# Safe Reinforcement Learning via Recovery-based Shielding with Gaussian Process Dynamics Models

arXiv: https://arxiv.org/abs/2602.12444 · HTML: https://mlsanity.com/paper/2602.12444/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Safe and constrained RL, one of 15 directions in Deep reinforcement learning methods.
- Builds on 23 of its 57 references that are in the map.
- First posted on arXiv 12 Feb 2026.

## Placed in

- Deep reinforcement learning methods / [Safe and constrained RL](https://mlsanity.com/deeprl/safe-and-constrained-rl/)

## Builds on (most cited first, 23 of 23)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [Prioritized Experience Replay](https://mlsanity.com/paper/1511.05952/) (2015)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [Safe Reinforcement Learning via Shielding](https://mlsanity.com/paper/1708.08611/) (2017)
- [On Kernelized Multi-armed Bandits](https://mlsanity.com/paper/1704.00445/) (2017)
- [End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks](https://mlsanity.com/paper/1903.08792/) (2019)
- [Learning-based Model Predictive Control for Safe Exploration](https://mlsanity.com/paper/1803.08287/) (2018)
- [Safe Reinforcement Learning in Constrained Markov Decision Processes](https://mlsanity.com/paper/2008.06626/) (2020)
- [OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research](https://mlsanity.com/paper/2305.09304/) (2023)
- [Uniform Error Bounds for Gaussian Process Regression with Application to Safe Control](https://mlsanity.com/paper/1906.01376/) (2019)
- [Constrained Policy Optimization via Bayesian World Models](https://mlsanity.com/paper/2201.09802/) (2022)
- [Robust Model Predictive Shielding for Safe Reinforcement Learning with Stochastic Dynamics](https://mlsanity.com/paper/1910.10885/) (2019)
- [MCMC for Variationally Sparse Gaussian Processes](https://mlsanity.com/paper/1506.04000/) (2015)
- [Shield Synthesis: Runtime Enforcement for Reactive Systems](https://mlsanity.com/paper/1501.02573/) (2015)
- [Linear model predictive safety certification for learning-based control](https://mlsanity.com/paper/1803.08552/) (2018)
- [Provably Safe Reinforcement Learning: Conceptual Analysis, Survey, and Benchmarking](https://mlsanity.com/paper/2205.06750/) (2022)
- [Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithms](https://mlsanity.com/paper/2310.03225/) (2023)
- [Policy Optimization with Linear Temporal Logic Constraints](https://mlsanity.com/paper/2206.09546/) (2022)
- [Probabilistic Shielding for Safe Reinforcement Learning](https://mlsanity.com/paper/2503.07671/) (2025)
- [Online Shielding for Stochastic Systems](https://mlsanity.com/paper/2012.09539/) (2020)
- [Do Androids Dream of Electric Fences? Safety-Aware Reinforcement Learning with Latent Shielding](https://mlsanity.com/paper/2112.11490/) (2021)
- [Leveraging Approximate Model-based Shielding for Probabilistic Safety Guarantees in Continuous Environments](https://mlsanity.com/paper/2402.00816/) (2024)

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

- [Uncertainty-Aware Predictive Safety Filters for Probabilistic Neural Network Dynamics](https://mlsanity.com/paper/2604.26836/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) is a powerful framework for optimal decision-making and control but often lacks provable guarantees for safety-critical applications. In this paper, we introduce a novel recovery-based shielding framework that enables safe RL with a provable safety lower bound for unknown and non-linear continuous dynamical systems. The proposed approach integrates a backup policy (shield) with the RL agent, leveraging Gaussian process (GP) based uncertainty quantification to predict potential violations of safety constraints, dynamically recovering to safe trajectories only when necessary. Experience gathered by the 'shielded' agent is used to construct the GP models, with policy optimization via internal model-based sampling - enabling unrestricted exploration and sample efficient learning, without compromising safety. Empirically our approach demonstrates strong performance and strict safety-compliance on a suite of continuous control environments.