Adaptive drift analysis

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

Abstract

We show that the (1+1) evolutionary algorithm using an arbitrary mutation rate p = c/n, c a constant, finds the optimum of any n-bit pseudo-Boolean linear function f in expected time Θ(n log n). Since previous work shows that universal drift functions cannot exist for c larger than a certain constant, we define drift functions depending on p and f. This seems to be the first time in the theory of evolutionary algorithms that drift functions are used that take into account the particular problem instance.

Original languageEnglish
Title of host publicationParallel Problem Solving from Nature, PPSN XI - 11th International Conference, Proceedings
Pages32-41
Number of pages10
EditionPART 1
DOIs
Publication statusPublished - 12 Nov 2010
Externally publishedYes
Event11th International Conference on Parallel Problem Solving from Nature, PPSN 2010 - Krakow, Poland
Duration: 11 Sept 201015 Sept 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 1
Volume6238 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference11th International Conference on Parallel Problem Solving from Nature, PPSN 2010
Country/TerritoryPoland
CityKrakow
Period11/09/1015/09/10

Fingerprint

Dive into the research topics of 'Adaptive drift analysis'. Together they form a unique fingerprint.

Cite this