KnightCap: a chess program that learns by combining TD(λ) with game-tree search Export

@inproceedings{citeulike:1989179, abstract = {In this paper we present TDLeaf(), a variation on the TD() algorithm that enables it to be used in conjunction with minimax search. We present some experiments in which our chess program, "KnightCap," used TDLeaf() 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...}, author = {Baxter, Jonathan and Trigdell, Andrew and Weaver, Lex}, booktitle = {Proc. 15th International Conf. on Machine Learning}, citeulike-article-id = {1989179}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.54.8263}, keywords = {chess, game\_playing, minimax\_search, reinforcement\_learning, temporal\_difference}, pages = {28--36}, posted-at = {2007-11-26 23:32:01}, priority = {2}, publisher = {Morgan Kaufmann, San Francisco, CA}, title = {KnightCap: a chess program that learns by combining TD(λ) with game-tree search}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.54.8263}, year = {1998} }

TY - CONF ID - citeulike:1989179 L3 - citeulike-article-id:1989179 N2 - In this paper we present TDLeaf(), a variation on the TD() algorithm that enables it to be used in conjunction with minimax search. We present some experiments in which our chess program, "KnightCap," used TDLeaf() 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... EP - 36 TI - KnightCap: a chess program that learns by combining TD(λ) with game-tree search PB - Morgan Kaufmann, San Francisco, CA T2 - Proc. 15th International Conf. on Machine Learning SP - 28 KW - chess KW - game_playing KW - minimax_search KW - reinforcement_learning KW - temporal_difference AU - Baxter, Jonathan AU - Trigdell, Andrew AU - Weaver, Lex PY - 1998/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.54.8263 ER -