A Connection between One-Step RL and Critic Regularization in Reinforcement Learning

Benjamin Eysenbach, Matthieu Geist, Sergey Levine, Ruslan Salakhutdinov

Research output: Contribution to journalConference articlepeer-review

Abstract

As with any machine learning problem with limited data, effective offline RL algorithms require careful regularization to avoid overfitting. One class of methods, known as one-step RL, perform just one step of policy improvement. These methods, which include advantage-weighted regression and conditional behavioral cloning, are thus simple and stable, but can have limited asymptotic performance. A second class of methods, known as critic regularization, perform many steps of policy improvement with a regularized objective. These methods typically require more compute but have appealing lower-bound guarantees. In this paper, we draw a connection between these methods: applying a multi-step critic regularization method with a regularization coefficient of 1 yields the same policy as one-step RL. While our theoretical results require assumptions (e.g., deterministic dynamics), our experiments nevertheless show that our analysis makes accurate, testable predictions about practical offline RL methods (CQL and one-step RL) with commonly-used hyperparameters.

Original languageEnglish (US)
Pages (from-to)9485-9507
Number of pages23
JournalProceedings of Machine Learning Research
Volume202
StatePublished - 2023
Externally publishedYes
Event40th International Conference on Machine Learning, ICML 2023 - Honolulu, United States
Duration: Jul 23 2023Jul 29 2023

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'A Connection between One-Step RL and Critic Regularization in Reinforcement Learning'. Together they form a unique fingerprint.

Cite this