STRUCTURE OPTIMIZATION STRATEGY OF NORMALIZED RBF NETWORKS

摘要

Aimed at studying normalized radial basis function network (NRBFN), this paper introduces the subtractive clustering based on a mountain function to construct the initial structure of NRBFN, adopts singular value decomposition (SVD) to analyze the relationship between neural nodes of the hidden layer and singular values, cumulative contribution ratio, index vector, and optimizes the structure of NRBFN. Finally, simulation and performance comparison show that the algorithm is feasible and effective.