A Niche Hierarchy Genetic Algorithms for Learning Wavelet Neural Networks

Based on the wavelet neural network (WNN) training algorithms and geometrical structure, a niche hierarchy genetic algorithm was proposed to improve the performance of wavelet networks. This evolutionary algorithm utilizes niche technology and the hierarchical chromosome to encode the structure and parameters of the wavelet network, and combines a genetic algorithm and evolutionary programming to construct and train the network simultaneously through evolution. By the evolutionary algorithm, the structure of WNN can be more reasonable, and the local minimum problem in the training process will be overcome efficiently. The experimental results show that the proposed method for the construction and training of the wavelet network is feasible and effective.