Revisiting CFR+ and Alternating Updates

10/26/2018

∙

The CFR+ algorithm for solving imperfect information games is a variant of the popular CFR algorithm, with faster empirical performance on a range of problems. It was introduced with a theoretical upper bound on solution error, but subsequent work showed an error in one step of the proof. We provide updated proofs to recover the original bound.

READ FULL TEXT

Revisiting CFR+ and Alternating Updates

Lazy-CFR: a fast regret minimization algorithm for extensive games with imperfect information

Rényi-Ulam Games and Online Computation with Imperfect Advice

A True O(n logn) Algorithm for the All-k-Nearest-Neighbors Problem

A Hennessy-Milner Theorem for ATL with Imperfect Information

Communication-Efficient Distributed SGD with Error-Feedback, Revisited

Trading Transforms of Non-weighted Simple Games and Integer Weights of Weighted Simple Games

Tighter Bound Estimation of Sensitivity Analysis for Incremental and Decremental Data Modification

Revisiting CFR+ and Alternating Updates

Related Research

Lazy-CFR: a fast regret minimization algorithm for extensive games with imperfect information

Rényi-Ulam Games and Online Computation with Imperfect Advice

A True O(n logn) Algorithm for the All-k-Nearest-Neighbors Problem

A Hennessy-Milner Theorem for ATL with Imperfect Information

Communication-Efficient Distributed SGD with Error-Feedback, Revisited

Trading Transforms of Non-weighted Simple Games and Integer Weights of Weighted Simple Games

Tighter Bound Estimation of Sensitivity Analysis for Incremental and Decremental Data Modification