检索结果-内蒙古大学图书馆

International Joint Conference on Neural Networks (IJCNN)

作者： G. Auda M. Kamel Pattern Analysis And Machine Intelligence Lab Systems Design Engineering Department University of Waterloo Canada

Cooperation by voting is one of the popular modular neural network decision-making strategies. Ensemble classifiers are multiple identical modules which use voting for post-learning classification. This paper suggests a new cooperation scheme for ensembles which utilizes voting in the learning process itself. According to the suggested scheme, different modules would, automatically, focus on different regions in the input space. Hence, temporal crosstalk decreases and decision boundaries are drawn accurately in complex overlapping regions of the input space.

关键词： Voting Neural networks Decision making Multi-layer neural network Filtering Crosstalk pattern analysis machine intelligence System analysis and design design engineering

来源：评论

学校读者我要写书评

暂无评论

An optimized two-step camera calibration method

An optimized two-step camera calibration method

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： H. Bacakoglu M. Kamel Pattern Analysis and Machine Intelligence Laboratory Department of Systems Design Engineering University of Waterloo Waterloo ONT Canada

An optimized two-step camera calibration algorithm is developed. The proposed method starts with the well known linear calibration which approximates the transformation as a 3/spl times/4 matrix. Based on the results of the linear calibration and the camera model we construct the 4/spl times/4 homogeneous transformation matrix. Quaternion algebra is used to extract the optimum rotation matrix and this optimization is later extended to the other calibration parameters. Our calibration method includes nonlinear optimization which takes into consideration lens distortions. The convergence rate of the nonlinear optimization is accelerated by three more objective functions we introduced. To assess the accuracy of our proposed quaternion method, Euclidean norm of the error matrix between the original and computed homogeneous transformation matrices is calculated and compared to those of the existing methods. Simulations show that the quaternion method yields more accurate results both before and after the nonlinear optimization.

关键词： Optimization methods Cameras Calibration Quaternions Linear approximation Algebra Lenses Nonlinear distortion Convergence Acceleration

来源：评论

学校读者我要写书评

暂无评论

A rough set approach for planner evaluation

A rough set approach for planner evaluation

引用

IEEE International Conference on systems, Man and Cybernetics

作者： B.K. Sy Wang Jiaxin Pattern Analysis and Machine Intelligence Group System Design Engineering Department University of Waterloo Waterloo ONT Canada Department of Computer Science CUNY Flushing NY USA The State Key Laboratory of Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University Beijing China

This paper discusses a rough set approach for evaluating solutions of scheduling problems. Algorithms for solving scheduling problems are planners and the scheduling problems are modelled as constraint satisfaction problems. Conventional approach for the analysis of algorithms often focuses on the time and representational complexities, and assumes an identical cost on all operations. The proposed rough set approach augments conventional approaches for the analysis of algorithms in two ways: 1) it permits the consideration of different costs arising from different operations; and 2) it allows one to define a new utility for a complexity analysis.

关键词： Algorithm design and analysis Processor scheduling Set theory machine intelligence Computer science Costs Search problems Educational institutions pattern analysis System analysis and design

来源：评论

学校读者我要写书评

暂无评论

Representing Discovered patterns Using Attributed Hypergraph 2

Representing Discovered Patterns Using Attributed Hypergraph

引用

2nd International Conference on Knowledge Discovery and Data Mining, KDD 1996

作者： Wang, Yang Wong, Andrew K.C. Department of Systems Design Engineering Pattern Analysis and Machine Intelligence Group University of Waterloo WaterlooONN2L 3G1 Canada

ISBN: (纸本)1577350049

One of the fundamental problems in knowledge discovery in databases and other applications of AI is how to represent knowledge and *** representation schemes have various *** this paper, we propose a new knowledge representation scheme using attributed hypergraph (AHG), which is simple yet general enough to directly encode different order patterns discovered from large *** AHG, both the qualitative and quantitative relations are represented as attributed *** representation is lucid and transparent for ***, patterns in AHG are easy to *** the discussion, some basic manipulations of AHG for data mining tasks are briefly *** paper ends with examples of pattern representation using AHG. © 1996 AAAI (***). All Rights Reserved.

关键词： Knowledge representation

来源：评论

学校读者我要写书评

暂无评论

A connectionist model for learning robotic grasps using reinforcement learning

A connectionist model for learning robotic grasps using rein...

引用

International Conference on Neural Networks

作者： M.A. Moussa M.S. Kamel Pattern Analysis and Machine Intelligence LaboratoryDepartment of Systems Design Engineering University of Waterloo ONT Canada

We present a connectionist model for learning robotic grasps. These grasps are represented as generic grasping functions. Our approach is to learn these functions by experimentation with the environment. Grasping rules are mapped to generic representations that can then be learned by experiments using neural networks. Furthermore, grasping rules acquired in this format can then be used on different objects using different grippers. During experimentation, reinforcement learning is used to minimize the number of failed experiments. Results show that the system is able to learn how to grasp various objects while maintaining a small number of experiments.

