Understanding particle swarms through simplification: a study of recombinant PSO

Simplified forms of the particle swarm algorithm are very beneficial in contributing to understanding of what makes a PSO swarm function in the way that it does. One of these forms, PSO with discrete recombination, is analyzed in depth, demonstrating not just improvements in performance to a standard PSO algorithm, but also significantly different behavior with a reduction in bursting patterns due to the removal of stochastic components from the update equations. This altered behavior accompanied by equal and improved performance leads to conjectures that bursts are not generally efficacious in the optimization process.

[1]  Tim M. Blackwell,et al.  The Lévy Particle Swarm , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[2]  Kevin D. Seppi,et al.  Exposing origin-seeking bias in PSO , 2005, GECCO '05.

[3]  J. Jaccard,et al.  LISREL Approaches to Interaction Effects in Multiple Regression , 1998 .

[4]  Andres Upegui,et al.  Particle Swarm Optimization with Discrete Recombination: An Online Optimizer for Evolvable Hardware , 2006, First NASA/ESA Conference on Adaptive Hardware and Systems (AHS'06).

[5]  James Kennedy,et al.  Defining a Standard for Particle Swarm Optimization , 2007, 2007 IEEE Swarm Intelligence Symposium.

[6]  Tim Blackwell,et al.  Origin of bursts , 2007, GECCO '07.

[7]  James Kennedy,et al.  Probability and dynamics in the particle swarm , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).