A study of crossover operators in genetic programming

William M. Spears, Vic Anand

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Holland’s analysis of the sources of power of genetic algorithms has served as guidance for the applications of genetic algorithms for more than 15 years. The technique of applying a recombination operator (crossover) to a population of individuals is a key to that power. Neverless, there have been a number of contradictory results concerning crossover operators with respect to overall performance. Recently, for example, genetic algorithms were used to design neural network modules and their control circuits. In these studies, a genedc algorithm without crossover outperformed a genetic algorithm with crossover. This report re-examines these studies, and concludes that the results were caused by a small population size. New results are presented that illustrate the effectiveness of crossover when the population size is larger. From a performance view, the results indicate that better neural networks can be evolved in a shorter time if the genetic algorithm uses crossover.

Original languageEnglish (US)
Title of host publicationMethodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings
EditorsZbigniew W. Ras, Maria Zemankova
PublisherSpringer
Pages409-418
Number of pages10
ISBN (Print)9783540545637
DOIs
StatePublished - 1991
Externally publishedYes
Event6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991 - Charlotte , United States
Duration: Oct 16 1991Oct 19 1991

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume542 LNAI Part F2
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991
Country/TerritoryUnited States
CityCharlotte
Period10/16/9110/19/91

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'A study of crossover operators in genetic programming'. Together they form a unique fingerprint.

Cite this