Skip to main content
Presentation
Intensification strategies for extremal optimisation
8th International conference on simulated evolution and learning (SEAL-2010)
  • Marcus Randall, Bond University
  • Andrew Lewis, Griffith University
Date of this Version
12-1-2010
Document Type
Conference Paper
Publication Details

Accepted Version.

Randall, M. & Lewis, A. (2010). Intensification strategies for extremal optimisation. Paper presented at the 8th International conference on simulated evolution and learning (SEAL-2010), Kanpur, India.

Access the conference website.

2010 HERDC submission. FoR Code: 010300

© Copyright Springer-Verlag Berlin Heidelberg, 2010

Disciplines
Abstract

It is only relatively recently that extremal optimisation (EO) has been applied to combinatorial optimisation problems. As such, there have been only a few attempts to extend the paradigm to include standard search mechanisms that are routinely used by other techniques such as genetic algorithms, tabu search and ant colony optimisation. The key way to begin this process is to augment EO with attributes that it naturally lacks. While EO does not get confounded by local optima and is able to move through search space unencumbered, one of the major issues is to provide it with better search intensification strategies. In this paper, two strategies that compliment EO's mechanics are introduced and are used to augment an existing solver framework. Results, for single and population versions of the algorithm, demonstrate that intensification aids the performance of EO.

Citation Information
Marcus Randall and Andrew Lewis. "Intensification strategies for extremal optimisation" 8th International conference on simulated evolution and learning (SEAL-2010) (2010) p. 115 - 124
Available at: http://works.bepress.com/marcus_randall/32/