Subgraph Bellman operators give a policy evaluation estimator whose finite-sample error nearly matches TD's optimal asymptotic variance while retaining MC's occupancy-adaptive sample complexity.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
To bootstrap or to rollout? An optimal and adaptive interpolation
Subgraph Bellman operators give a policy evaluation estimator whose finite-sample error nearly matches TD's optimal asymptotic variance while retaining MC's occupancy-adaptive sample complexity.