Radial Basis Function Neural Network

The Inverse Network

A radial basis function neural network is used as an inverse network for characterizing the defect profiles. The radial basis function is a three-layer function approximation network, as shown at right. The structure of the network is similar to that of a wavelet basis function neural network. The difference lies in the fact that the radial basis function uses a single set of basis functions (the scaling functions in the wavelet basis function neural network). The training algorithm for the radial basis function is similar to that of the wavelet basis function neural network with the centers of the basis functions determined by using a clustering algorithm. The spread of each basis function is proportional to the cluster size. Alternatively, it may be set to some common constant value for all bases. The output interconnection weights are then determined by a matrix inversion step.

On to information on optimization of the networks.