Corpus ID: 2935600

KnightCap: A chess program that learns by combining TD( ) with minimax search

@inproceedings{Baxter1997KnightCapAC,
  title={KnightCap: A chess program that learns by combining TD( ) with minimax search},
  author={J. Baxter and A. Tridgell and Lex Weaver},
  booktitle={ICML 1997},
  year={1997}
}
  • J. Baxter, A. Tridgell, Lex Weaver
  • Published in ICML 1997
  • Computer Science
  • In this paper we present TDLeaf( ), a variation on the TD( ) algorithm that enables it to be used in conjunction with minimax search. We p resent some experiments in which our chess program, “KnightCap,” used TDL eaf( ) to learn its evaluation function while playing on the Free Ineternet Chess Server (FICS, fics.onenet.net). It improved from a 1650 rating to a 2100 rating in just 308 games and 3 days of play. We discuss some of the reasons for this success and also the relationship between our… CONTINUE READING
    18 Citations

    Figures and Topics from this paper

    Explore Further: Topics Discussed in This Paper

    Using Reinforcement Learning in Chess Engines
    • 11
    • Highly Influenced
    • PDF
    Learning to play chess using TD ( λ )-learning with database games
    • 4
    • PDF
    Learning to Play Board Games using Temporal Dierence Methods
    • 12
    • PDF
    Tuning evaluation functions by maximizing concordance
    • 19
    • PDF
    Programming backgammon using self-teaching neural nets
    • 165
    • PDF
    Learning to Play Draughts using temporal difference learning with neural networks and databases
    • 13
    • PDF
    Self-Play and Using an Expert to Learn to Play Backgammon with Temporal Difference Learning
    • M. Wiering
    • Computer Science
    • J. Intell. Learn. Syst. Appl.
    • 2010
    • 28
    • PDF
    Applying Artificial Neural Network combined with TD(λ) to computer Chinese chess
    • 1

    References

    SHOWING 1-10 OF 41 REFERENCES
    TD-Gammon, a Self-Teaching Backgammon Program, Achieves Master-Level Play
    • 768
    • PDF
    Learning to Play the Game of Chess
    • 190
    • PDF
    Best-First Fixed-Depth Minimax Algorithms
    • 90
    • PDF
    Coevolution of a Backgammon Player
    • 130
    • PDF
    Practical Issues in Temporal Difference Learning
    • 423
    Reinforcement Learning: A Survey
    • 6,648
    • PDF
    Analysis of Temporal-Diffference Learning with Function Approximation
    • 1,253
    • PDF
    Neuro-Dynamic Programming
    • 4,382
    An Analysisof TemporalDif ferenceLearningwith FunctionApproximation
    • IEEE Transactionson AutomaticControl,
    • 1997