Combining AIVAT variance reduction with anytime-valid confidence sequences lets poker agent evaluations stop at a median 74x fewer hands at plus or minus 1 BB, with exact finite-sample certification currently limited to games with an independent payoff bound.
Strategy evaluation in extensive games with importance sampling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.GT 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games
Combining AIVAT variance reduction with anytime-valid confidence sequences lets poker agent evaluations stop at a median 74x fewer hands at plus or minus 1 BB, with exact finite-sample certification currently limited to games with an independent payoff bound.