Evolving the Structure of the Particle Swarm Optimization Algorithms

A new model for evolving the structure of a Particle Swarm Optimization (PSO) algorithm is proposed in this paper. The model is a hybrid technique that combines a Genetic Algorithm (GA) and a PSO algorithm. Each GA chromosome is an array encoding a meaning for updating the particles of the PSO algorithm. The evolved PSO algorithm is compared to a human-designed PSO algorithm by using ten artificially constructed functions and one real-world problem. Numerical experiments show that the evolved PSO algorithm performs similarly and sometimes even better than standard approaches for the considered problems.

• Type: Conference
• Year: 2006
• Where: EVOW
• Authors: Laura Diosan, Mihai Oltean
• Keywords: Artificial Intelligence, Evolved Pso Algorithm, EVOW 2006, Human-designed Pso Algorithm, PSO Algorithm