RBF网学习的进化优选算法

被引:20
作者
魏海坤
徐嗣鑫
宋文忠
机构
[1] 东南大学自动化研究所!南京,东南大学自动化研究所!南京,东南大学自动化研究所!南京
关键词
RBF神经网络; 正交最小二乘算法; 进化算法; 选择路径;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
讨论了用正交最小二乘算法训练RBF网的不足之处 ,然后引入了选择路径的概念 ,在此基础上 ,提出了RBF网隐层节点选取的进化优选算法 .仿真结果表明 ,在不同的精度要求下 ,用进化优选算法均能设计出比正交最小二乘算法更小的RBF网 .
引用
收藏
页码:604 / 608
页数:5
相关论文
共 5 条
[1]  
Orthogonal least squares learning algorithms for radial basis function networks. Chen S,Cowan C F N and Grant P M. IEEE Transactions on Neural Networks . 1991
[2]  
A survey of evolution strategies. Back T,Hoffmeister F and Schwefel H P. Proc. of Fourth Int. Conf. on Genetic Algorithms[C], San Diego, CA . 1991
[3]  
On the efficiency of the orthogonal least squares training method for radial basis function networks. Sherstinsky A and Picard R W. IEEE Transactions on Neural Networks . 1996
[4]  
Networks for approximation and learning. Piggio T and Gorosi F. Proceedings of the IEEE . 1990
[5]  
The effective number of parameters: an analysis of generalization and regularization in nonlinear learning systems. Moody J E. In Advances in Neural Information Processing Systems 4[C], San Mateo, CA . 1992