Learning a Game Strategy Using Pattern-Weights and Self-play

  • Ari Shapiro
  • Gil Fuchs
  • Robert Levinson
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2883)


This paper demonstrates the use of pattern-weights in order to develop a strategy for an automated player of a non-cooperative version of the game of Diplomacy. Diplomacy is a multi-player, zero-sum and simultaneous move game with imperfect information. Pattern-weights represent stored knowledge of various aspects of a game that are learned through experience. An automated computer player is developed without any initial strategy and is able to learn important strategic aspects of the game through self-play by storing pattern-weights and using temporal difference learning.


Game Play Game Board Game Strategy Heuristic Evaluation Game Graph 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Korf, R.: Problem solving and search. CS261A Course Reader, UCLA (2002) Google Scholar
  2. 2.
    Gould, J., Levinson, R.: Experience-based adaptive search. In: Michalski, R., Tecuci, G. (eds.) Machine Learning 4: A Multi-Strategy Approach, pp. 579–604. Morgan Kaufman Publishers, San Francisco (1994)Google Scholar
  3. 3.
    Hasbro: Rules of Diplomacy, 4th edn. (2000),
  4. 4.
    Kraus, S., Lehmann, D.: Designing and building a negotiating automated agent. Computational Intelligence 11, 132–171 (1995)CrossRefGoogle Scholar
  5. 5.
    Kraus, S., Lehmann, D., Ephrati, E.: An automated Diplomacy player. In: Levy, D., Beal, D. (eds.) Heuristic Programming in Artificial Intelligence, pp. 136–153. Ellis Horwood Limited (1989)Google Scholar
  6. 6.
    Hall, M., Loeb, D.: Thoughts on programming a diplomat. In: van den Herik, J., Allis, V. (eds.) Heuristic Programming in Artificial Intelligence 3, pp. 123–145. Ellis Horwood Limited (1992)Google Scholar
  7. 7.
    Levinson, R.: Experience-based creativity. In: Dartnall, T. (ed.) Artificial Intelligence and Creativity: An Interdisciplinary Approach, pp. 161–179. Kluwer Academic Press, Dordrecht (1994)Google Scholar
  8. 8.
    Levinson, R., Fuchs, G.: A pattern-weight formulation of search knowledge. Computational Intelligence 17, 783–811 (2001)CrossRefGoogle Scholar
  9. 9.
    Tesauro, G.: Temporal difference learning and TD-Gammon. Communications of the ACM 38 (1995)Google Scholar
  10. 10.
    Shapiro, A., Tsafnat, G.: FreeDip, open source software (2002),
  11. 11.
    Nelson, M.: Opening’s custodian report for 1995, (1996)
  12. 12.
    Lorber, S.: Diplomacy AI software (2002),

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Ari Shapiro
    • 1
  • Gil Fuchs
    • 2
  • Robert Levinson
    • 2
  1. 1.Computer Science DepartmentUniversity of CaliforniaLos Angeles
  2. 2.Computer and Information SciencesUniversity of CaliforniaSanta Cruz

Personalised recommendations