局部线性模型在小波神经网络中的应用(英文)  

The Application of Local Linear Model in Wavelet Neural Network

作　　者：金丽婷[1] 黄敏[1] 崔宝同[1] 

机构地区：[1]江南大学通信与控制工程学院,江苏无锡214122

出　　处：《内蒙古师范大学学报（自然科学汉文版）》2008年第1期32-36,共5页Journal of Inner Mongolia Normal University(Natural Science Edition)

基　　金：Project Supported by the National Natural Science Foundation of China(60674026);the Jiangsu Provincial Natural Science Foundation of China(BK2007016)

摘　　要：把一种局部线性模型应用于小波神经网络中,把网络隐层与输出层之间的直接连接权值用一种线性模型来代替以减少隐层节点数目.对带局部线性模型的小波神经网络与一般小波神经网络的性能进行比较,试验结果证明了带局部线性模型的小波神经网络的优越性.The application of local linear model in wavelet neural network （WNN） is proposed. A linear model is adopted instead of the straightforward connection weights between the hidden layer and output layer for more parsimonious hidden units. The performance is compared with the WNN without the local linear model in the appli- cation of a two variable function approximation. The experimental results demonstrate the superiority of the local linear model in WNN.

关 键 词：小波神经网络 局部线性模型 函数逼近 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]