Analysis of objectives relationships in multiobjective problems using trade-off region maps

Pinheiro, Rodrigo L., Landa-Silva, Dario and Atkin, Jason (2015) Analysis of objectives relationships in multiobjective problems using trade-off region maps. In: Proceedings of the 2015 Genetic and Evolutionary Computation Conference (GECCO 2015), 11-15 July 2015, Madrid, Spain.

Full text not available from this repository.

Abstract

Understanding the relationships between objectives in many-objective optimisation problems is desirable in order to develop more effective algorithms. We propose a techniquefor the analysis and visualisation of complex relationships between many (three or more) objectives. This technique looks at conflicting, harmonious and independent objectives relationships from different perspectives. To do that, it uses correlation, trade-off regions maps and scatter-plots in a four step approach. We apply the proposed technique to a set of instances of the well-known multiobjective multidimensional knapsack problem. The experimental results show that with the proposed technique we can identify local and complex relationships between objectives, trade-offs not derived from pairwise relationships, gaps in the fitness landscape, and regions of interest. Such information can be used to tailor the development of algorithms.

Item Type: Conference or Workshop Item (Paper)
RIS ID: https://nottingham-repository.worktribe.com/output/756950
Additional Information: Published in: GECCO '15 : Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation. New York : ACM, 2015. ISBN 9781450334723, pp. 735-742. doi: 10.1145/2739480.2754721
Keywords: multiobjective optimization, search landscape analysis, knapsack problem, personnel scheduling
Schools/Departments: University of Nottingham, UK > Faculty of Science > School of Computer Science
Depositing User: Landa-Silva, Dario
Date Deposited: 21 Jan 2016 08:47
Last Modified: 04 May 2020 17:13
URI: https://eprints.nottingham.ac.uk/id/eprint/31299

Actions (Archive Staff Only)

Edit View Edit View