Title :
Evolving wavelet neural networks
Author :
Yao, Susu ; Wei, Chengjian ; He, Zhenya
Author_Institution :
Dept. of Radio Eng., Southeast Univ., Nanjing, China
Abstract :
A wavelet neural network with evolutionary programming is proposed in this paper. Unlike the conventional backpropagation training algorithm, the evolutionary programming does not require gradient information and can provide a stochastic optimal search. The proposed method is used to approximate nonlinear functions and solve classification problems. Some experimental results are proposed to show the potential of the evolving wavelet neural networks
Keywords :
feedforward neural nets; function approximation; genetic algorithms; learning (artificial intelligence); pattern classification; search problems; wavelet transforms; classification problems; evolutionary programming; evolving wavelet neural networks; nonlinear functions; stochastic optimal search; Continuous wavelet transforms; Feedforward neural networks; Feeds; Function approximation; Genetic programming; Neural networks; Neurons; Signal analysis; Wavelet analysis; Wavelet transforms;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488903
