A New Method of Multilayer Perceptron Encoding

One of the central issues in neural network research is how to find an optimal MultiLayer Perceptron architecture. The number of neurons, their organization in layers, as well as their connection scheme have a considerable influence on network learning, and on the capacity for generalization [7]. A solution to find out these parameters is needed: The neuro-evolution ([1,2,4,5]). The novelty is to emphasize the network performance aspects, and the network simplification achieved by reducing the network topology. All these genetic manipulations on the network architecture should not decrease the neural network performance.