A Comparative Study of Four Parallel and Distributed PSO Methods

We present four new parallel and distributed particle swarm optimization methods consisting in a genetic algorithm whose individuals are co-evolving swarms, an “island model”-based multi-swarm system, where swarms are independent and interact by means of particle migrations at regular time steps, an...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:New generation computing Jg. 29; H. 2; S. 129 - 161
Hauptverfasser: Vanneschi, Leonardo, Codecasa, Daniele, Mauri, Giancarlo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Heidelberg Verlag Omsha Tokio 01.04.2011
Schlagworte:
Artificial Intelligence
Computer Hardware
Computer Science
Computer Systems Organization and Communication Networks
Software Engineering/Programming and Operating Systems
Parallel and Distributed Algorithms
Optimization
Swarm Intelligence
ISSN:0288-3635, 1882-7055
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present four new parallel and distributed particle swarm optimization methods consisting in a genetic algorithm whose individuals are co-evolving swarms, an “island model”-based multi-swarm system, where swarms are independent and interact by means of particle migrations at regular time steps, and their respective variants enriched by adding a repulsive component to the particles. We study the proposed methods on a wide set of problems including theoretically hand-tailored benchmarks and complex real-life applications from the field of drug discovery, with a particular focus on the generalization ability of the obtained solutions. We show that the proposed repulsive multi-swarm system has a better optimization ability than all the other presented methods on all the studied problems. Interestingly, the proposed repulsive multi-swarm system is also the one that returns the most general solutions.
ISSN:0288-3635
1882-7055
DOI:10.1007/s00354-010-0102-z