On the selection of fitness landscape analysis metrics for continuous optimization problems

Yuan Sun, Saman K Halgamuge, Michael Kirley, Mario A Muñoz

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

11 Citations (Scopus)

Abstract

Selecting the best algorithm for a given optimization problem is non-trivial due to large number of existing algorithms and high complexity of problems. A possible way to tackle this challenge is to attempt to understand the problem complexity. Fitness Landscape Analysis (FLA) metrics are widely used techniques to extract characteristics from problems. Based on the extracted characteristics, machine learning methods are employed to select the optimal algorithm for a given problem. Therefore, the accuracy of the algorithm selection framework heavily relies on the choice of FLA metrics. Although researchers have paid great attention to designing FLA metrics to quantify the problem characteristics, there is still no agreement on which combination of FLA metrics should be employed. In this paper, we present some well-performed FLA metrics, discuss their contributions and limitations in detail, and map each FLA metric to the captured problem characteristics. Moreover, computational complexity of each FLA metric is carefully analysed. We propose two criteria to follow when selecting FLA metrics. We hope our work can help researchers identify the best combination of FLA metrics.

Original languageEnglish
Title of host publication2014 7th International Conference on Information and Automation for Sustainability
Subtitle of host publicationSharpening the Future with Sustainable Technology
EditorsAnjula De Silva
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)9781479945986
DOIs
Publication statusPublished - 26 Mar 2014
EventIEEE International Conference on Information and Automation for Sustainability 2014 - Galadari Hotel, Colombo, Sri Lanka
Duration: 22 Dec 201424 Dec 2014
Conference number: 7

Conference

ConferenceIEEE International Conference on Information and Automation for Sustainability 2014
Abbreviated titleICIAfS 2014
CountrySri Lanka
CityColombo
Period22/12/1424/12/14

Keywords

  • Continuous optimization problem
  • fitness landscape analysis
  • problem characteristics
  • problem difficulty

Cite this

Sun, Y., Halgamuge, S. K., Kirley, M., & Muñoz, M. A. (2014). On the selection of fitness landscape analysis metrics for continuous optimization problems. In A. De Silva (Ed.), 2014 7th International Conference on Information and Automation for Sustainability: Sharpening the Future with Sustainable Technology [7069635] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIAFS.2014.7069635