No Cover Image

Conference Paper/Proceeding/Abstract 410 views 35 downloads

Emergence of structural bias in differential evolution

Bas van Stein, Fabio Caraffini Orcid Logo, Anna V. Kononova

Proceedings of the Genetic and Evolutionary Computation Conference Companion

Swansea University Author: Fabio Caraffini Orcid Logo

  • 62443_VoR.pdf

    PDF | Version of Record

    © 2021 Copyright held by the owner/author(s). Released under the terms of a CC-BY license.

    Download (1.33MB)

DOI (Published version): 10.1145/3449726.3463223

Abstract

Heuristic optimisation algorithms are in high demand due to the overwhelming amount of complex optimisation problems that need to be solved. The complexity of these problems is well beyond the boundaries of applicability of exact optimisation algorithms and therefore require modern heuristics to fin...

Full description

Published in: Proceedings of the Genetic and Evolutionary Computation Conference Companion
ISBN: 978-1-4503-8351-6
Published: New York, NY, USA ACM 2021
URI: https://cronfa.swan.ac.uk/Record/cronfa62443
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: Heuristic optimisation algorithms are in high demand due to the overwhelming amount of complex optimisation problems that need to be solved. The complexity of these problems is well beyond the boundaries of applicability of exact optimisation algorithms and therefore require modern heuristics to find feasible solutions quickly. These heuristics and their effects are almost always evaluated and explained by particular problem instances. In previous works, it has been shown that many such algorithms show structural bias, by either being attracted to a certain region of the search space or by consistently avoiding regions of the search space, on a special test function designed to ensure uniform 'exploration' of the domain. In this paper, we analyse the emergence of such structural bias for Differential Evolution (DE) configurations and, specifically, the effect of different mutation, crossover and correction strategies. We also analyse the emergence of the structural bias during the run-time of each algorithm. We conclude with recommendations of which configurations should be avoided in order to run DE unbiased.
College: Faculty of Science and Engineering