关键词： Learning Grippers Grasping Humans Shape Neural networks Robot sensing systems pattern analysis machine intelligence System analysis and design

来源：评论

学校读者我要写书评

暂无评论

Modular neural network architectures for classification

Modular neural network architectures for classification

引用

International Conference on Neural Networks

作者： G. Auda M. Kamel H. Raafat Pattern Analysis And Machine Intelligence Laboratory Systems Design Eneering Department University of Waterloo Canada University of Waterloo Waterloo ON CA Department of Mathematics and Computer Science Kuwait University Kuwait

One of the major drawbacks of the current neural network generation is the inability to cope with the increase of size/complexity of classification tasks. Modular neural network classifiers attempt to solve this problem through a "divide and conquer" approach. However. The performance of the modular neural network classifiers is sensitive to efficiency of the "task decomposition" technique and the "multi-module decision-making" strategy. After a brief review of previous work with emphasis on five published modular classifiers-decoupled nets, ART-BP, hierarchical network, multiple experts, and multiple identical networks (majority vote and average output decisions)-this paper introduces the cooperative modular neural network (CMNN). The CMNN classifier outperforms the surveyed nets due to its novel task decomposition and multi-module decision-making techniques.

关键词： Neural networks Decision making Multi-layer neural network Computer architecture pattern analysis machine intelligence System analysis and design Mathematics Computer science Voting

来源：评论

学校读者我要写书评

暂无评论

A connectionist model of human grasps and its application to robot grasping

A connectionist model of human grasps and its application to...

引用

International Conference on Neural Networks

作者： M.A. Moussa M.S. Kamel Pattern Analysis and Machine Intelligence Laboratory Department of Systems Design Engineering University of Waterloo Waterloo ON Canada

Presents a connectionist model that can learn human grasping rules. The model domain is a taxonomy of manufacturing grasps. The input vector specifies task and object attributes such as dexterity and shape, and the model predicts the required grasp. Experiments show that the model is capable of learning grasping rules and predicting correct grasps for new objects that it has not been trained on. The authors finally discuss how this model can be used as part of a robotic system for grasping objects.

关键词： Humans Service robots Taxonomy Robot sensing systems Grippers Shape Predictive models Electrical equipment industry Robot control Neural networks

来源：评论

学校读者我要写书评

暂无评论

A segmentation criterion for digital image compression

A segmentation criterion for digital image compression

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： M.R. El-Sakka M.S. Kamel Pattern Analysis and Machine Intelligence Laboratory Systems Design Engineering Department University of Waterloo Waterloo ONT Canada University of Waterloo Waterloo ON CA Dept. of Syst. Design Eng. Waterloo Univ. Ont. Canada

This paper is concerned with segmenting light intensity images for the sake of compressing them using lossy compression techniques. Among the most commonly used techniques for image segmentation is quad-tree partitioning. In this technique, block variance based criteria are usually used to measure the smoothness of the segmented blocks and to consequently classify them. Block variance, however, does not consider the pixel value distribution within the block. Instead of using the block variance as a segmentation and classification measure, we propose using the mean squared deviation from the neighboring pixels mean. The proposed measure is capable of differentiating between blocks not only according to block pixel values but also according to their distribution within the block. This leads to a much better image segmentation and consequently to higher image compression ratios with lower image degradation. The results show the superiority of the proposed measure over the block variance measure.

关键词： Image segmentation Digital images Image coding Degradation

来源：评论

学校读者我要写书评

暂无评论

DYKOR - A METHOD FOR GENERATING THE CONTENT OF EXPLANATIONS IN KNOWLEDGE systems

引用

KNOWLEDGE-BASED systems 1994年第3期7卷 177-188页

作者： LOPEZSUAREZ, A KAMEL, M Pattern Analysis and Machine Intelligence Laboratory Department of Systems Design Engineering University of Waterloo Waterloo Ontario Canada N2L 3G1

The paper presents a methodology for improving the organization of knowledge bases and demonstrates its application for generating the content of explanations. The DyKOr (Dynamic Knowledge Organization) method combines information that is usually available through execution traces with existing domain knowledge using techniques from machine learning including knowledge compilation, explanation-based learning, and conceptual clustering. These techniques allow the separation of the knowledge needed to solve a problem from that which is not required, and the identification of information that is related to the problem but is not explicitly stated. Thus, the analysis performed through the methodology can considerably improve the quality and content of explanations. The paper describes the implementation of the methodology and how it can be integrated into typical rule-based expert systems. Illustrations of how the method can be used to produce the content for explanations are presented in the context of typical consultation and problem solving expert systems. A discussion of how the information produced by the method can be used to prepare explanations for users with different levels of expertise is also presented.

关键词： KNOWLEDGE ORGANIZATION RULE-BASED systems EXPLANATION-BASED LEARNING

来源：评论

学校读者我要写书评

暂无评论

Fuzzy clustering neural network system design and implementation

Fuzzy clustering neural network system design and implementa...

引用

Midwest Symposium on Circuits and systems (MWSCAS)

作者： D. Zhang M. Kamel M.I. Elmasry Electrical and Computer Engineering VLSI Research Group Pattern Analysis and Machine Intelligence Systems Design Engineering University of Waterloo Waterloo ONT Canada

In this paper, a novel VLSI system design methodology for fuzzy clustering neural networks (FCNN) is presented. This methodology emphasizes a coordination between model definition, architectural description, and hardware implementation, Two mapping strategies, from FCNN model to parallel architecture and from architecture to systolic implementation, are discussed. The system design is achieved by 1) developing an effective FCNN model, where a direct fuzzy competitive learning algorithm between the nodes is adopted; 2) designing the corresponding parallel architecture with special feedforward and feedback paths; 3) building the systolic array (SA) suitable for VLSI.

关键词： Fuzzy neural networks Fuzzy systems Neural networks Very large scale integration System analysis and design design engineering Hardware Clustering algorithms Systolic arrays Computational modeling

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：