摘要
文章提出了一种自动“删减”隐层神经元的ＲＢＦ神经网络学习算法。模拟结果表明，该算法训练的ＲＢＦ网络不仅结构得以优化，同时性能良好，可以成功地应用于模式分类和时间序列预测问题中。
In this paper, an adaptive scheme for pruning the hidden neurons of a radial basis function neural network is presented. Trained by the proposed learning algorithms, RBF network with optimized structure and better performance can be obtained, which has been confirmed on data classification problem and time series prediction.
出处
《微电子学与计算机》
CSCD
北大核心
2000年第4期14-18,共5页
Microelectronics & Computer
基金
安徽省教委自然科学基金
