关键词: BP算法, 权值平衡算法, 小波网络, 算法改进

Abstract: The characters of BP Algorithm is analyzed in this paper. On the basis of the Weight Balance Algorithm, BP Algorithm is modified by WNN. And the BP Weight Balance Algorithm based on WNN is proposed and the concrete steps of this Algorithm are presented. This Improved Weight Balance Algorithm adopts one kind of continuous Wavelet functions—Gauss Wavelet function to replace Sigmoid Function to be implicit layer challenging function, and also bring in form factor υk to balance the value of weighting regulation in BP. Theoretical analysis and Simulation result show it is an effective algorithm, which not only is simple and efficient like BP Algorithm but also can increase the learning speed and precision, avoid the problem that BP neural network is easy to be trapped in local minima and its convergence speed is slow.

Key words: BP Algorithm, Weight Balance Algorithm, Wavelet Neural Networks, Algorithm Improvement