Frankenstein's PSO: A Composite Particle Swarm Optimization Algorithm

• Published in: IEEE transactions on evolutionary computation Vol. 13; no. 5; pp. 1120 - 1132
• Main Authors: Montes de Oca, M.A., Stutzle, T., Birattari, M., Dorigo, M.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.10.2009; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithm design and analysis; Algorithms; Analogies; Applied sciences; Artificial intelligence; Computer science; control theory; systems; Continuous optimization; Convergence; Design optimization; Empirical analysis; Evolutionary algorithms; Exact sciences and technology; experimental analysis; integration of algorithmic components; Learning and adaptive systems; Minimization methods; Optimization; Particle swarm optimization; particle swarm optimization (PSO); Particulate composites; Performance evaluation; run-time distributions; Runtime; Scholarships; swarm intelligence; Topology; Performance evaluation; Continuous optimization; Evolutionary algorithm; Empirical method; Algorithmics; Neural network; Particle swarm optimization; Time distribution; experimental analysis; integration of algorithmic components; Swarm intelligence; Reliability; particle swarm optimization (PSO); Artificial intelligence; run-time distributions
• ISSN: 1089-778X; 1941-0026
• DOI: 10.1109/TEVC.2009.2021465

During the last decade, many variants of the original particle swarm optimization (PSO) algorithm have been proposed. In many cases, the difference between two variants can be seen as an algorithmic component being present in one variant but not in the other. In the first part of the paper, we present the results and insights obtained from a detailed empirical study of several PSO variants from a component difference point of view. In the second part of the paper, we propose a new PSO algorithm that combines a number of algorithmic components that showed distinct advantages in the experimental study concerning optimization speed and reliability. We call this composite algorithm Frankenstein's PSO in an analogy to the popular character of Mary Shelley's novel. Frankenstein's PSO performance evaluation shows that by integrating components in novel ways effective optimizers can be designed.