Adaptively choosing niching parameters in a PSO

Niching techniques play an important role in evolutionary algorithms. Existing niching methods often require userspecified parameters, limiting their usefulness. This paper proposes a niching method for Particle Swarm Optimisation (PSO) where population statistics are used to adaptively determine the niching parameters during a run. The proposed niching method is compared to another niching based PSO, SPSO. Our results show that the proposed adaptive niching PSO can solve difficult multimodal functions more reliably and with fewer evaluations. Categories and Subject Descriptors G.1 [Numerical Analysis]: Optimisation; F.2.1 [Analysis of Algorithms and Problem Complexity]: Numerical Algorithms and Problems General Terms Algorithms Keywords Evolutionary Computation, Particle Swarms, Multimodal Function Optimisation

• Type: Conference
• Year: 2006
• Where: GECCO
• Authors: Stefan Bird, Xiaodong Li
• Keywords: GECCO 2006, Niching Method, Optimization, Particle Swarm Optimisation, Particle Swarms