A reverse-KL perturbed FTRL estimator has unbiased, conditionally-zero-variance perturbation updates that improve last-iterate convergence in sampled extensive-form games.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GT 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Power of Perturbation under Sampling in Solving Extensive-Form Games
A reverse-KL perturbed FTRL estimator has unbiased, conditionally-zero-variance perturbation updates that improve last-iterate convergence in sampled extensive-form games.