# Beyond Hard Constraints: Budget-Conditioned Reachability For Safe Offline Reinforcement Learning

arXiv: https://arxiv.org/abs/2603.22292 · HTML: https://mlsanity.com/paper/2603.22292/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Safe offline RL, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 17 of its 31 references that are in the map.
- First posted on arXiv 8 Mar 2026.

## Placed in

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

## Builds on (most cited first, 17 of 17)

- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [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)
- [An Introduction to Variational Autoencoders](https://mlsanity.com/paper/1906.02691/) (2019)
- [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)
- [DualDICE: Behavior-Agnostic Estimation of Discounted Stationary Distribution Corrections](https://mlsanity.com/paper/1906.04733/) (2019)
- [Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes](https://mlsanity.com/paper/2408.03539/) (2024)
- [Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research](https://mlsanity.com/paper/2310.08710/) (2023)
- [OptiDICE: Offline Policy Optimization via Stationary Distribution Correction Estimation](https://mlsanity.com/paper/2106.10783/) (2021)
- [Extreme Q-Learning: MaxEnt RL without Entropy](https://mlsanity.com/paper/2301.02328/) (2023)
- [Offline RL with No OOD Actions: In-Sample Learning via Implicit Value Regularization](https://mlsanity.com/paper/2303.15810/) (2023)
- [COptiDICE: Offline Constrained Reinforcement Learning via Stationary Distribution Correction Estimation](https://mlsanity.com/paper/2204.08957/) (2022)
- [Safe Offline Reinforcement Learning with Feasibility-Guided Diffusion Model](https://mlsanity.com/paper/2401.10700/) (2024)
- [Offline Multi-Agent Reinforcement Learning with Implicit Global-to-Local Value Regularization](https://mlsanity.com/paper/2307.11620/) (2023)
- [Escaping from Zero Gradient: Revisiting Action-Constrained Reinforcement Learning via Frank-Wolfe Policy Optimization](https://mlsanity.com/paper/2102.11055/) (2021)
- [Hamilton-Jacobi Reachability in Reinforcement Learning: A Survey](https://mlsanity.com/paper/2407.09645/) (2024)

## Built on by (most cited first, 1 of 1)

- [Decoupled Guidance Diffusion for Adaptive Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2605.02777/) (2026)

## Abstract (arXiv)

Sequential decision making using Markov Decision Process underpins many realworld applications. Both model-based and model free methods have achieved strong results in these settings. However, real-world tasks must balance reward maximization with safety constraints, often conflicting objectives, that can lead to unstable min/max, adversarial optimization. A promising alternative is safety reachability analysis, which precomputes a forward-invariant safe state, action set, ensuring that an agent starting inside this set remains safe indefinitely. Yet, most reachability based methods address only hard safety constraints, and little work extends reachability to cumulative cost constraints. To address this, first, we define a safetyconditioned reachability set that decouples reward maximization from cumulative safety cost constraints. Second, we show how this set enforces safety constraints without unstable min/max or Lagrangian optimization, yielding a novel offline safe RL algorithm that learns a safe policy from a fixed dataset without environment interaction. Finally, experiments on standard offline safe RL benchmarks, and a real world maritime navigation task demonstrate that our method matches or outperforms state of the art baselines while maintaining safety.