A novel multiplicative neural network architecture motivated by spiking neuron model

Abstract

In this paper, learning algorithm for a multiplicative neural network motivated by spiking neuron model (MSN) is proposed and tested for various applications where a multilayer perceptron (MLP) neural network is conventionally used. It is observed that the inclusion of a few more biological phenomena in the formulation of artificial neural network models make them more prevailing. Several benchmark and real-life problems of classification and function-approximation are illustrated.