Pith. sign in

REVIEW 1 cited by

Regularized Behavior Value Estimation

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 2103.09575 v1 pith:67LYCPFQ submitted 2021-03-17 cs.LG

classification cs.LG
keywords r-bvetrainingbehaviorduringlearningpolicyvaluechallenge
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Offline reinforcement learning restricts the learning process to rely only on logged-data without access to an environment. While this enables real-world applications, it also poses unique challenges. One important challenge is dealing with errors caused by the overestimation of values for state-action pairs not well-covered by the training data. Due to bootstrapping, these errors get amplified during training and can lead to divergence, thereby crippling learning. To overcome this challenge, we introduce Regularized Behavior Value Estimation (R-BVE). Unlike most approaches, which use policy improvement during training, R-BVE estimates the value of the behavior policy during training and only performs policy improvement at deployment time. Further, R-BVE uses a ranking regularisation term that favours actions in the dataset that lead to successful outcomes. We provide ample empirical evidence of R-BVE's effectiveness, including state-of-the-art performance on the RL Unplugged ATARI dataset. We also test R-BVE on new datasets, from bsuite and a challenging DeepMind Lab task, and show that R-BVE outperforms other state-of-the-art discrete control offline RL methods.

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 vs. Online Learning in Model-based RL: Lessons for Data Collection Strategies

    cs.LG 2025-09 conditional novelty 5.0 of 10

    Offline-trained world-model agents in DreamerV3 underperform online agents due to out-of-distribution states at test time; adding about 10% self-generated data or exploratory data largely recovers performance.

Pith tools