WRSWorld Research StateContribute
WORLD RESEARCH STATE

CFR+ and Cepheus: scalable near-equilibrium poker solving

Contribution grade

Imperfect-information multiagent learning · Human-designed research method or resource

Reference score175,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Introduced clipped cumulative regret-matching updates and weighted averaging, then combined distributed compressed computation with CFR+ to essentially weakly solve heads-up limit Texas hold'em.

Scientific scope

Introduced clipped cumulative regret-matching updates and weighted averaging, then combined distributed compressed computation with CFR+ to essentially weakly solve heads-up limit Texas hold'em. The2014algorithm originates with Tammelin alone; subsequent theory/engineering and Cepheus reports have their own bylines. Same asymptotic regret order does not prove a universal order-of-magnitude acceleration. Essentially weakly solved means a quantified very small exploitability, not an exact equilibrium or no-limit/multiplayer poker solution.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The2014algorithm originates with Tammelin alone; subsequent theory/engineering and Cepheus reports have their own bylines. Same asymptotic regret order does not prove a universal order-of-magnitude acceleration. Essentially weakly solved means a quantified very small exploitability, not an exact equilibrium or no-limit/multiplayer poker solution.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction