A global learning algorithm for a RBF network

Citation data:

Neural Networks, ISSN: 0893-6080, Vol: 12, Issue: 3, Page: 527-540

Publication Year:
Usage 15
Abstract Views 15
Captures 9
Readers 7
Exports-Saves 2
Citations 58
Citation Indexes 58
Repository URL:
https://works.bepress.com/qiuming-zhu/22; https://digitalcommons.unomaha.edu/compscifacpub/49
Zhu, Qiuming; Cai, Yao; Liu, Luzheng
Elsevier BV
Computer Science; Neuroscience; RBF neural networks; Competitive neuron layer; Maximum likelihood classification; Hyper-ellipsoidal subspace; Subclass clustering; Computer Sciences
article description
This article presents a new learning algorithm for the construction and training of a RBF neural network. The algorithm is based on a global mechanism of parameter learning using a maximum likelihood classification approach. The resulting neurons in the RBF network partitions a multidimensional pattern space into a set of maximum-size hyper-ellipsoid subspaces in terms of the statistical distributions of the training samples. An important feature of the algorithm is that the learning process includes both the tasks of discovering a suitable network structure and of determining the connection weights. The entire network and its parameters are thought of evolved gradually in the learning process.