# Harnessing Density Ratios for Online Reinforcement Learning

arXiv: https://arxiv.org/abs/2401.09681 · HTML: https://mlsanity.com/paper/2401.09681/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data.
- First posted on arXiv 18 Jan 2024.

## Placed in

- Offline RL and imitation learning from data / [Offline RL algorithms](https://mlsanity.com/offlinerl/offline-rl-algorithms/)
- Deep reinforcement learning methods / [Deep RL algorithms and representations](https://mlsanity.com/deeprl/deep-rl-algorithms-and-representations/)

## Built on by (most cited first, 15 of 15)

- [The Importance of Online Data: Understanding Preference Fine-tuning via Coverage](https://mlsanity.com/paper/2406.01462/) (2024)
- [Do We Need to Verify Step by Step? Rethinking Process Supervision from a Theoretical Perspective](https://mlsanity.com/paper/2502.10581/) (2025)
- [Offline Reinforcement Learning: Role of State Aggregation and Trajectory Data](https://mlsanity.com/paper/2403.17091/) (2024)
- [Outcome-Based Online Reinforcement Learning: Algorithms and Fundamental Limits](https://mlsanity.com/paper/2505.20268/) (2025)
- [RL in Latent MDPs is Tractable: Online Guarantees via Off-Policy Evaluation](https://mlsanity.com/paper/2406.01389/) (2024)
- [Hybrid Reinforcement Learning Breaks Sample Size Barriers in Linear MDPs](https://mlsanity.com/paper/2408.04526/) (2024)
- [Provably and Practically Efficient Adversarial Imitation Learning with General Function Approximation](https://mlsanity.com/paper/2411.00610/) (2024)
- [Hybrid Preference Optimization for Alignment: Provably Faster Convergence Rates by Combining Offline Preferences with Online Exploration](https://mlsanity.com/paper/2412.10616/) (2024)
- [A Natural Extension To Online Algorithms For Hybrid RL With Limited Coverage](https://mlsanity.com/paper/2403.09701/) (2024)
- [Efficient and Sharp Off-Policy Evaluation in Robust Markov Decision Processes](https://mlsanity.com/paper/2404.00099/) (2024)
- [Greedy Algorithm for Structured Bandits: A Sharp Characterization of Asymptotic Success / Failure](https://mlsanity.com/paper/2503.04010/) (2025)
- [Fitted Q-Evaluation without Bellman Completeness via Occupancy Weighting](https://mlsanity.com/paper/2512.23805/) (2025)
- [A Tutorial: An Intuitive Explanation of Offline Reinforcement Learning Theory](https://mlsanity.com/paper/2508.07746/) (2025)
- [A Unifying View of Coverage in Linear Off-Policy Evaluation](https://mlsanity.com/paper/2601.19030/) (2026)
- [On the Complexity of Offline Reinforcement Learning with Q^star-Approximation and Partial Coverage](https://mlsanity.com/paper/2602.12107/) (2026)

## Abstract (arXiv)

The theories of offline and online reinforcement learning, despite having evolved in parallel, have begun to show signs of the possibility for a unification, with algorithms and analysis techniques for one setting often having natural counterparts in the other. However, the notion of density ratio modeling, an emerging paradigm in offline RL, has been largely absent from online RL, perhaps for good reason: the very existence and boundedness of density ratios relies on access to an exploratory dataset with good coverage, but the core challenge in online RL is to collect such a dataset without having one to start. In this work we show -- perhaps surprisingly -- that density ratio-based algorithms have online counterparts. Assuming only the existence of an exploratory distribution with good coverage, a structural condition known as coverability (Xie et al., 2023), we give a new algorithm (GLOW) that uses density ratio realizability and value function realizability to perform sample-efficient online exploration. GLOW addresses unbounded density ratios via careful use of truncation, and combines this with optimism to guide exploration. GLOW is computationally inefficient; we complement it with a more efficient counterpart, HyGLOW, for the Hybrid RL setting (Song et al., 2022) wherein online RL is augmented with additional offline data. HyGLOW is derived as a special case of a more general meta-algorithm that provides a provable black-box reduction from hybrid RL to offline RL, which may be of independent interest.