PSO Facing Non-Separable and Ill-Conditioned Problems - Archive ouverte HAL Access content directly
Reports (Research Report) Year : 2008

PSO Facing Non-Separable and Ill-Conditioned Problems

Abstract

This report investigates the behavior of particle swarm optimization (PSO) on ill-conditioned functions. We find that PSO performs very well on separable, ill-conditioned functions. If the function is rotated such that it becomes non-separable, the performance declines dramatically. On non-separable, ill-conditioned functions we find the search costs (number of function evaluations) of PSO increasing roughly proportional with the condition number. We never observe premature convergence, but on non-separable, ill-conditioned problems PSO is outperformed by a contemporary evolution strategy by orders of magnitude. The strong dependency of PSO on rotations originates from random events that are only independent within the given coordinate system. We argue that invariance properties, like rotational invariance, are desirable, because they increase the predictive power of performance results.
Fichier principal
Vignette du fichier
RR-6447.pdf (748.2 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

inria-00250078 , version 1 (09-02-2008)
inria-00250078 , version 2 (11-02-2008)

Identifiers

  • HAL Id : inria-00250078 , version 2

Cite

Nikolaus Hansen, Raymond Ros, Nikolas Mauny, Marc Schoenauer, Anne Auger. PSO Facing Non-Separable and Ill-Conditioned Problems. [Research Report] RR-6447, INRIA. 2008. ⟨inria-00250078v2⟩
415 View
566 Download

Share

Gmail Mastodon Facebook X LinkedIn More