多层前向小世界神经网络的逼近与容错性能

李小虎 1; 杜海峰 2; 张进华 1; 王孙安 1

您当前的位置：

首页 >

文章列表页 >

多层前向小世界神经网络的逼近与容错性能

更新时间：2025-07-09

- 多层前向小世界神经网络的逼近与容错性能
- Approximation and Fault-Tolerance Performances of Multilayer Feedforward Small World Neural Network
- 西安交通大学学报 2010年44卷第7期页码：59-63
- 作者机构：
  
  西安交通大学机械工程学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2009-11-23. 作者简介: 李小虎(1976-),男,博士生;杜海峰(联系人),男,教授,博士生导师. 基金项目: 国家自然科学基金资助项目(70671083,50505034)
- DOI：
  中图分类号： TP183
- 网络首发：2010-07-10，
  
  纸质出版：2010
- 稿件说明：
移动端阅览
李小虎 1, 杜海峰 2, 张进华 1, 等. 多层前向小世界神经网络的逼近与容错性能[J]. 西安交通大学学报, 2010,44(7):59-63.

Approximation and Fault-Tolerance Performances of Multilayer Feedforward Small World Neural Network[J]. 2010, 44(7): 59-63.
李小虎 1, 杜海峰 2, 张进华 1, 等. 多层前向小世界神经网络的逼近与容错性能[J]. 西安交通大学学报, 2010,44(7):59-63. DOI：

Approximation and Fault-Tolerance Performances of Multilayer Feedforward Small World Neural Network[J]. 2010, 44(7): 59-63. DOI：

摘要

基于Watts-Strogatz网络模型的构造思想

对多层前向神经网络中的规则连接依重连概率进行重连

构建了一种多层前向小世界神经网络模型.对该网络模型进行简要的数学描述

并以函数逼近和网络容错仿真考察了构建的小世界神经网络的性能.结果表明

与规则或随机连接的网络相比

当重连概率处于0.1～0.2时

小世界神经网络具有更优的逼近性能

且当网络学习速率参数在0.1～0.3之间时

对小世界神经网络的逼近性能影响较小.此外

当网络权值故障率小于30%时

重连概率不大于0.8的小世界神经网络与规则网络同样具有较优的容错性能

而当故障率大于40%时

重连概率较大的小世界神经网络和随机连接的神经网络的容错性能要明显优于规则网络.

Abstract

Based on the idea of Watts-Strogatz(W-S)network model

a multilayer feedforward small world neural network

which relies heavily on the rewiring probability

is constructed by reconnecting the regular links. The network model is described mathematically

and the function approximation and network fault-tolerance simulations are employed to investigate the performances of small world neural network. The results show that the small world neural network is endowed with better approximation performance compared with the regular or random connecting network; and when the learning rate parameter of network gets between 0.1 to 0.3

it exerts less impact on the approximation performance of small world neural network. In addition

when the network weight failure is less than 30%

small world neural network with rewiring probability less than 0.8 and regular network have the same better performance on fault-tolerance. When failure rate is more than 40%

small world neural network with high rewiring probability and random connecting neural network have distinctly better performance on fault-tolerance than the regular neural network has.

关键词

Keywords

references

李翔. 从复杂到有序:神经网络智能控制理论新进展[M]. 上海:上海交通大学出版社, 2006.

SPORNS O, HONEY C J. Small worlds inside big brains [J]. Proceedings of the National Academy of Sciences of the United States of America, 2006,103(51):19219-19220.

袁著祉, 陈增强,李翔. 联接主义智能控制综述 [J]. 自动化学报, 2002,28(1):38-59.

YUAN Zhuzhi, CHEN Zengqiang, LI Xiang. Connectionism intelligent control: a survey [J]. Acta Automatica Sinica, 2002,28(1):38-59.

XIA Y S, WANG J. A general methodology for designing globally convergent optimization neural networks [J]. IEEE Trans on Neural Networks, 1998,9(6):1331-1343.

MURATA N. Network information criterion determining the number of hidden units for an artificial neural network model [J]. IEEE Trans on Neural Networks, 1994, 5(6):865-872.

HORNIK K, STINCHCOMBE M, WHITE H. Multilayer feedforward networks are universal approximators [J]. Neural Networks, 1989,2(2):359-366.

WATTS D J, STROGATZ S H. Collective dynamics of small-world networks [J]. Nature, 1998,393(4): 440-442.

ALBERT R, BARABA'SI A L. Statistical mechanics of complex networks [J]. Reviews Modern Physics, 2002,74(1):47-97.

杜海峰, 庄健, 张进华,等. 用于函数优化的小世界优化算法 [J]. 西安交通大学学报, 2005,39(9):1011-1015.

DU Haifeng, ZHUANG Jian, ZHANG Jinhua, et al. Small-world phenomenon for function optimization [J]. Journal of Xi'an Jiaotong University, 2005,39(9):1011-1015.

SPORNS O, TONONI G, EDELMAN G M. Connectivity and complexity: the relationship between neuroanatomy and brain dynamics [J]. Neural Networks, 2000,13(8):909-922.

BASSETT D S, BULLMORE E. Small-world brain networks [J]. Neuroscientist, 2006,12(6):512-523.

SIMARD D, NADEAU L, KROGER H. Faster learning in small-world neural networks [J]. Physics Letters: A, 2005,336(1):8-15.

杜海峰, 李树茁, MARCUS W F, 等. 基于先验知识和模块性的网络社区结构探测算法 [J]. 西安交通大学学报, 2007,41(6):750-754.

DU Haifeng, LI Shuzhuo, MARCUS W F, et al. Detecting algorithm based on prior knowledge and modularity for networked community structure [J]. Journal of Xi'an Jiaotong University, 2007,41(6): 750-754.

KLEINBERG J. The small-world phenomenon and decentralized search [J]. SIAM News, 2004,37(3):1-3.

阮晓钢. 人工智能之路神经计算科学 [M]. 北京: 国防工业出版社, 2006.

阎平凡. 对多层前向神经网络研究的几点看法 [J]. 自动化学报, 1997,23(1):129-135.

YAN Pingfan. Some views on the research of multilayer feed forward neural networks [J]. Acta Automatica Sinica, 1997,23(1):129-135.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

动态短信通信复杂网络演化模型研究

扰动环境下作业车间网络多瓶颈识别方法研究

多源多工序加工过程波动扩散网络建模与波动源辨识

面向误差传递的复杂机械产品装配单元划分方法

面向高海拔空冷发电机防晕性能演化规律的电热耦合研究