A novel neural network training algorithm based on particle swarm optimizer and optimal foraging theory

A novel neural network training algorithm based on particle swarm optimization (PSO) and optimal foraging theory (OFT) is presented in this paper. The proposed algorithm, called PSOOFT, makes use of two mechanisms of OFT: a reproduction strategy to enhance the ability to converge rapidly to good solutions and a patch-choice based scheme to keep a right balance of exploration, and exploitation, The PSOOFT is applied for training a multi-layer feed-forward neural network for three benchmark classification problems. The performance of PSOOFT used for neural network training is compared to that of Back Propagation (BP), genetic algorithm (GA) and standard PSO (SPSO), demonstrating its effectiveness and efficiency