Inductive Genetic Programming with Immune Network Dynamics

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "Nikolay I. Nikolaev and Hitoshi Iba and Vanio Slavov",
  title =        "Inductive Genetic Programming with Immune Network
  booktitle =    "Advances in Genetic Programming 3",
  publisher =    "MIT Press",
  year =         "1999",
  editor =       "Lee Spector and William B. Langdon and 
                 Una-May O'Reilly and Peter J. Angeline",
  chapter =      "15",
  pages =        "355--376",
  address =      "Cambridge, MA, USA",
  month =        jun,
  keywords =     "genetic algorithms, genetic programming",
  ISBN =         "0-262-19423-6",
  URL =          "",
  size =         "22 pages",
  abstract =     "This chapter presents an immune version of Genetic
                 Programming (GP). This is a GP version that conducts
                 progressive search controlled by a dynamic fitness
                 function. The new fitness function is based on analogy
                 with a model of the biological immune system, such that
                 the programs are viewed as lymphocyte clones that
                 compete to recognize most of the examples, viewed as
                 antigens. The programs are reinforced with rewards for
                 matched important examples and stimulated to match
                 different examples. Examples recognized by a small
                 number of programs are considered important. The
                 motivation for using the immune dynamics for GP
                 navigation is to maintain a high population diversity
                 and to achieve enhanced search performance. Empirical
                 evidence for the efficacy of this immune version on
                 practical inductive machine learning and time-series
                 prediction tasks is provided",
  notes =        "AiGP3 See",

Genetic Programming entries for Nikolay Nikolaev Hitoshi Iba Vanio Slavov