Ramped Half-n-Half Initialisation Bias in GP

Created by W.Langdon from gp-bibliography.bib Revision:1.4340

  author =       "Edmund Burke and Steven Gustafson and Graham Kendall",
  title =        "Ramped Half-n-Half Initialisation Bias in {GP}",
  booktitle =    "Genetic and Evolutionary Computation -- GECCO-2003",
  editor =       "E. Cant{\'u}-Paz and J. A. Foster and K. Deb and 
                 D. Davis and R. Roy and U.-M. O'Reilly and H.-G. Beyer and 
                 R. Standish and G. Kendall and S. Wilson and 
                 M. Harman and J. Wegener and D. Dasgupta and M. A. Potter and 
                 A. C. Schultz and K. Dowsland and N. Jonoska and 
                 J. Miller",
  year =         "2003",
  pages =        "1800--1801",
  address =      "Chicago",
  publisher_address = "Berlin",
  month =        "12-16 " # jul,
  volume =       "2724",
  series =       "LNCS",
  ISBN =         "3-540-40603-4",
  publisher =    "Springer-Verlag",
  keywords =     "genetic algorithms, genetic programming, poster",
  URL =          "http://www.cs.nott.ac.uk/~smg/research/publications/gecco-poster-2003.ps",
  URL =          "http://www.cs.nott.ac.uk/~smg/research/publications/gecco-poster-2003.pdf",
  DOI =          "doi:10.1007/3-540-45110-2_71",
  abstract =     "Tree initialisation techniques for genetic programming
                 (GP) are examined in [4,3], highlighting a bias in the
                 standard implementation of the initialisation method
                 Ramped Half-n-Half (RHH) [1]. GP trees typically evolve
                 to random shapes, even when populations were initially
                 full or minimal trees [2]. In canonical GP, unbalanced
                 and sparse trees increase the probability that bigger
                 subtrees are selected for recombination, ensuring code
                 growth occurs faster and that subtree crossover will
                 have more difficultly in producing trees within
                 specified depth limits. The ability to evolve tree
                 shapes which allow more legal crossover operations, by
                 providing more possible crossover points (by being
                 bushier), and control code growth is critical. The GP
                 community often uses RHH [4]. The standard
                 implementation of the RHH method selects either the
                 grow or full method with 0.5 probability to produce a
                 tree. If the tree is already in the initial population
                 it is discarded and another is created by grow or full.
                 As duplicates are typically not allowed, this standard
                 implementation of RHH favours full over grow and
                 possibly biases the evolutionary process.",
  notes =        "GECCO-2003. A joint meeting of the twelfth
                 International Conference on Genetic Algorithms
                 (ICGA-2003) and the eighth Annual Genetic Programming
                 Conference (GP-2003)",

Genetic Programming entries for Edmund Burke Steven M Gustafson Graham Kendall