Size of Random Programs to ensure Uniformity

Created by W.Langdon from gp-bibliography.bib Revision:1.4504

  author =       "W. B. Langdon",
  title =        "Size of Random Programs to ensure Uniformity",
  booktitle =    "Proceedings of the Fourteenth Belgium/Netherlands
                 Conference on Artificial Intelligence (BNAIC'02)",
  year =         "2002",
  editor =       "Hendrik Blockeel and Marc Denecker",
  pages =        "459--460",
  address =      "Leuven, Belgium",
  month =        "21-22 " # oct,
  organisation = "BNVKI, Dutch and the Belgian AI Association",
  keywords =     "genetic algorithms, genetic programming",
  URL =          "",
  URL =          "",
  size =         "2 pages",
  abstract =     "Fitness distributions (landscapes) of programs tend to
                 a limit as they get bigger. Markov chain convergence
                 theorems give general upper bounds on the linear
                 program sizes needed for convergence. Tight bounds
                 (exponential in N, N log(N) and smaller) are given in
                 \cite{langdon:2002:crlp} for the outputs of five
                 computer models (any, average, cyclic, bit flip and
                 Boolean). Mutation randomises a genetic algorithm
                 population in 0.25(l+1)(log(l)+4) generations. While
                 \cite{langdon:2002:foga} considers convergence of
                 functions. We restate the results 0.5N(log(m)+4) and
                 O(N)-O(N^{3/2}) for a genetic programming (GP) like
  notes =        "2 page summary of


                 Katholieke Universiteit Leuven and Universite Libre de
                 Bruxelles in collaboration with PharmaDM and under the
                 auspices of BNVKI/AIABN (the Belgian-Dutch Association
                 for Artificial Intelligence), SIKS (School for
                 Information and Knowledge Systems), and SNN (the
                 Foundation for Neural Networks).",

Genetic Programming entries for William B Langdon