On the use of multiplicative neuron in feedforward neural networks

Neurons are functional units and can be considered as generators of function spaces. Neuron modelling concerns relating the function to the structure of the neuron on the basis of its operation. Most existing neuron models are based on the summing operation of the inputs. In this paper we present a new neuron model, multiplicative neuron, that performs multiplication operation instead of simple summation. The computational and learning capabilities of the model have been tested on some functional mapping and time series prediction problems. Simulation results show that the proposed neuron model, when used in a feedforward neural network, performs better than existing multilayer networks (MLN).