Counterfactual Regret Minimization in Sequential Security Games

Authors

  • Viliam Lisy University of Alberta
  • Trevor Davis University of Alberta
  • Michael Bowling University of Alberta

DOI:

https://doi.org/10.1609/aaai.v30i1.10051

Keywords:

counterfactual regret minimization, sequential game, imperfect information, extensive form game

Abstract

Many real world security problems can be modelled as finite zero-sum games with structured sequential strategies and limited interactions between the players. An abstract class of games unifying these models are the normal-form games with sequential strategies (NFGSS). We show that all games from this class can be modelled as well-formed imperfect-recall extensive-form games and consequently can be solved by counterfactual regret minimization. We propose an adaptation of the CFR+ algorithm for NFGSS and compare its performance to the standard methods based on linear programming and incremental game generation. We validate our approach on two security-inspired domains. We show that with a negligible loss in precision, CFR+ can compute a Nash equilibrium with five times less computation than its competitors.

Downloads

Published

2016-02-21

How to Cite

Lisy, V., Davis, T., & Bowling, M. (2016). Counterfactual Regret Minimization in Sequential Security Games. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1). https://doi.org/10.1609/aaai.v30i1.10051

Issue

Section

Technical Papers: Game Theory and Economic Paradigms