| 一种用于径向基函数（RBF）神经网络训练的有效方法 |
| |
| 引用本文： | 孙毅刚 战强. 一种用于径向基函数（RBF）神经网络训练的有效方法[J]. 哈尔滨工业大学学报, 1997, 29(4): 103-106 |
| |
| 作者姓名： | 孙毅刚 战强 |
| |
| 作者单位： | 哈尔滨工业大学机器人研究所 |
| |
| 摘 要： | 提出了一种用于径向基函数神经网络训练的新方法，即Ｇａｕｓｓ－Ｊｏｒｄａｎ与广义逆的复合法。仿真结果表明，此方法训练速度快，实时性强，其收敛性和收敛严谨的比正交最小二乘算法效果好。
|
| 关 键 词： | 径向基函数 神经网络 仿真 复合法 |
A Useful Method for the Training of Radial Basis Function Neural Networks |
| |
| Sun Yigang Zhan Qiang. A Useful Method for the Training of Radial Basis Function Neural Networks[J]. Journal of Harbin Institute of Technology, 1997, 29(4): 103-106 |
| |
| Authors: | Sun Yigang Zhan Qiang |
| |
| Affiliation: | Robot Research Institute |
| |
| Abstract: | This paper proposes a new method for the training of Radial Basis Function (RBF) neural networks, which is a compound algorithm of Gauss-Jordan algorithm and General Inverse algorithm. The simulating results show that the training of the method is very quickly convergent, a good real-time way and has better quality than Orthogonal Least Squares Algothrism in Least Mean Square error and convergence. |
| |
| Keywords: | Radial Basis Function (RBF) neural network simulation orthogonal least square algorithm blending algorithm |
| 本文献已被 CNKI 维普 等数据库收录！ |
