摘要: We present an advanced intelligent system for playing three-player tarok card game. The is based on alpha-beta search with several enhancements such as fuzzy transposition table, which clusters strategically similar positions into generalised game states. Unknown distribution of other players' cards addressed by Monte Carlo sampling. Experimental results show additional reduction in size the expanded 9-ply game-tree a factor 184. Human players judge resulting program to play reasonably well.
core.ac.uk
bocosoft.com 参考文章(5)
Ian Frank, David Basin, A theoretical and empirical investigation of search in imperfect information games Theoretical Computer Science. ,vol. 252, pp. 217- 256 ,(2001) , 10.1016/S0304-3975(00)00083-9
J. Schaeffer, The history heuristic and alpha-beta search enhancements in practice IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 11, pp. 1203- 1212 ,(1989) , 10.1109/34.42858
Darse Billings, Aaron Davidson, Jonathan Schaeffer, Duane Szafron, The challenge of poker Artificial Intelligence. ,vol. 134, pp. 201- 240 ,(2002) , 10.1016/S0004-3702(01)00130-8
T.A. Marsland, A REVIEW OF GAME-TREE PRUNING† ICGA Journal. ,vol. 9, pp. 3- 19 ,(1986) , 10.3233/ICG-1986-9102
M. L. Ginsberg, GIB: imperfect information in a computationally challenging game Journal of Artificial Intelligence Research. ,vol. 14, pp. 303- 358 ,(2001) , 10.1613/JAIR.820