Advertisement

Crossover in Grammatical Evolution

  • Michael O'Neill
  • Conor Ryan
  • Maarten Keijzer
  • Mike Cattolico
Article

Abstract

We present an investigation into crossover in Grammatical Evolution that begins by examining a biologically-inspired homologous crossover operator that is compared to standard one and two-point operators. Results demonstrate that this homologous operator is no better than the simpler one-point operator traditionally adopted.

An analysis of the effectiveness of one-point crossover is then conducted by determining the effects of this operator, by adopting a headless chicken-type crossover that swaps randomly generated fragments in place of the evolved strings. Experiments show detrimental effects with the utility of the headless chicken operator.

Finally, the mechanism of crossover in GE is analysed and termed ripple crossover, due to its defining characteristics. An experiment is described where ripple crossover is applied to tree-based genetic programming, and the results show that ripple crossover is more effective in exploring the search space of possible programs than sub-tree crossover by examining the rate of premature convergence during the run. Ripple crossover produces populations whose fitness increases gradually over time, slower than, but to an eventual higher level than that of sub-tree crossover.

grammatical evolution genetic programming ripple crossover homologous crossover headless chicken crossover sub-tree crossover 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Kluwer Academic Publishers 2003

Authors and Affiliations

  • Michael O'Neill
    • 1
  • Conor Ryan
    • 1
  • Maarten Keijzer
    • 2
  • Mike Cattolico
    • 3
  1. 1.Department of Computer Science & Information SystemsUniversity of LimerickIreland
  2. 2.Free UniversityAmsterdam
  3. 3.Tiger Mountain Scientific, IncUSA

Personalised recommendations