# A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning

arXiv: https://arxiv.org/abs/2312.07685 · HTML: https://mlsanity.com/paper/2312.07685/

## 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.
- Builds on 26 of its 34 references that are in the map.
- First posted on arXiv 12 Dec 2023.

## Placed in

- Offline RL and imitation learning from data / [Offline RL algorithms](https://mlsanity.com/offlinerl/offline-rl-algorithms/)

## Builds on (most cited first, 26 of 26)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [OpenAI Gym](https://mlsanity.com/paper/1606.01540/) (2016)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Offline Reinforcement Learning with Implicit Q-Learning](https://mlsanity.com/paper/2110.06169/) (2021)
- [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018)
- [A Minimalist Approach to Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06860/) (2021)
- [Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction](https://mlsanity.com/paper/1906.00949/) (2019)
- [AWAC: Accelerating Online Reinforcement Learning with Offline Datasets](https://mlsanity.com/paper/2006.09359/) (2020)
- [Behavior Regularized Offline Reinforcement Learning](https://mlsanity.com/paper/1911.11361/) (2019)
- [MOPO: Model-based Offline Policy Optimization](https://mlsanity.com/paper/2005.13239/) (2020)
- [Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning](https://mlsanity.com/paper/1910.11956/) (2019)
- [Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble](https://mlsanity.com/paper/2110.01548/) (2021)
- [Offline-to-Online Reinforcement Learning via Balanced Replay and Pessimistic Q-Ensemble](https://mlsanity.com/paper/2107.00591/) (2021)
- [Pessimistic Bootstrapping for Uncertainty-Driven Offline Reinforcement Learning](https://mlsanity.com/paper/2202.11566/) (2022)
- [Policy Expansion for Bridging Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2302.00935/) (2023)
- [Combining Self-Supervised Learning and Imitation for Vision-Based Rope Manipulation](https://mlsanity.com/paper/1703.02018/) (2017)
- [Dexterous Manipulation with Deep Reinforcement Learning: Efficient, General, and Low-Cost](https://mlsanity.com/paper/1810.06045/) (2018)
- [Reincarnating Reinforcement Learning: Reusing Prior Computation to Accelerate Progress](https://mlsanity.com/paper/2206.01626/) (2022)
- [Efficient Reinforcement Learning for Autonomous Driving with Parameterized Skills and Priors](https://mlsanity.com/paper/2305.04412/) (2023)
- [Adaptive Behavior Cloning Regularization for Stable Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2210.13846/) (2022)
- [ACE: Cooperative Multi-agent Q-learning with Bidirectional Action-Dependency](https://mlsanity.com/paper/2211.16068/) (2022)

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

- [Efficient Online Reinforcement Learning Fine-Tuning Need Not Retain Offline Data](https://mlsanity.com/paper/2412.07762/) (2024)
- [Penalizing Infeasible Actions and Reward Scaling in Reinforcement Learning with Offline Data](https://mlsanity.com/paper/2507.08761/) (2025)
- [Adaptive Action Chunking via Multi-Chunk Q Value Estimation](https://mlsanity.com/paper/2605.10044/) (2026)
- [Offline-to-Online Multi-Agent Reinforcement Learning with Offline Value Function Memory and Sequential Exploration](https://mlsanity.com/paper/2410.19450/) (2024)
- [Towards Fast Safe Online Reinforcement Learning via Policy Finetuning](https://mlsanity.com/paper/2412.04426/) (2024)
- [VLM Q-Learning: Aligning Vision-Language Models for Interactive Decision-Making](https://mlsanity.com/paper/2505.03181/) (2025)
- [Entropy Regularized Task Representation Learning for Offline Meta-Reinforcement Learning](https://mlsanity.com/paper/2412.14834/) (2024)
- [Enhancing Offline Model-Based RL via Active Model Selection: A Bayesian Optimization Perspective](https://mlsanity.com/paper/2502.11480/) (2025)
- [Behavior-Adaptive Q-Learning: A Unifying Framework for Offline-to-Online RL](https://mlsanity.com/paper/2511.03695/) (2025)
- [Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2602.18117/) (2026)
- [From Static Constraints to Dynamic Adaptation: Sample-Level Constraint Relaxation for Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2511.03828/) (2025)
- [SMAC: Score-Matched Actor-Critics for Robust Offline-to-Online Transfer](https://mlsanity.com/paper/2602.17632/) (2026)
- [Efficient and Uncertainty-Aware Diffusion Framework for Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2605.30776/) (2026)
- [Diffusion Policy with Behavioral Advantage Correction for Offline Reinforcement Learning](https://mlsanity.com/paper/2608.02332/) (2026)
- [Policy Plasticity Matters in Offline-to-Online Reinforcement Learning: Refitting Offline Policies for Online Adaptation](https://mlsanity.com/paper/2609.33127/) (2026)

## Abstract (arXiv)

Offline-to-online Reinforcement Learning (O2O RL) aims to improve the performance of offline pretrained policy using only a few online samples. Built on offline RL algorithms, most O2O methods focus on the balance between RL objective and pessimism, or the utilization of offline and online samples. In this paper, from a novel perspective, we systematically study the challenges that remain in O2O RL and identify that the reason behind the slow improvement of the performance and the instability of online finetuning lies in the inaccurate Q-value estimation inherited from offline pretraining. Specifically, we demonstrate that the estimation bias and the inaccurate rank of Q-value cause a misleading signal for the policy update, making the standard offline RL algorithms, such as CQL and TD3-BC, ineffective in the online finetuning. Based on this observation, we address the problem of Q-value estimation by two techniques: (1) perturbed value update and (2) increased frequency of Q-value updates. The first technique smooths out biased Q-value estimation with sharp peaks, preventing early-stage policy exploitation of sub-optimal actions. The second one alleviates the estimation bias inherited from offline pretraining by accelerating learning. Extensive experiments on the MuJoco and Adroit environments demonstrate that the proposed method, named SO2, significantly alleviates Q-value estimation issues, and consistently improves the performance against the state-of-the-art methods by up to 83.1%.