| 一种改进的RBF神经网络学习算法 |
| |
| 引用本文： | 王洪斌,杨香兰,王洪瑞. 一种改进的RBF神经网络学习算法[J]. 系统工程与电子技术, 2002, 24(6): 103-105 |
| |
| 作者姓名： | 王洪斌 杨香兰 王洪瑞 |
| |
| 作者单位： | 燕山大学电气工程学院,河北,秦皇岛,066004 |
| |
| 摘 要： | 提出了一种改进的RBF神经网络学习算法 ,分别通过减聚类和监督学习算法对网络参数和权值进行训练 ,既可以根据样本合理地聚类、确定RBF径向基函数的个数和相应参数 ,又具有较强的网络映射能力 ,从而不仅使RBF神经网络结构得以优化 ,性能也得到了提高。仿真结果表明了该学习算法的实用性和有效性
|
| 关 键 词： | RBF神经网络 减聚类算法 监督学习算法 |
| 文章编号： | 1001-506X(2002)06-0103-03 |
| 修稿时间： | 2001-04-18 |
An Improved Learning Algorithm for RBF Neural Networks |
| |
| WANG Hong bin,YANG Xiang lan,WANG Hong rui. An Improved Learning Algorithm for RBF Neural Networks[J]. System Engineering and Electronics, 2002, 24(6): 103-105 |
| |
| Authors: | WANG Hong bin YANG Xiang lan WANG Hong rui |
| |
| Abstract: | An improved learning algorithm for RBF neural networks is presented. By using subtractive clustering method and supervised learning algorithm to train the networks, it can cluster properly based on the set of data, thus obtaining both the parameters of the neurons and the number of the hidden neurons automatically. The RBF has an optimized structure and a better performance. The simulation indicates the practicability and efficiency of the new algorithm. |
| |
| Keywords: | RBF neural networks Subtractive clustering method Supervised learning algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
