Deep Counterfactual Regret Minimization

TL;DR

This paper introduces Deep CFR, a neural network-based approach that eliminates the need for game abstraction in large imperfect-information games, achieving strong results in poker.

Contribution

It presents the first successful non-tabular CFR variant using deep learning to directly approximate strategies in large games.

Findings

Deep CFR outperforms traditional tabular CFR in large poker games.

The method effectively approximates CFR behavior without manual abstraction.

Deep CFR demonstrates strong convergence properties in complex game settings.

Abstract

Counterfactual Regret Minimization (CFR) is the leading framework for solving large imperfect-information games. It converges to an equilibrium by iteratively traversing the game tree. In order to deal with extremely large games, abstraction is typically applied before running CFR. The abstracted game is solved with tabular CFR, and its solution is mapped back to the full game. This process can be problematic because aspects of abstraction are often manual and domain specific, abstraction algorithms may miss important strategic nuances of the game, and there is a chicken-and-egg problem because determining a good abstraction requires knowledge of the equilibrium of the game. This paper introduces Deep Counterfactual Regret Minimization, a form of CFR that obviates the need for abstraction by instead using deep neural networks to approximate the behavior of CFR in the full game. We show that Deep CFR is principled and achieves strong performance in large poker games. This is the first non-tabular variant of CFR to be successful in large games.