EE-POLITEX interleaves an exploration policy with on-policy rollouts and uses least-squares Monte Carlo value estimation, claiming a sublinear regret bound under weaker exploration assumptions, but the proof has a norm-mismatch gap.
Approximate policy iteration: A survey and some new methods
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Exploration-Enhanced POLITEX
EE-POLITEX interleaves an exploration policy with on-policy rollouts and uses least-squares Monte Carlo value estimation, claiming a sublinear regret bound under weaker exploration assumptions, but the proof has a norm-mismatch gap.