检索结果-内蒙古大学图书馆

Comments on "Comparative analysis of backpropagation and the extended Kalman filter for training multilayer perceptrons"

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 1994年第8期16卷 862-863页

作者： P. Sarat Chandran Centre for Signal Processing School of Electrical and Electronic Engineering Nanyang Technological University Singapore

In this note, the connection between the backpropagation algorithm and the extended Kalman filter is analyzed using an alternate form of representation for the Kalman gain term and shown to be much simpler than that r... 详细信息

关键词： Multi-layer neural network Kalman filters backpropagation algorithms Neural networks Equations Covariance matrix Algorithm design and analysis State estimation Nonhomogeneous media Feedforward neural networks

来源：评论

学校读者我要写书评

暂无评论

The effects of analog hardware properties on backpropagation networks with on-chip learning

The effects of analog hardware properties on backpropagation...

引用

International Conference on Neural Networks

作者： B.K. Dolenko H.C. Card Department of Electrical and Computer Engineering University of Manitoba Winnipeg MAN Canada

Results of simulations performed assuming both forward and backward computation done on-chip using analog components are presented. Aspects of analog hardware studied are component variability (variability in multiplier gains and zero offsets), limited voltage ranges, and components (multipliers) that only approximate the computations in the backpropagation algorithm. It is shown that backpropagation networks can learn to compensate for all these shortcomings of analog circuits except for zero offsets. Variability in multiplier gains is not a problem, and learning is still possible despite limited voltage ranges and function approximations. Fixed component variation from fabrication is shown to be less detrimental to learning than component variation due to noise.< >

关键词： Network-on-a-chip Computer networks backpropagation algorithms Analog computers Neural network hardware Voltage Neural networks Silicon Computational modeling Analog circuits

来源：评论

学校读者我要写书评

暂无评论

Efficient symmetric comparison-based self-diagnosis using backpropagation artificial neural networks

Efficient symmetric comparison-based self-diagnosis using ba...

引用

IEEE International Conference on Performance, Computing and Communications (IPCCC)

作者： Mourad Elhadef Amiya Nayak College of Engineering and Computer Science Abu Dhabi University Abu Dhabi UAE School of Information Technology and Engineering University of Ottawa Ottawa Canada

One of the main challenging problems in distributed systems is the comparison-based self-diagnosis which aims at identifying the set faulty of nodes (or units) based on the matching and mismatching among the system's nodes. The comparison-based self-diagnosis approach assigns tasks to the nodes that need to be diagnosed. The outcomes from each pair of units performing the same task are compared, and based on such comparison their fault status is identified. In this work, we consider symmetric t-comparison-based diagnosable systems in which at most t nodes can fail permanently at the same time. We introduce a novel backpropagation neural network-based (BPNN) approach to implement a new fault identification algorithm. The BPNN diagnosis algorithm uses the comparison approach to collect the agreements and disagreements among the nodes, and then uses these comparison outcomes to identify which nodes are faulty and which ones are fault-free. The BPNN-based diagnosis algorithm needs first to undergo an extensive training phase using various faulty situations. Results from a thorough simulation study demonstrate the effectiveness of the BPNN-based self diagnosis algorithm for randomly generated diagnosable systems, making it a viable addition to existing diagnosis algorithms. The novel approach is shown to scale very well for large systems.

关键词： Artificial neural networks Fault diagnosis Testing backpropagation algorithms Educational institutions Computer science Information technology Neural networks Performance evaluation Wireless communication

来源：评论

学校读者我要写书评

暂无评论

Functional site prediction on the DNA sequence by artificial neural networks

Functional site prediction on the DNA sequence by artificial...

引用

IEEE International Joint Symposia on Intelligence and Systems

作者： Hatzigeorgiou, A Mache, N Reczko, M European Molecular Biology Lab Heidelberg Germany

ISBN: (纸本)0818677287

A modular system of neural networks is used to identify genes in DNA sequences of eukaryotic organisms. The identification task is decomposed into the detection of distinct signals using separate neural network modules. Such signals are coding regions, splice sites and transcription start regions (cap-site). A focus of the work is the use of Back-percolation, cascade correlation, and Time-Delay neural networks. These give, in this in this particular application, better generalization than the well known backpropagation algorithm. This system achieves a prediction accuracy comparable to the traditionally design ed gene identification packages and is able to produce more accurate protein sequences from the constructed gene structures.

