A Recursive Orthogonal Least Squares Algorithm for Training RBF Networks
Neural Processing Letters
A classification technique based on radial basis function neural networks
Advances in Engineering Software
Hi-index | 0.00 |
This paper proposed an improved APC-III algorithm to determine RBFNN's hidden-layer structure. The improved algorithm is used to obtain the different radial basis width and less hidden-layer node number. The RBFNN was constructed with the hidden-layer structure that got by the improved algorithm. Two nonlinear system's model simulations were carried on with the RBFNN. The simulation results show that the structure produced RBFNN by improved algorithm is simple and the generalization performance is better than the original algorithm.