Comparison of two interval models for interval-valued particle swarm optimization
Hidehiko Okada · 2013
The author previously proposed an extension of particle swarm optimization (PSO). The proposed method extends the processes of PSO to handle interval numbers as genotype values so that PSO can be applied directly to interval-valued optimization problems. The interval PSO (IPSO) can employ either of two interval models, the lower and upper model or the center and width model, for specifying genotype values. Ability of the IPSO in searching for solutions may depend on the model. In this paper, the author compares the two models to investigate which model contributes better for the IPSO to find better solutions. IPSO is applied to evolutionary training of interval-valued neural networks. A result of preliminary study indicates that the CW model is slightly better than the LU model and both models contribute well: the IPSO with the LU/CW model could evolve neural networks with a small error.