关键词： DNA Sequence Neural network Modular systems backpropagation algorithms Gene Annotation Genes signals

来源：评论

学校读者我要写书评

暂无评论

Continuous-wave terahertz diffraction tomography 46

Continuous-wave terahertz diffraction tomography

引用

46th International Conference on Infrared, Millimeter and Terahertz Waves (IRMMW-THz)

作者： Jin, Xiaoyu Wang, Dayong Zhao, Jie Wang, Yunxin Rong, Lu Beijing Univ Technol Fac Sci Beijing 100124 Peoples R China Beijing Engn Res Ctr Precis Measurement Technol & Beijing 100124 Peoples R China

ISBN: (纸本)9781728194240

We implement diffraction tomography using a continuous-wave terahertz (CW- THz) source. An off-axis digital holographic interferometry system was built to measure the complex amplitude distributions of a sample for various rotation angles. Based on the Rytov approximation, the three-dimensional (3D) refractive index (RI) distributions of the polystyrene foam spheres was reconstructed by the filtered backpropagation algorithm.

关键词： Three-dimensional displays Diffraction Refractive index backpropagation algorithms Tomography Filtering algorithms Rotation measurement

来源：评论

学校读者我要写书评

暂无评论

Convergence of Gradient Descent Algorithm for Diagonal Recurrent Neural Networks

Convergence of Gradient Descent Algorithm for Diagonal Recur...

引用

2nd International Conference on Bio-Inspired Computing

作者： Xu, Dongpo Li, Zhengxue Wu, Wei Ding, Xiaoshuai Qu, Di Dalian Univ Technol Dept Appl Math Dalian 116024 Peoples R China

ISBN: (纸本)9781424441051

Recurrent neural networks have been used for analysis and prediction of time series. This paper is concerned with the convergence of the gradient descent algorithm for training the diagonal recurrent neural networks. The existing convergence results consider the online gradient training algorithm based on the assumption that a very large number of (or infinitely many in theory) training samples of the time series are available, and accordingly the stochastic process theory is used to establish some convergence results of probability nature. In this paper, we consider the case that only a small number of training samples of the time series are available such that the stochastic treatment of the problem is no longer appropriate. Instead, we use the offline gradient descent algorithm for training the diagonal recurrent neural network, and we accordingly prove some convergence results of deterministic nature. The monotonicity of the error function in the iteration is also guaranteed.

关键词： Convergence Recurrent neural networks Neurons Neural networks Stochastic processes Time series analysis backpropagation algorithms Mathematics Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

A Fuzzy Neural Network Model and Its Hardware Implementation

引用

IEEE Transactions on Fuzzy Systems 1993年第3期1卷 171-183页

作者： Kuo, Yau-Hwang Kao, Cheng-I. Chen, Jiahn-Jung Institute of Information Engineering National Cheng Kung University Tainan 70101 Taiwan Institute for Information Industry Taipei Taiwan Computer and Communication Research Laboratory of Industrial Technology Research Institute Hsinchu Taiwan

In this paper, a fuzzy classifier based on a four-layered feedforward neural network model is proposed. This connectionist fuzzy classifier, called CFC, realizes the “weighted Euclidean distance” fuzzy classification concept in a massively parallel manner to recognize input patterns. CFC employs a hybrid supervised/unsupervised learning scheme to organize referenced pattern vectors. This scheme not only overcomes the major drawbacks of the multilayer perceptron models with the back-propagation algorithm: local minimal problem and long training time, but also avoids the disadvantage of the huge storage space requirement of the probabilistic neural network (PNN). According to experimental results, CFC shows better accuracy on speech recognition than several existing methods, even in a noisy environment. Moreover, it has a higher stability on recognition rates for different environmental conditions. To implement CFC with digital hardware, a massively parallel hardware architecture is developed. In this architecture, the techniques of bus-oriented multiprocessor, systolic processor structure, and pipelining are applied to provide a low-cost, high-performance fuzzy classification scheme. The systolic VLSI processor for membership grade computation has been modeled with the VHDL language and synthesized by a high-level synthesis tool. Fuzzy neural network, fuzzy classification, hierarchical clustering, speech recognition, massively parallel hardware, systolic VLSI processor, VHDL language, high-level synthesis. © 1993 IEEE

关键词： Fuzzy neural networks Neural network hardware Neural networks Feedforward neural networks Pattern recognition Unsupervised learning Multilayer perceptrons backpropagation algorithms Multi-layer neural network Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Neural network-based detection of mechanical, sensor and biological faults in deep-trough hydroponics

Neural network-based detection of mechanical, sensor and bio...

