CONVENTIONAL MODELING OF THE MULTILAYER PERCEPTRON USING POLYNOMIAL BASIS FUNCTIONS

In this letter, we present a new technique for modeling the multilayer perceptron (MLP) neural network, in which input and hidden units are represented by polynomial basis functions (PBF's). The MLP output is expressed as a linear combination of the PBF's and can therefore be expressed as a polynomial function of its inputs. Thus the MLP is isomorphic to conventional polynomial discriminant classifiers or Volterra filters. The modeling technique is successfully applied to several trained MLP networks.