Pith. sign in

REVIEW 1 cited by

Near-Optimal Offline Reinforcement Learning via Double Variance Reduction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2102.01748 v1 pith:4SMMISY5 submitted 2021-02-02 cs.LG cs.AIstat.ML

classification cs.LGcs.AIstat.ML
keywords offlineopdvrepsilonmdpsoptimalpolicyreductionvariance
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

We consider the problem of offline reinforcement learning (RL) -- a well-motivated setting of RL that aims at policy optimization using only historical data. Despite its wide applicability, theoretical understandings of offline RL, such as its optimal sample complexity, remain largely open even in basic settings such as \emph{tabular} Markov Decision Processes (MDPs). In this paper, we propose Off-Policy Double Variance Reduction (OPDVR), a new variance reduction based algorithm for offline RL. Our main result shows that OPDVR provably identifies an $\epsilon$-optimal policy with $\widetilde{O}(H^2/d_m\epsilon^2)$ episodes of offline data in the finite-horizon stationary transition setting, where $H$ is the horizon length and $d_m$ is the minimal marginal state-action distribution induced by the behavior policy. This improves over the best known upper bound by a factor of $H$. Moreover, we establish an information-theoretic lower bound of $\Omega(H^2/d_m\epsilon^2)$ which certifies that OPDVR is optimal up to logarithmic factors. Lastly, we show that OPDVR also achieves rate-optimal sample complexity under alternative settings such as the finite-horizon MDPs with non-stationary transitions and the infinite horizon MDPs with discounted rewards.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Offline Learning for Combinatorial Multi-armed Bandits

    cs.LG 2025-01 conditional novelty 7.0 of 10

    A pessimistic lower-confidence-bound algorithm achieves suboptimality bounds for offline combinatorial multi-armed bandits with probabilistically triggered arms, under coverage conditions requiring observation of each...

Pith tools