Code Growth in Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.2031

@PhdThesis{soule:thesis,
  author =       "Terence Soule",
  title =        "Code Growth in Genetic Programming",
  school =       "University of Idaho",
  year =         "1998",
  address =      "Moscow, Idaho, USA",
  month =        "15 " # may,
  keywords =     "genetic algorithms, genetic programming, bloat",
  URL =          "http://www.cs.uidaho.edu/~tsoule/research/the3.ps",
  size =         "101 pages",
  abstract =     "Genetic programming is a technique for the automatic
                 generation of computer programs loosely based on the
                 theory of evolution. It has produced successful
                 solutions to a wide variety of problems and can be
                 effective even in noisy and changing
                 environments.

                 However, genetic programming produces solutions with
                 large amounts of unnecessary code. The amount of
                 unnecessary code increases over time and is not
                 proportional to increases in the quality of the
                 solutions produced. Thus, this additional code
                 seriously hinders the genetic programming processes by
                 requiring extra resources without producing equivalent
                 returns.

                 This dissertation examines the causes of this code
                 growth. We use three test problems from very different
                 fields of interest to confirm the generality of the
                 results. We tested the destructive hypothesis, that
                 code growth is a protective response to the
                 destructiveness of crossover, as a potential cause of
                 code growth. It is a definite cause, but is not
                 sufficient to explain all growth. We propose a second
                 cause of code growth removal bias to explain the
                 remaining growth. Testing shows that removal bias does
                 occur and that it produces growth sufficient to explain
                 the discrepancy. We also examine the relationship
                 between code size and code shape, demonstrating that
                 sparser program trees produce more rapid growth.
                 Finally, we examine parsimony pressure as a potential
                 solution to the code growth phenomenon.",
}

Genetic Programming entries for Terence Soule