A black-box theorem shows that if un-averaged SA iterates concentrate, then Polyak-averaged iterates concentrate at the optimal 1/k rate, with applications to reinforcement learning algorithms.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A General-Purpose Theorem for High-Probability Bounds of Stochastic Approximation with Polyak Averaging
A black-box theorem shows that if un-averaged SA iterates concentrate, then Polyak-averaged iterates concentrate at the optimal 1/k rate, with applications to reinforcement learning algorithms.