The relationship between evolvability and bloat

Jeffrey K. Bassett, Mark Coletti, Kenneth A. De Jong

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

2 Scopus citations

Abstract

Bloat is a common problem with Evolutionary Algorithms (EAs) that use variable length representation. By creating unnecessarily large individuals it results in longer EA runtimes and solutions that are difficult to interpret. The causes of bloat are still uncertain, but one theory suggests that it occurs when the phenotype (e.g. behaviors) of the parents are not successfully inherited by their offspring. Noting the similarity to evolvability theory, which measures heritability of fitness, we hypothesize that reproductive operators with high evolvability will be less likely to cause bloat. We set out to design a new crossover operator for Pittsburgh approach classifier systems that has high phenotypic heritability. We saw an opportunity using the nearest neighbor representation to perform crossover cuts in phenotype space rather than on the genomes. We demonstrate that our operator tends to be less susceptible to bloat and has higher evolvability than a standard Pittsburgh approach crossover operator. Our hope is that this will lead to a general approach to reducing bloat for any representation.

Original languageEnglish
Title of host publicationProceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
Pages1899-1900
Number of pages2
DOIs
StatePublished - 2009
Externally publishedYes
Event11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009 - Montreal, QC, Canada
Duration: Jul 8 2009Jul 12 2009

Publication series

NameProceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009

Conference

Conference11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
Country/TerritoryCanada
CityMontreal, QC
Period07/8/0907/12/09

Keywords

  • Bloat
  • Evolvability
  • Homologous crossover
  • Learning classifier systems
  • Pittsburgh approach

Fingerprint

Dive into the research topics of 'The relationship between evolvability and bloat'. Together they form a unique fingerprint.

Cite this