引用

4th International Workshop on Artificial Intelligence in Agriculture

作者： Ferentinos, KP Albright, LD Selman, B Cornell Univ Dept Biol & Environm Engn Ithaca NY 14853 USA Cornell Univ Dept Comp Sci Ithaca NY 14853 USA

In this work, two separate fault detection models are developed: one for the detection of faulty operation of a deep-trough hydroponic system which is caused by mechanical, actuator or sensor faults, and one for the detection of a category of biological faults (i.e. specific stressed situations of the plants), namely the "transpiration fault". The neural network methodology was proved to be successful in the task of fault detection, in both applications. The general fault detection model was capable of detecting a faulty situation in a very short time, in most cases within 20 or 40 min. In the case of the biological fault detection model, the "transpiration fault" was generally detected within 2-3 h. The results show that both neural networks have useful generalization capabilities. The influences of the conditions of the plants to the measured root zone variables are also investigated and they show that biological faults in general cannot be detected in this kind of cultivation systems using the measurable variables used in this work. The most probable explanation is the inertia of overwhelming mass of the nutrient solution (compared to the mass of the plants), in a way that it becomes a limitation factor to the influence of plants condition to their root zone microenvironment. (C) 2003 Elsevier Science B.V. All rights reserved.

关键词： fault detection hydroponics feedforward neural networks biological faults backpropagation algorithms

来源：评论

学校读者我要写书评

暂无评论

Agent-based pattern mining of discredited activities in public services

Agent-based pattern mining of discredited activities in publ...

引用

IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT 2006)

作者： Li, Chunsheng Gao, Yatian Daqing Petr Inst Coll Comp & Informat Technol Heilongjiang 163318 Peoples R China

ISBN: (纸本)9780769527499

Along with the applications of computer and database techniques, a large even huge amount of data has been produced and collected in the field of public service. Many valuable regulations hide in the raw data. It is necessary that new technique is studied to help the manager make proper decision. We propose a data mining model which is based on multi-agent technique to mine the discredit activities in the water sale public service. The Multilayer feed-forward neural network (MFNN) trained by the improved backpropagation (BP) algorithm and decision tree algorithm have been employed in the model. Five agents have been developed to mine the discredit patterns and evaluate the users' credit according to the model. A case study of Daqing Oilfield water supplying payment system has been implemented for verifying this model.

关键词： Multi-layer neural network backpropagation algorithms Application software Computer applications Databases Data mining Marketing and sales Feedforward systems Neural networks Feedforward neural networks

来源：评论

学校读者我要写书评

暂无评论

Training simultaneous recurrent neural network with resilient propagation for static optimization.

引用

International journal of neural systems 2002年第3-4期12卷 203-218页

作者： Serpen, Gursel Corra, Joel Electrical Engineering and Computer Science The University of Toledo Toledo 43606 OH United States

This paper proposes a non-recurrent training algorithm, resilient propagation, for the Simultaneous Recurrent Neural network operating in relaxation-mode for computing high quality solutions of static optimization problems. Implementation details related to adaptation of the recurrent neural network weights through the non-recurrent training algorithm, resilient backpropagation, are formulated through an algebraic approach. Performance of the proposed neuro-optimizer on a well-known static combinatorial optimization problem, the Traveling Salesman Problem, is evaluated on the basis of computational complexity measures and, subsequently, compared to performance of the Simultaneous Recurrent Neural network trained with the standard backpropagation, and recurrent backpropagation for the same static optimization problem. Simulation results indicate that the Simultaneous Recurrent Neural network trained with the resilient backpropagation algorithm is able to locate superior quality solutions through comparable amount of computational effort for the Traveling Salesman Problem.

关键词： Recurrent neural network Resilient propagation Traveling salesman problem Neural network training Static optimization Nonlinear neural network dynamics backpropagation Recurrent backpropagation static optimization backpropagation algorithms travelling salesman problems back propagation recurrent neural nets

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：