摘要
径向基函数网络与BP网络在网络结构上都属于前向网络,但它们对网络权值训练所采用的算法是完全不同的。另外,径向基函数网络的网络结构与模糊系统有很紧密的关联。该文从径向基函数网络的结构入手,分别对其所具有的特点、权值训练、网络设计方法及其应用等方面,通过分析与实例,采用对比的方式,给予实验的验证。
Radial basis function networks(RBFN for short)and BP networks both belong to the feed-forward networks in the structure,but they have quite different way to train the weights of their networks.On the other hand,the structure of the RBFN has the relationship with the fuzzy logic system.The paper analyses the characteristics,the ways of weight train and the design of networks through the network construction.The numerical examples are tested by the comparative way.
出处
《计算机工程与应用》
CSCD
北大核心
2002年第3期85-87,200,共4页
Computer Engineering and Applications
基金
中国科学院重点项目资助
