# Safe offline RL

Area: [Offline RL and imitation learning from data](https://mlsanity.com/offlinerl/index.md). HTML: https://mlsanity.com/offlinerl/safe-offline-rl/

## Key facts

- Safe offline RL: 115 papers on arXiv since 2019 (data through 14 Oct 2026).
- 21 of its 115 mapped papers (18%) cite “Constraints Penalized Q-learning for Safe Offline Reinforcement Learning” (2021, arXiv 2107.09003).
- 16 of its 115 mapped papers (14%) cite “Datasets and Benchmarks for Offline Safe Reinforcement Learning” (2023, arXiv 2306.09303).
- 1 new papers 17 Sep–14 Oct.

## Most cited here

- [Constraints Penalized Q-learning for Safe Offline Reinforcement Learning](https://mlsanity.com/paper/2107.09003/) (2021), cited by 21 papers here
- [Datasets and Benchmarks for Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2306.09303/) (2023), cited by 16 papers here
- [Safe Offline Reinforcement Learning with Feasibility-Guided Diffusion Model](https://mlsanity.com/paper/2401.10700/) (2024), cited by 15 papers here
- [Constrained Decision Transformer for Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2302.07351/) (2023), cited by 14 papers here
- [Safe Offline Reinforcement Learning with Real-Time Budget Constraints](https://mlsanity.com/paper/2306.00603/) (2023), cited by 9 papers here
- [Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare Settings](https://mlsanity.com/paper/2107.11003/) (2021), cited by 5 papers here
- [SaFormer: A Conditional Sequence Modeling Approach to Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2301.12203/) (2023), cited by 5 papers here
- [Surgical Robot Transformer (SRT): Imitation Learning for Surgical Tasks](https://mlsanity.com/paper/2407.12998/) (2024), cited by 5 papers here
- [OASIS: Conditional Distribution Shaping for Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2407.14653/) (2024), cited by 5 papers here
- [Offline Safe Reinforcement Learning Using Trajectory Classification](https://mlsanity.com/paper/2412.15429/) (2024), cited by 5 papers here

## New (17 Sep–14 Oct)

- [Finite-Sample Probabilistic Safety Certification for AI-Based Grid-Edge Coordination](https://mlsanity.com/paper/2609.28182/) (2026-09-23)