A critical assessment of some variants of particle swarm optimization

Among the variants of the basic Particle Swarm Optimization algorithm as first proposed in 1995 discussed in literature, an interesting one, which combines both approaches, is the one proposed by Miranda and Fonseca in 2002, which seems to produce significant improvements. We analyze the effects of two modifications introduced in that work (adaptive parameter setting and selection based on an evolution strategies-like approach) separately, reporting results obtained on a set of multimodal benchmark functions, which showi that they may have opposite and complementary effects. In particular, using only parameter adaptation when optimizing ’harder’ functions yields better results than when both modifications are applied. We also propose a justification for this, based on recent analyses in which particle swarm optimizers are studied as dynamical systems.