Empirical Comparison of Incremental Reuse Strategies in Genetic Programming for Keep-Away Soccer

Created by W.Langdon from gp-bibliography.bib Revision:1.4221

  author =       "William H. Hsu and Scott J. Harmon and 
                 Edwin Rodriguez and Christopher Zhong",
  title =        "Empirical Comparison of Incremental Reuse Strategies
                 in Genetic Programming for Keep-Away Soccer",
  booktitle =    "Late Breaking Papers at the 2004 Genetic and
                 Evolutionary Computation Conference",
  year =         "2004",
  editor =       "Maarten Keijzer",
  address =      "Seattle, Washington, USA",
  month =        "26 " # jul,
  keywords =     "genetic algorithms, genetic programming",
  URL =          "http://www.cs.bham.ac.uk/~wbl/biblio/gecco2004/LBP010.pdf",
  abstract =     "Easy missions approaches to machine learning seek to
                 synthesise solutions for complex tasks from those for
                 simpler ones. In genetic programming, this has been
                 achieved by identifying goals and fitness functions for
                 subproblems of the overall problem. Solutions evolved
                 for these subproblems are then reused to speed up
                 learning, either as automatically defined functions
                 (ADFs) or by seeding a new GP population. Previous
                 positive results using both approaches for learning in
                 multi-agent systems (MAS) showed that incremental reuse
                 using easy missions achieves comparable or better
                 overall fitness than monolithic simple GP. A key
                 unresolved issue dealt with hybrid reuse using ADF plus
                 easy missions. Results in the keep-away soccer domain
                 (a test bed for MAS learning) were also inconclusive on
                 whether compactness inducing reuse helped or hurt
                 overall agent performance. In this paper, we compare
                 monolithic (simple GP and GP with ADFs) and easy
                 missions reuse to two types of GP learning systems with
                 incremental reuse: GP/ADF hybrids with easy missions
                 and single-mission incremental ADFs. As hypothesised,
                 pure easy missions reuse achieves results competitive
                 with the best hybrid approaches in this domain. We
                 interpret this finding and suggest a theoretical
                 approach to characterising incremental reuse and code
  notes =        "Part of \cite{keijzer:2004:GECCO:lbp}",

Genetic Programming entries for William H Hsu Scott J Harmon Edwin Rodriguez Christopher Zhong