The Model of NW Multilayer Feedforward Small-World Artificial Neural Networks and It’s Applied

Abstract

AbstractTo find the optimal neural network structure , the structure of multi-layer forward neural networks model is studied based on the research methods from the complex network, a new neural networks model, NW multilayer forward small world artificial neural networks can be proposed, whose structure of layer is between the regular model and the stochastic model. At first, regular of the multilayer feed-forward neural network neurons randomized cross-layer link back layer with a probability p to construct the new neural networks model. Secondly, the cross-layer small world artificial neural networks are used for function approximation under different re-wiring probability. The counts of convergence under different probability are compared by setting a same precision. Simulation shows that the small-world neural network has a better convergence speed than regular network and random network nearly p=0.08, the optimum performance of the NW multilayer forward small world artificial neural networks is proved in the right side of probability increases.KeywordsNW small-world networksmultilayer forward neural networksfunction approximation