Solving Large Imperfect Information Games Using CFR+

TL;DR

This paper introduces CFR+, an improved algorithm for solving large imperfect information games like poker, which significantly outperforms previous methods in speed and memory efficiency.

Contribution

CFR+ is a novel algorithm that enhances counterfactual regret minimization, achieving faster convergence and reduced memory usage compared to existing approaches.

Findings

CFR+ converges faster than traditional CFR methods.

CFR+ requires less memory for large game trees.

CFR+ outperforms previous algorithms by an order of magnitude in computation time.

Abstract

Counterfactual Regret Minimization and variants (e.g. Public Chance Sampling CFR and Pure CFR) have been known as the best approaches for creating approximate Nash equilibrium solutions for imperfect information games such as poker. This paper introduces CFR$^+$, a new algorithm that typically outperforms the previously known algorithms by an order of magnitude or more in terms of computation time while also potentially requiring less memory.