A novel similarity-based mutant vector generation strategy for differential evolution

Eduardo Segredo, Eduardo Lalla-Ruiz, Emma Hart

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

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

The mutant vector generation strategy is an essential component of Differential Evolution (de), introduced to promote diversity, resulting in exploration of novel areas of the search space. However, it is also responsible for promoting intensification, to improve those solutions located in promising regions. In this paper we introduce a novel similarity-based mutant vector generation strategy for de, with the goal of inducing a suitable balance between exploration and exploitation, adapting its behaviour depending on the current state of the search. In order to achieve this balance, the strategy considers similarities among individuals in terms of their Euclidean distance in the decision space. A variant of de incorporating the novel mutant vector generation strategy is compared to well-known explorative and exploitative adaptive de variants. An experimental evaluation performed on a well-known suite of large-scale continuous problems shows that the new de algorithm that makes use of the similarity-based approach provides better performance in comparison to the explorative and exploitative de variants for a wide range of the problems tested, demonstrating the ability of the new component to properly balance exploration and exploitation.

Original languageEnglish
Title of host publicationProceedings of GECCO '18
Subtitle of host publicationGenetic and Evolutionary Computation Conference, 15-19 July 2018, Kyoto, Japan
Place of PublicationNew York
PublisherAssociation for Computing Machinery (ACM)
Pages881-888
Number of pages8
ISBN (Electronic)978-1-4503-5618-3
DOIs
Publication statusPublished - 2 Jul 2018
EventGenetic and Evolutionary Computation Conference, GECCO 2018 - Kyoto Terrsa, Kyoto, Japan
Duration: 15 Jul 201819 Jul 2018
http://gecco-2018.sigevo.org/index.html/tiki-index.php

Conference

ConferenceGenetic and Evolutionary Computation Conference, GECCO 2018
Abbreviated titleGECCO 2018
CountryJapan
CityKyoto
Period15/07/1819/07/18
OtherA recombination of the 27th International Conference on Genetic Algorithms (ICGA) and the 23rd Annual Genetic Programming Conference (GP)
Internet address

Keywords

  • Differential evolution
  • Diversity
  • Global optimization
  • Large-scale optimization
  • Similarity
  • Heuristis
  • Metaheuristics
  • Artificial Intelligence
  • Continuous Optimization

Cite this