Crossover in Grammatical Evolution: A Smooth Operator?

Created by W.Langdon from gp-bibliography.bib Revision:1.4524

  author =       "Michael O'Neill and Conor Ryan",
  title =        "Crossover in Grammatical Evolution: A Smooth
  booktitle =    "Genetic Programming, Proceedings of EuroGP'2000",
  year =         "2000",
  editor =       "Riccardo Poli and Wolfgang Banzhaf and 
                 William B. Langdon and Julian F. Miller and Peter Nordin and 
                 Terence C. Fogarty",
  volume =       "1802",
  series =       "LNCS",
  pages =        "149--162",
  address =      "Edinburgh",
  publisher_address = "Berlin",
  month =        "15-16 " # apr,
  organisation = "EvoNet",
  publisher =    "Springer-Verlag",
  keywords =     "genetic algorithms, genetic programming, grammatical
  ISBN =         "3-540-67339-3",
  URL =          "",
  URL =          "",
  DOI =          "doi:10.1007/978-3-540-46239-2_11",
  abstract =     "Grammatical Evolution is an evolutionary algorithm
                 which can produce code in any language, requiring as
                 inputs a BNF grammar definition describing the output
                 language, and the fitness function. The usefulness of
                 crossover in GP systems has been hotly debated for some
                 time, and this debate has also arisen with respect to
                 Grammatical Evolution. This paper serves to analyse the
                 crossover operator in our algorithm by comparing the
                 performance of a variety of crossover operators.
                 Results show that the standard one point crossover
                 employed by Grammatical Evolution is not as destructive
                 as it might originally appear, and is useful in
                 performing a global search over the course of entire
                 runs. This is attributed to the fact that prior to the
                 crossover event the parent chromosomes undergo
                 alignment which facilitates the swapping of blocks
                 which are more likely to be in context.",
  notes =        "EuroGP'2000, part of \cite{poli:2000:GP}",

Genetic Programming entries for Michael O'Neill Conor Ryan