Effects of Tree Size and State Number on GP-Automata Bidding Strategies

Created by W.Langdon from gp-bibliography.bib Revision:1.4504

  author =       "Charles W. {Richter Jr.} and Daniel Ashlock and 
                 Gerald Sheble",
  title =        "Effects of Tree Size and State Number on GP-Automata
                 Bidding Strategies",
  booktitle =    "Genetic Programming 1998: Proceedings of the Third
                 Annual Conference",
  year =         "1998",
  editor =       "John R. Koza and Wolfgang Banzhaf and 
                 Kumar Chellapilla and Kalyanmoy Deb and Marco Dorigo and 
                 David B. Fogel and Max H. Garzon and 
                 David E. Goldberg and Hitoshi Iba and Rick Riolo",
  pages =        "329--337",
  address =      "University of Wisconsin, Madison, Wisconsin, USA",
  publisher_address = "San Francisco, CA, USA",
  month =        "22-25 " # jul,
  publisher =    "Morgan Kaufmann",
  keywords =     "genetic algorithms, genetic programming",
  ISBN =         "1-55860-548-7",
  URL =          "http://www.dakotarichter.com/papers/GPA_GP98.pdf",
  URL =          "http://www.dakotarichter.com/papers/gp98/GP98.html",
  size =         "9 pages",
  abstract =     "The impending deregulation of the electrical industry
                 in the USA promises to open a multi-billion dollar
                 industry to competition. Current research indicates
                 that the double auction will be at the heart of several
                 regional electrical commodity exchanges. The authors
                 are attempting to design comprehensive profitable
                 bidding strategies for traders. The advantages of the
                 strategies detailed here come from using data
                 structures which combine genetic programming and finite
                 state automata termed GP-Automata. Adaptive strategies
                 encoded by two populations of GP-Automata are tested in
                 an auction simulator modelling distribution companies
                 and generation companies buying and selling power via a
                 double auction. In addition to evolving profitable
                 bidding strategies, the resulting strategies can also
                 be designed to imitate certain types of trading
                 behaviours. These strategies can be used directly in
                 on-line trading, or as realistic competitors in an
                 off-line trading simulator. In this paper we report the
                 results of specific experiments which test the effect
                 of changing the size of the GP trees, and the effect of
                 changing the number of states.",
  notes =        "GP-98",

Genetic Programming entries for Charles W Richter Jr Daniel Ashlock Gerald B Sheble