Timezone: »

Monte Carlo Sampling for Regret Minimization in Extensive Games
Marc Lanctot · Kevin G Waugh · Martin A Zinkevich · Michael Bowling

Mon Dec 07 07:00 PM -- 11:59 PM (PST) @ None #None

Sequential decision-making with multiple agents and imperfect information is commonly modeled as an extensive game. One efficient method for computing Nash equilibria in large, zero-sum, imperfect information games is counterfactual regret minimization (CFR). In the domain of poker, CFR has proven effective, particularly when using a domain-specific augmentation involving chance outcome sampling. In this paper, we describe a general family of domain independent CFR sample-based algorithms called Monte Carlo counterfactual regret minimization (MCCFR) of which the original and poker-specific versions are special cases. We start by showing that MCCFR performs the same regret updates as CFR on expectation. Then, we introduce two sampling schemes: {\it outcome sampling} and {\it external sampling}, showing that both have bounded overall regret with high probability. Thus, they can compute an approximate equilibrium using self-play. Finally, we prove a new tighter bound on the regret for the original CFR algorithm and relate this new bound to MCCFRs bounds. We show empirically that, although the sample-based algorithms require more iterations, their lower cost per iteration can lead to dramatically faster convergence in various games.

Author Information

Marc Lanctot (University of Alberta)
Kevin G Waugh (University of Alberta)
Martin A Zinkevich (Yahoo! Inc.)
Michael Bowling (DeepMind / University of Alberta)

More from the Same Authors