# Finetuning from Offline Reinforcement Learning: Challenges, Trade-offs and Practical Solutions

arXiv: https://arxiv.org/abs/2303.17396 · HTML: https://mlsanity.com/paper/2303.17396/

## Key facts

- Cited by 26 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 20 of its 28 references that are in the map.
- First posted on arXiv 30 Mar 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, 20 of 20)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [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)
- [Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning](https://mlsanity.com/paper/1910.00177/) (2019)
- [AWAC: Accelerating Online Reinforcement Learning with Offline Datasets](https://mlsanity.com/paper/2006.09359/) (2020)
- [Maximum a Posteriori Policy Optimisation](https://mlsanity.com/paper/1806.06920/) (2018)
- [Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewards](https://mlsanity.com/paper/1707.08817/) (2017)
- [Critic Regularized Regression](https://mlsanity.com/paper/2006.15134/) (2020)
- [Offline-to-Online Reinforcement Learning via Balanced Replay and Pessimistic Q-Ensemble](https://mlsanity.com/paper/2107.00591/) (2021)
- [Online Decision Transformer](https://mlsanity.com/paper/2202.05607/) (2022)
- [Acme: A Research Framework for Distributed Reinforcement Learning](https://mlsanity.com/paper/2006.00979/) (2020)
- [Offline Reinforcement Learning: Fundamental Barriers for Value Function Approximation](https://mlsanity.com/paper/2111.10919/) (2021)
- [AW-Opt: Learning Robotic Skills with Imitation and Reinforcement at Scale](https://mlsanity.com/paper/2111.05424/) (2021)
- [Offline Reinforcement Learning Hands-On](https://mlsanity.com/paper/2011.14379/) (2020)

## Built on by (most cited first, 26 of 26)

- [Reinforcement Learning with Action Chunking](https://mlsanity.com/paper/2507.07969/) (2025)
- [Blending Supervised and Reinforcement Fine-Tuning with Prefix Sampling](https://mlsanity.com/paper/2507.01679/) (2025)
- [PROTO: Iterative Policy Regularized Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2305.15669/) (2023)
- [Decision ConvFormer: Local Filtering in MetaFormer is Sufficient for Decision Making](https://mlsanity.com/paper/2310.03022/) (2023)
- [Accelerating Exploration with Unlabeled Prior Data](https://mlsanity.com/paper/2311.05067/) (2023)
- [Fast Adaptation with Behavioral Foundation Models](https://mlsanity.com/paper/2504.07896/) (2025)
- [Optimistic Critic Reconstruction and Constrained Fine-Tuning for General Offline-to-Online RL](https://mlsanity.com/paper/2412.18855/) (2024)
- [A Simple Unified Uncertainty-Guided Framework for Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2306.07541/) (2023)
- [Small Dataset, Big Gains: Enhancing Reinforcement Learning by Offline Pre-Training with Model Based Augmentation](https://mlsanity.com/paper/2312.09844/) (2023)
- [Active Advantage-Aligned Online Reinforcement Learning with Offline Data](https://mlsanity.com/paper/2502.07937/) (2025)
- [Improving Robotic Generalist Policies via Flow Reversal Steering](https://mlsanity.com/paper/2606.13675/) (2026)
- [Planning to Go Out-of-Distribution in Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2310.05723/) (2023)
- [Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignment](https://mlsanity.com/paper/2406.11039/) (2024)
- [Offline Behavior Distillation](https://mlsanity.com/paper/2410.22728/) (2024)
- [Generative Actor Critic](https://mlsanity.com/paper/2512.21527/) (2025)
- [Safe and Interpretable Estimation of Optimal Treatment Regimes](https://mlsanity.com/paper/2310.15333/) (2023)
- [Improving Offline-to-Online Reinforcement Learning with Q Conditioned State Entropy Exploration](https://mlsanity.com/paper/2310.19805/) (2023)
- [Snapshot Reinforcement Learning: Leveraging Prior Trajectories for Efficiency](https://mlsanity.com/paper/2403.00673/) (2024)
- [Efficient Offline Reinforcement Learning: First Imitate, then Improve](https://mlsanity.com/paper/2406.13376/) (2024)
- [Fine-Tuning without Performance Degradation](https://mlsanity.com/paper/2505.00913/) (2025)
- [COOPO: Cyclic Offline-Online Policy Optimization Algorithm](https://mlsanity.com/paper/2605.18675/) (2026)
- [Beyond One-Size-Fits-All: Diagnosis-Driven Online Reinforcement Learning with Offline Priors](https://mlsanity.com/paper/2606.25527/) (2026)
- [Robo-ValueRL: Reliable Value Estimation for Offline-to-Online Reinforcement Learning](https://mlsanity.com/paper/2607.09866/) (2026)
- [Evaluating Fuzz Testing for Reinforcement Learning Agents](https://mlsanity.com/paper/2607.24577/) (2026)
- [Schrödinger--Föllmer Actor--Critic: Diffusion Policy Improvement with Finite-Sample Analysis](https://mlsanity.com/paper/2609.33239/) (2026)
- [From Static Policies to Adaptive Priors in Offline Reinforcement Learning](https://mlsanity.com/paper/2609.35880/) (2026)

## Abstract (arXiv)

Offline reinforcement learning (RL) allows for the training of competent agents from offline datasets without any interaction with the environment. Online finetuning of such offline models can further improve performance. But how should we ideally finetune agents obtained from offline RL training? While offline RL algorithms can in principle be used for finetuning, in practice, their online performance improves slowly. In contrast, we show that it is possible to use standard online off-policy algorithms for faster improvement. However, we find this approach may suffer from policy collapse, where the policy undergoes severe performance deterioration during initial online learning. We investigate the issue of policy collapse and how it relates to data diversity, algorithm choices and online replay distribution. Based on these insights, we propose a conservative policy optimization procedure that can achieve stable and sample-efficient online learning from offline pretraining.