Particle Swarm Guided Evolution Strategy for Real-Parameter Optimization

Evolution strategy (ES) and particle swarm optimization (PSO) are two of the most popular research topics for tackling real-parameter optimization problems in evolutionary computation. Both of them have strengths and weaknesses for their different search behaviors and methodologies. In ES, mutation, as the main operator, tries to find good solutions around each individual. While in PSO, particles are moving toward directions determined by certain global information, such as the global best particle. In order to leverage the specialties offered by both sides to our advantage, this paper combines the essential mechanism of ES and the key concept of PSO to develop a new hybrid optimization methodology, called particle swarm guided evolution strategy. We introduce swarm intelligence to the ES mutation framework to create a new mutation operator, called guided mutation, and integrate the guided mutation operator into ES. Numerical experiments are conducted on a set of benchmark functions, and the experimental results indicate that PSGES is a promising optimization methodology as well as an interesting research direction.