Using landscape topology to compare continuous metaheuristics: a framework and case study on EDAs and ridge structure

Morgan, R. and Gallagher, M. (2012) Using landscape topology to compare continuous metaheuristics: a framework and case study on EDAs and ridge structure. Evolutionary Computation, 20 2: 277-299. doi:10.1162/EVCO_a_00070

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
UQ276871_OA.pdf Full text (open access) application/pdf 1.29MB 0

Author Morgan, R.
Gallagher, M.
Title Using landscape topology to compare continuous metaheuristics: a framework and case study on EDAs and ridge structure
Journal name Evolutionary Computation   Check publisher's open access policy
ISSN 1063-6560
1530-9304
Publication date 2012-07
Sub-type Article (original research)
DOI 10.1162/EVCO_a_00070
Open Access Status File (Publisher version)
Volume 20
Issue 2
Start page 277
End page 299
Total pages 23
Place of publication Cambridge, MA, United States
Publisher MIT Press
Collection year 2013
Language eng
Abstract In this paper we extend a previously proposed randomized landscape generator in combination with a comparative experimental methodology to study the behavior of continuous metaheuristic optimization algorithms. In particular, we generate twodimensional landscapes with parameterized, linear ridge structure, and perform pairwise comparisons of algorithms to gain insight into what kind of problems are easy and difficult for one algorithm instance relative to another.We apply thismethodology to investigate the specific issue of explicit dependency modeling in simple continuous estimation of distribution algorithms. Experimental results reveal specific examples of landscapes (with certain identifiable features) where dependency modeling is useful, harmful, or has little impact on mean algorithm performance. Heat maps are used to compare algorithm performance over a large number of landscape instances and algorithm trials. Finally, we perform ameta-search in the landscape parameter space to find landscapes which maximize the performance between algorithms. The results are related to some previous intuition about the behavior of these algorithms, but at the same time lead to new insights into the relationship between dependency modeling in EDAs and the structure of the problem landscape. The landscape generator and overall methodology are quite general and extendable and can be used to examine specific features of other algorithms.
Keyword Problem generator
Experimental analysis
Fitness landscape
Estimation of distribution algorithms
Q-Index Code C1
Q-Index Status Confirmed Code
Institutional Status UQ
Additional Notes Posted Online: 7 May 2012.

Document type: Journal Article
Sub-type: Article (original research)
Collections: Official 2013 Collection
School of Information Technology and Electrical Engineering Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 2 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 3 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Mon, 02 Jul 2012, 08:31:37 EST by System User on behalf of School of Information Technol and Elec Engineering