reality in traffic scenes: A view synthesis approach

引用

IFIP 1st International Workshop on Entertainment Computing, IWEC 2002

作者： Martínez, A.B. Arboleda, J.P. Martín, E.X. Torrens, C. Dept. of Automatic Control and Computer Engineering UPC Spain

ISBN: (纸本)9781475751536

Our research group, interested in outdoor scenes, has developed a methodology to register a CAD model of a city, with images taken with video cameras installed in a car, while driving city streets. So, we can merge video captured data and the render of elements of the model into one image, forcing real and virtual camera points of view, orientations and parameters to coincide. Consequently, we generate real-video sequences and can insert, in real time, virtual objects congruently placed in relation to real objects of the scene. With the aim of increasing realistic likeness of inserted objects, we propose an alternate method that uses view synthesis techniques instead of render techniques. From a finite set of pictures of a real object, taken from different points of view, we are able to generate any point of view object projection, to be inserted in the physical coherent location. © 2003 by Springer Science+Business Media New York.

关键词： Video cameras

来源：评论

学校读者我要写书评

暂无评论

The design and application of DWT-domain optimum decoders

引用

1st International Workshop on Digital Watermarking, IWDW 2002

作者： Hu, Yongjian Kwong, Sam Chan, Y.K. Dept. of Computer Science City University of Hong Kong Kowloon Hong Kong Dept. of Automatic Control Engg South China University of Tech Guangzhou510641 China School of Info. Sci. and Tech Sun Yat-Sen University Guangzhou510275 China

ISBN: (纸本)3540012176

Based on Bayes theory of hypothesis testing, a new DWTdomain decoder structure for image watermarking has been proposed in this work. The statistical distribution of wavelet coefficients is deliberately described with the Laplacian model so that the decoding algorithm could couple effectiveness and simplicity. Under the Neyman-Pearson criterion, the decision rule is optimized by minimizing the probability of missing the watermark for a given false detection rate. Compared with other domain decoders, the proposed DWT-domain decoder has more flexibility in constructing new watermarking algorithms by using visual models that have varying spatial support. © Springer-Verlag Berlin Heidelberg 2003.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Robust fault detection using interval models

Robust fault detection using interval models

引用

European control Conference (ECC)

作者： S. Tornil T. Escobet L. Travé-Massuyès Automatic Control Dept. Technical University of Catalonia Terrassa Spain Laborarory for Analysis and Architecture of Systems Toulouse Cedex 4 France

The problem of model-based fault detection in the presence of both parametric uncertainty and noise is addressed in this paper. Intervals are used to represent the uncertainty in the system parameters and interval extensions of parity equations are used as adaptive threshold selectors. A proper combination in time of different (interval) parity equations, together with a robust indicator, is used to maintain a good sensitivity to faults whilst avoiding the effect of noise. The results in the application to the detection of different faults in a DC motor are used to show the good properties of the proposed method.

关键词： Mathematical model Robustness Noise Fault detection Uncertainty Adaptation models Wrapping

来源：评论

学校读者我要写书评

暂无评论

Application of New Type BP Neural Networks for Magnetic Measurement

引用

High Technology Letters 2002年第2期8卷 83-86页

作者：张旭 Che Rensheng Kinouchi Y Luo Xiaochuan Dept. of Automatic Measurement and Control School of Computer Science and EngineeringHarbin Institute of Technology Harbin 150001 P.R.China Dept. of Electronic Engineering University of Tokushima Japan

Magnetic measurement is a typical inverse problem in Biomedical field. In this kind of problem we always need to locate the positions and moments of one or more magnetic dipoles. Although using the traditional methods to solve this kind of inverse problem has all kinds of shortcomings, BPNN (Back Propagation Neural Networks) method can be used to solve this typical inverse problem fast enough for real time measurement. In the traditional BPNN method, gradient descent search method is performed for error propagation. In this paper the authors propose a new algorithm that Newton method is performed for error propagation. For the cost function is highly nonconvex in the magnetic measurement problem, the new kind of BPNN can get convergent results quickly and precisely. A simulation result for this method is also presented.

关键词： magnetic measurement BP neural network gradient method Newton Gauss method

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：