Better fixed-arity unbiased black-box algorithms

Research output: Chapter in Book/Report/Conference proceedingConference Proceeding (Non-Journal item)

Abstract

In their GECCO'12 paper, Doerr and Doerr proved that the k-ary unbiased black-box complexity of OneMax on n bits is O(n/k) for 2 ≤ k ≤ log2 n. We propose an alternative strategy for achieving this unbiased black-box complexity when 3 ≤ k ≤ log2 n. While it is based on the same idea of block-wise optimization, it uses k-ary unbiased operators in a different way. For each block of size 2k −1 − 1 we set up, in O(k) queries, a virtual coordinate system, which enables us to use an arbitrary unrestricted algorithm to optimize this block. This is possible because this coordinate system introduces a bijection between unrestricted queries and a subset of k-ary unbiased operators. We note that this technique does not depend on OneMax being solved and can be used in more general contexts. This together constitutes an algorithm which is conceptually simpler than the one by Doerr and Doerr, and in the same time achieves better constant multiples in the asymptotic notation. Our algorithm works in (2 + o(1)) · n/(k − 1), where o(1) relates to k. Our experimental evaluation of this algorithm shows its efficiency already for 3 ≤ k ≤ 6.

Original languageEnglish
Title of host publicationGECCO 2018 Companion - Proceedings of the 2018 Genetic and Evolutionary Computation Conference Companion
PublisherAssociation for Computing Machinery, Inc
Pages322-323
Number of pages2
ISBN (Electronic)9781450357647
DOIs
Publication statusPublished - 06 Jul 2018
Externally publishedYes
Event2018 Genetic and Evolutionary Computation Conference, GECCO 2018 - Kyoto, Japan
Duration: 15 Jul 201819 Jul 2018

Publication series

NameGECCO 2018 Companion - Proceedings of the 2018 Genetic and Evolutionary Computation Conference Companion

Conference

Conference2018 Genetic and Evolutionary Computation Conference, GECCO 2018
Country/TerritoryJapan
CityKyoto
Period15 Jul 201819 Jul 2018

Keywords

  • Black-box complexity
  • OneMax
  • Unbiased variation

Fingerprint

Dive into the research topics of 'Better fixed-arity unbiased black-box algorithms'. Together they form a unique fingerprint.

Cite this