Temporal Difference Learning and TD-Gammon

Communications of the ACM

Volumn 38, Issue 3, 1995, Pages 58-68

  Tesau, Covid ^a   Tesau, Gerald ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

ARTIFICIAL INTELLIGENCE; BACKPROPAGATION; ERRORS; GAME THEORY; HEURISTIC PROGRAMMING; KNOWLEDGE ENGINEERING; LEARNING SYSTEMS; NEURAL NETWORKS;

BOARD GAMES; COMPUTER GAMES; GAME LEARNING PROGRAM; NEURAL NETWORK TRAINING; REINFORCEMENT LEARNING; TD-GAMMON (SOFTWARE PACKAGE); TEMPORAL DIFFERENCE LEARNING;

COMPUTER SOFTWARE;

EID: 0029276036     PISSN: 00010782     EISSN: 15577317     Source Type: Journal    
DOI: 10.1145/203330.203343     Document Type: Article

References (6)

1. Berliner, 11. 72 64 1 Sci. Amer. 243 Computer backgammon 1980
2. Epstein, S, Lowards an ideal trainer Mach. Learning 15, 8, (1994), 251–277.
3. Fahlman, S. E and Lebiere, C. Hie cascade-correlation learning architecture. In 1). S. Touretzky, Ed. Advances in Neural Information Processing Systems 2, Morgan Kaufmann, San Mateo, Calif., (1990), 524–532.
4. Fawcett, T, E. and Utgoff, I*. E. Automatic feature generation for problem solving systems. In i>. Sieeman and P Edwards Eds., Machine Learning: Proceedings of the Ninth International Workshop, Morgan Kaufmann, San Mateo, Calif., 1992, 144–153.
5. Hornik, K., Stinchcombe, M, and White, 1L Multilayer feedforward networks are universal approximators. Neural Networks 2, (1989), 359–366.
6. 6, Isabelle, J.-F. Auto-apprentissage, a Paide de reseaux de neurones, de fonctions heuristiques ulilisees dans les jeux strategiques. Master's thesis