Neutral Variations Cause Bloat in Linear GP

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

  author =       "Markus Brameier and Wolfgang Banzhaf",
  title =        "Neutral Variations Cause Bloat in Linear GP",
  booktitle =    "Genetic Programming, Proceedings of EuroGP'2003",
  year =         "2003",
  editor =       "Conor Ryan and Terence Soule and Maarten Keijzer and 
                 Edward Tsang and Riccardo Poli and Ernesto Costa",
  volume =       "2610",
  series =       "LNCS",
  pages =        "286--296",
  address =      "Essex",
  publisher_address = "Berlin",
  month =        "14-16 " # apr,
  organisation = "EvoNet",
  publisher =    "Springer-Verlag",
  keywords =     "genetic algorithms, genetic programming: Poster",
  ISBN =         "3-540-00971-X",
  URL =          "",
  DOI =          "doi:10.1007/3-540-36599-0_26",
  abstract =     "In this contribution we investigate the influence of
                 different variation effects on the growth of code. A
                 mutation-based variant of linear GP is applied that
                 operates with minimum structural step sizes. Results
                 show that neutral variations are a direct cause for
                 (and not only a result of) the emergence and the growth
                 of intron code. The influence of non-neutral variations
                 has been found to be considerably smaller. Neutral
                 variations turned out to be beneficial by solving two
                 classification problems more successfully.",
  notes =        "EuroGP'2003 held in conjunction with EvoWorkshops

                 Section 2.3 PerlGP 'In PerlGP \cite{maccallum03},
                 evolved code is expanded from a tree-based genotype
                 into a string before being evaluated with Perl's eval()
                 function. The trees of each individual are built (and
                 later, mutated) according to a grammar and are strongly
                 typed. In this application, we want the evolved code to
                 look like the example given in Figure 3; that is to
                 say, the solution should be some arithmetic expression
                 containing constants and RE matches against a protein
                 sequence. The matches() function feeds the number of
                 separate RE matches into the arithmetic expression. If
                 the result of the expression for a given sequence is
                 greater than zero, it is predicted/classified as
                 nuclear, otherwise it is non-nuclear.'


Genetic Programming entries for Markus Brameier Wolfgang Banzhaf