An improved representation for evolving programs

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@Article{Withall:2009:GPEM,
  author =       "M. S. Withall and C. J. Hinde and R. G. Stone",
  title =        "An improved representation for evolving programs",
  journal =      "Genetic Programming and Evolvable Machines",
  year =         "2009",
  volume =       "10",
  number =       "1",
  pages =        "37--70",
  month =        mar,
  keywords =     "genetic algorithms, genetic programming, Perl",
  ISSN =         "1389-2576",
  DOI =          "doi:10.1007/s10710-008-9069-7",
  URL =          "http://results.ref.ac.uk/Submissions/Output/2828879",
  size =         "34 pages",
  abstract =     "A representation has been developed that addresses
                 some of the issues with other Genetic Program
                 representations while maintaining their advantages.
                 This combines the easy reproduction of the linear
                 representation with the inheritable characteristics of
                 the tree representation by using fixed-length blocks of
                 genes representing single program statements. This
                 means that each block of genes will always map to the
                 same statement in the parent and child unless it is
                 mutated, irrespective of changes to the surrounding
                 blocks. This method is compared to the variable length
                 gene blocks used by other representations with a clear
                 improvement in the similarity between parent and child.
                 In addition, a set of list evaluation and manipulation
                 functions was evolved as an application of the new
                 Genetic Program components. These functions have the
                 common feature that they all need to be 100percent
                 correct to be useful. Traditional Genetic Programming
                 problems have mainly been optimisation or approximation
                 problems. The list results are good but do highlight
                 the problem of scalability in that more complex
                 functions lead to a dramatic increase in the required
                 evolution time.",
  notes =        "Individuals represented as list of 8-bit integers.
                 Modulus operator used to convert to required range.
                 Perl exec. Checks for infinite for loops (if exceeded
                 programs aborted and given low fitness). Read-only and
                 read-write variables (ie memory). Syntax corrected by
                 post operation fixup. sumlist, avelist, listmax,
                 listmin, reverse, sort. Function set includes Double,
                 If, For, Assign, End, etc. (except two case) 100percent
                 fitness on verification set. Huge impact of Swap on
                 sort.",
  uk_research_excellence_2014 = "D - Journal article",
}

Genetic Programming entries for Mark S Withall Chris J Hinde Roger G Stone

Citations