Notice
This is not the latest version of this item. The latest version can be found at:https://dspace.mit.edu/handle/1721.1/142962.2
World-class interpretable poker
Author(s)
Bertsimas, Dimitris; Paskov, Alex
Download10994_2022_Article_6179.pdf (1.547Mb)
Publisher with Creative Commons License
Publisher with Creative Commons License
Creative Commons Attribution
Terms of use
Metadata
Show full item recordAbstract
Abstract
We address the problem of interpretability in iterative game solving for imperfect-information games such as poker. This lack of interpretability has two main sources: first, the use of an uninterpretable feature representation, and second, the use of black box methods such as neural networks, for the fitting procedure. In this paper, we present advances on both fronts. Namely, first we propose a novel, compact, and easy-to-understand game-state feature representation for Heads-up No-limit (HUNL) Poker. Second, we make use of globally optimal decision trees, paired with a counterfactual regret minimization (CFR) self-play algorithm, to train our poker bot which produces an entirely interpretable agent. Through experiments against Slumbot, the winner of the most recent Annual Computer Poker Competition, we demonstrate that our approach yields a HUNL Poker agent that is capable of beating the Slumbot. Most exciting of all, the resulting poker bot is highly interpretable, allowing humans to learn from the novel strategies it discovers.
Date issued
2022-06-09Publisher
Springer US
Citation
Bertsimas, Dimitris and Paskov, Alex. 2022. "World-class interpretable poker."
Version: Final published version