Abstract
A common evaluation function for playing Chinese Checkers with two or more players has been the single-agent distance across the board. This is an abstraction of a perfect heuristic, because it ignores the interactions between the players in the game. Previous work has studied these heuristics for smaller versions of the game, including 6-piece data for a board with 49 locations and 81 locations which have 13.98 million and 324.5 million combinations respectively. The single-agent solution to the full game of Chinese Checkers has 81 locations and 10 pieces per player. This results in 1.88 trillion possible positions and is stored using 500 GB of disk space. In this paper we report results from a preliminary study on how to best use the data to improve the play of a Chinese Checkers program.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Due to symmetry either the start or the goal state can be used, although the choice influences whether subsets of data can be efficiently loaded.
- 2.
In our original implementation the algorithms were indistinguishable. Improving the efficiency of our TD learning, by taking advantage of the binary features, significantly improved the performance of the TD player.
- 3.
It can change by more than one because we may jump over an opponent’s piece, which is not accounted for in the single-agent data.
References
8 longest 7-man checkmates. http://tb7.chessok.com/articles/Top8DTM_eng. Acces-sed 11 May 2015
Bell, G.I.: The shortest game of Chinese Checkers and related problems. CoRR abs/0803.1245 (2008). http://arxiv.org/abs/0803.1245
Bouzy, B.: Old-fashioned computer Go vs Monte-Carlo Go. In: IEEE Symposium on Computational Intelligence in Games (CIG) (2007). Invited Tutorial
Browne, C., Powley, E.J., Whitehouse, D., Lucas, S.M., Cowling, P.I., Rohlfshagen, P., Tavener, S., Perez, D., Samothrakis, S., Colton, S.: A survey of monte carlo tree search methods. IEEE Trans. Comput. Intell. AI Games 4(1), 1–43 (2012)
Campbell, M., Hoane Jr., A.J., Hsu, F.: Deep blue. Artif. Intell. 134(1–2), 57–83 (2002)
Chaslot, G.M.J.B., Winands, M.H.M., van den Herik, H.J., Uiterwijk, J.W.H.M., Bouzy, B.: Progressive strategies for Monte-Carlo tree search. New Math. Nat. Comput. 4(3), 343–357 (2008)
Fang, H., Hsu, T., Hsu, S.-C.: Construction of Chinese chess endgame databases by retrograde analysis. In: Marsland, T., Frank, I. (eds.) CG 2001. LNCS, vol. 2063, pp. 96–114. Springer, Heidelberg (2002)
Gelly, S., Silver, D.: Combining online and offline knowledge in UCT. In: Ghahramani, Z. (ed.) Machine Learning. ACM International Conference Proceeding Series, vol. 227, pp. 273–280. ACM, New York (2007)
Gelly, S., Silver, D.: Achieving master level play in 9 x 9 computer go. In: Fox, D., Gomes, C.P. (eds.) AAAI, pp. 1537–1540. AAAI Press, Menlo Park (2008)
Huang, S.-C., Arneson, B., Hayward, R.B., Müller, M., Pawlewicz, J.: MoHex 2.0: a pattern-based MCTS hex player. In: van den Herik, H.J., Iida, H., Plaat, A. (eds.) CG 2013. LNCS, vol. 8427, pp. 60–71. Springer, Heidelberg (2014)
Hutton, A.: Developing Computer Opponents for Chinese Checkers. Master’s thesis, University of Glasgow (2001)
Kocsis, L., Szepesvári, C.: Bandit based Monte-Carlo planning. In: Fürnkranz, J., Scheffer, T., Spiliopoulou, M. (eds.) ECML 2006. LNCS (LNAI), vol. 4212, pp. 282–293. Springer, Heidelberg (2006)
Lanctot, M., Winands, M.H.M., Pepels, T., Sturtevant, N.R.: Monte Carlo tree search with heuristic evaluations using implicit minimax backups. In: 2014 IEEE Conference on Computational Intelligence and Games, CIG 2014, Dortmund, Germany, 26–29 August 2014, pp. 341–348. IEEE (2014)
Lorentz, R.J.: Amazons discover Monte-Carlo. In: van den Herik, H.J., Xu, X., Ma, Z., Winands, M.H.M. (eds.) CG 2008. LNCS, vol. 5131, pp. 13–24. Springer, Heidelberg (2008)
Nalimov, E., Haworth, G.M., Heinz, E.A.: Space-efficient indexing of endgame tables for chess. ICGA J. 23(3), 148–162 (2000)
Nijssen, J.P.A.M., Winands, M.H.M.: Enhancements for multi-player Monte-Carlo tree search. In: van den Herik, H.J., Iida, H., Plaat, A. (eds.) CG 2010. LNCS, vol. 6515, pp. 238–249. Springer, Heidelberg (2011)
Pearl, J.: The solution for the branching factor of the alpha-beta pruning algorithm and its optimality. Commun. ACM 25(8), 559–564 (1982)
Roschke, M., Sturtevant, N.R.: UCT enhancements in Chinese Checkers using an endgame database. In: Cazenave, T., Winands, M.H.M., Iida, H. (eds.) Computer Games (CGW 2013). CCIS, vol. 408, pp. 57–70. Springer International Publishing, Switzerland (2014)
Samadi, M., Asr, F.T., Schaeffer, J., Azimifar, Z.: Extending the applicability of pattern and endgame databases. IEEE Trans. Comput. Intell. AI Games 1(1), 28–38 (2009)
Samadi, M., Schaeffer, J., Asr, F.T., Samar, M., Azimifar, Z.: Using abstraction in two-player games. In: ECAI, pp. 545–549 (2008)
Schaeffer, J.: The history heuristic and alpha-beta search enhancements in practice. IEEE Trans. Pattern Anal. Mach. Intell. 11(11), 1203–1212 (1989)
Schaeffer, J.: One Jump Ahead - Challenging Human Supremacy in Checkers. Springer, New York (1997)
Schaeffer, J., Björnsson, Y., Burch, N., Lake, R., Lu, P., Sutphen, S.: Building the checkers 10-piece endgame databases. Adv. Comput. Games 10, 193–210 (2003)
Sturtevant, N.R., Rutherford, M.J.: Minimizing writes in parallel external memory search. In: International Joint Conference on Artificial Intelligence (IJCAI) (2013)
Sturtevant, N.R.: A comparison of algorithms for multi-player games. In: Schaeffer, J., Müller, M., Björnsson, Y. (eds.) CG 2002. LNCS, vol. 2883, pp. 108–122. Springer, Heidelberg (2003)
Sturtevant, N.R.: An analysis of UCT in multi-player games. ICGA J. 31(4), 195–208 (2008)
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)
Tong, K.B.: Intelligent Strategy for Two-person Non-random Perfect Information Zero-sum Game. Master’s thesis, Chinese University of Hong Kong (2003)
Acknowledgments
This paper benefited from research by a summer student, Evan Boucher, who worked on the problem of determining the true distance of a state from the goal given the modulo distance.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Sturtevant, N.R. (2016). Challenges and Progress on Using Large Lossy Endgame Databases in Chinese Checkers. In: Cazenave, T., Winands, M., Edelkamp, S., Schiffel, S., Thielscher, M., Togelius, J. (eds) Computer Games. CGW GIGA 2015 2015. Communications in Computer and Information Science, vol 614. Springer, Cham. https://doi.org/10.1007/978-3-319-39402-2_1
Download citation
DOI: https://doi.org/10.1007/978-3-319-39402-2_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-39401-5
Online ISBN: 978-3-319-39402-2
eBook Packages: Computer ScienceComputer Science (R0)