Numerical-node building block analysis of genetic programming with simplification

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

@TechReport{Wong:2006:CS-TR-06-15,
  author =       "Phillip Wong and Mengjie Zhang",
  title =        "Numerical-node building block analysis of genetic
                 programming with simplification",
  institution =  "Computer Science, Victoria University of Wellington",
  year =         "2006",
  number =       "CS-TR-06-15",
  address =      "New Zealand",
  month =        dec,
  keywords =     "genetic algorithms, genetic programming,
                 Simplification, building blocks, numerical-nodes",
  URL =          "http://www.mcs.vuw.ac.nz/comp/Publications/archive/CS-TR-06/CS-TR-06-15.pdf",
  URL =          "http://www.mcs.vuw.ac.nz/comp/Publications/archive/CS-TR-06/CS-TR-06-15.ps.gz",
  abstract =     "This paper investigates the effects on building blocks
                 of using simplification in a GP system to combat the
                 problem of code bloat. The evolved genetic programs are
                 simplified online during the evolutionary process using
                 algebraic simplification rules and hashing techniques.
                 A simplified form of building block (numerical-nodes)
                 are tracked throughout several individual GP runs both
                 when using and not using simplification. The results
                 suggest that simplification disrupts existing potential
                 building blocks during the evolution process. However,
                 the result s also suggest that simplification is
                 capable of creating new building blocks which are used
                 to form a more accurate solution than the standard GP.
                 The effectiveness of GP systems simplification can be
                 correlated to the creation of these new building
                 blocks.",
  size =         "17 pages",
}

Genetic Programming entries for Phillip Wong Mengjie Zhang

Citations