HBPI-UCRL gives the first explicit PAC sample-complexity bound for parallel hierarchical RL, with a factor-S improvement over flat BPI-UCRL in sparse-reward goal-directed tasks under a restrictive assumption.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sample Efficient Hierarchical Reinforcement Learning via Best Policy Identification
HBPI-UCRL gives the first explicit PAC sample-complexity bound for parallel hierarchical RL, with a factor-S improvement over flat BPI-UCRL in sparse-reward goal-directed tasks under a restrictive assumption.