A New Type of BP Wavelet Neural Network Based on Fast Training Algorithm

摘要：

A new type of BP wavelet neural network with local linear character is proposed in this paper. The activation function of the network in hidden layer neurons is substituted for wavelet functions. The connection weights between the hidden layer and output layer of the above WNN are replaced by a local linear model. The network is trained using fast training algorithm (adaptive adjustment of learning rate). The performance of this network is simulated and the results are reported.

作者: Liting Jin Baotong Cui

作者单位: Institute of Automation Southern Yangtze University Wuxi, 214122, P. R. China

会议类型: 国际会议

会议名称: Fourth International Conference on Impulsive and Hybrid Dynamical Systems(ICIHDS 2007)(第四届国际脉冲和混合动力系统学术会议)

会议地点: 南宁

会议语种:英文

在线出版日期: 2007-07-20（万方平台首次上网日期，不代表论文的发表时间）