CoDE Publications CoDE Publications
IRIDIA Publications IRIDIA Publications
SMG Publications
WIT Publications
WIT Publications
SMG Publications
Home People Research Activities Publications Teaching Resources
By Class By Topic By Year Technical Reports
By Class By Topic By Year Technical Reports
login
M. Montes de Oca, T. Stützle, M. Birattari, and M. Dorigo. Frankenstein's PSO: A Composite Particle Swarm Optimization Algorithm. Technical Report TR/IRIDIA/2007-006, IRIDIA, Université Libre de Bruxelles, Brussels, Belgium, March 2007.

Abstract

During the last decade, many modifications of the original particle swarm optimization (PSO) algorithm have been proposed. In many cases, it is claimed that the modified variant is superior to some reference variant in some way. The differences between two variants can often be seen as an algorithmic component being present in one variant but not in the other. From this perspective, the question arises as to whether it is possible to integrate different algorithmic components into a single PSO variant that performs better than the variants from which its components are taken. In this paper, we take this perspective to design a new PSO algorithm whose components were selected after a careful evaluation of their impact on optimization speed and reliability. We call this composite algorithm Frankenstein's PSO in an analogy to the popular character of Mary Shelley's novel. The evaluation of Frankenstein's PSO performance suggests that the answer to the driving question is positive. We present the process that guided us in selecting and adapting the algorithmic components included in Frankenstein's PSO. The performance of the composite algorithm is validated via a comparison with the variants from which the components were taken on a number of well-known benchmark problems.


Updated: 2017-03-27