Abstract
Computer chess research has traditionally focused on creating the strongest possible chess engine. Recently, however, attempts have been made to create engines that mimic the playing strength and style of human players. Our research proposes enhancements of models developed in this vein that more accurately imitate master-level players, as well as improve the prediction accuracy of existing models on weaker players. Our proposed enhancements are simple to apply by post-processing the output of existing chess engines. The performance of our enhancements was evaluated and compared using two metrics, prediction accuracy and average centipawn loss. We found that using an ensemble model over search depths maximised prediction accuracy, while an evaluation window filtering approach was preferable with respect to average centipawn loss.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
- 3.
In this paper, “depth” refers to the average depth of the MCTS tree, as described in the Leela chess documentation (https://lczero.org/dev/wiki/technical-explanation-of-leela-chess-zero/).
- 4.
- 5.
- 6.
Portable Game Notation: https://www.chessprogramming.org/Portable_Game_Notation.
- 7.
- 8.
A ply is a single move made by one of the players.
- 9.
- 10.
- 11.
- 12.
- 13.
It is necessary to average the absolute difference since engine evaluations are always from white’s perspective.
- 14.
In other words, the accuracy obtained when calculated using correct move predictions up to a given depth. For instance, the cumulative accuracy at depth 2 includes all the correct predictions at depth 1, as well as the extra correct predictions at depth 2.
- 15.
In the case of a tie, the point was shared.
References
Ferreira, D.: Determining the strength of chess players based on actual play. ICGA J. 35, 3–19 (2012)
Jacob, A.P., et al.: Modeling strong and human-like gameplay with KL-regularized search. In: Chaudhuri, K., Jegelka, S., Song, L., Szepesvari, C., Niu, G., Sabato, S. (eds.) Proceedings of the 39th International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 162. PMLR, 17–23 July 2022. https://proceedings.mlr.press/v162/jacob22a.html
Kiefer, J.: Sequential minimax search for a maximum. Proc. Am. Math. Soc. 4(3) (1953). http://www.jstor.org/stable/2032161
McCarthy, J.: Chess as the drosophila of AI. In: Marsland, T.A., Schaeffer, J. (eds.) Computers, Chess, and Cognition. Springer, New York (1990). https://doi.org/10.1007/978-1-4613-9080-0_14
McIlroy-Young, R., Sen, S., Kleinberg, J., Anderson, A.: Aligning superhuman AI with human behavior: chess as a model system. In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. KDD ’20, Association for Computing Machinery, New York, NY, USA (2020). https://doi.org/10.1145/3394486.3403219
McIlroy-Young, R., Wang, R., Sen, S., Kleinberg, J., Anderson, A.: Learning models of individual behavior in chess. In: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. KDD ’22. Association for Computing Machinery, New York, NY, USA (2022). https://doi.org/10.1145/3534678.3539367
Munos, R.: From bandits to Monte-Carlo tree search: the optimistic principle applied to optimization and planning. Found. Trends® Mach. Learn. 7(1) (2014)
Rosin, C.: Multi-armed bandits with episode context. Ann. Math. Artif. Intell. 61 (2010). https://doi.org/10.1007/s10472-011-9258-6
Silver, D., et al.: A general reinforcement learning algorithm that masters chess, shogi, and go through self-play. Science 362(6419) (2018). https://doi.org/10.1126/science.aar6404
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Barrish, D., Kroon, S., van der Merwe, B. (2024). Making Superhuman AI More Human in Chess. In: Hartisch, M., Hsueh, CH., Schaeffer, J. (eds) Advances in Computer Games. ACG 2023. Lecture Notes in Computer Science, vol 14528. Springer, Cham. https://doi.org/10.1007/978-3-031-54968-7_1
Download citation
DOI: https://doi.org/10.1007/978-3-031-54968-7_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-54967-0
Online ISBN: 978-3-031-54968-7
eBook Packages: Computer ScienceComputer Science (R0)