A fast on-line learning algorithm for multidimensional fuzzy neural networks
Mohamed S. Ibrahim · 2005
Abstract:-This paper presents a fast on-line algorithm for constructing multidimensional fuzzy neural networks based on the possibility theory (MFNN-P).The multidimensional fuzzy neurons have a center vector and radius vector and the dimensions of these vectors are the same as the dimensions of the input vectors. The possibility measure is incorporated at the neurons in the output layer. A new on-line algorithm is developed. Simulation results show that the proposed network can deal with nonlinearities and uncertainties of the system. Key –Words:- fuzzy, neural network, possibility theory, multidimensional neurons.