RBF neural networks for classification using new kernel functions

Authors:
Hamparsum Bozdogan;Zhenqiu Liu
Affiliations:
Department of Statistics, The University of Tennessee Knoxville, TN;Community Health Research Group, Suite 309 600 Henley Street, The University of Tennessee Knoxville, TN
Venue:
Second international workshop on Intelligent systems design and application
Year:
2002

Citing 2
Cited 0

Fat-shattering and the learnability of real-valued functions

Journal of Computer and System Sciences
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we introduce a new class of kernel functions1 and try to combine kmean, EM, and new kernels together. The resulting network is easy to use , and has favorable classification accuracy. Experiments show that our new model has better performance than traditional RBF neural networks and is super compared with MLP and other nonparametric classification tools.