一种新的径向基概率神经网络模型(Ⅰ):基本理论

A NEW MODEL ON RADIAL BASIS PROBABILISTIC NEURAL NETWORKS （Ⅰ）:BASIC THEORY

摘要: 文中在径向基函数网络（ＲＢＦＮ）和概率神经网络（ＰＮＮ）的基础上，提出了一种径向基概率神经网络（ＲＢＰＮＮ）模型，这种网络保留了前两种网络模型的优点，既可以减少网络连接权值的训练时间，又能减少网络隐单元的数目，同时，网络用于测试的时间也较ＲＢＦＮ明显地下降．

Abstract: Based on radial basis function network （RBFN） and probabilistic neural network （PNN）, a new radial basis probabilistic neural network （RBPNN） model is proposed. This new network inherits the advantages of RBFNs and PNNs, and it can reduce not only the computation complexity for the network but also the number of the hidden nodes of the network. Specifically, the testing time for this new model is much shorter than that for the RBFN.