# Safe Reinforcement Learning via Shielding

arXiv: https://arxiv.org/abs/1708.08611 · HTML: https://mlsanity.com/paper/1708.08611/

## Key facts

- Cited by 449 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Aug 2017.

## Built on by (most cited first, 30 of 449)

- [Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning](https://mlsanity.com/paper/2108.06266/) (2021)
- [Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark](https://mlsanity.com/paper/2304.03279/) (2023)
- [A Review of Safe Reinforcement Learning: Methods, Theory and Applications](https://mlsanity.com/paper/2205.10330/) (2022)
- [A Survey of Safety and Trustworthiness of Large Language Models through the Lens of Verification and Validation](https://mlsanity.com/paper/2305.11391/) (2023)
- [A Survey on Interpretable Reinforcement Learning](https://mlsanity.com/paper/2112.13112/) (2021)
- [Constrained Variational Policy Optimization for Safe Reinforcement Learning](https://mlsanity.com/paper/2201.11927/) (2022)
- [Safe Multi-Agent Reinforcement Learning via Shielding](https://mlsanity.com/paper/2101.11196/) (2021)
- [Safe Reinforcement Learning via Shielding under Partial Observability](https://mlsanity.com/paper/2204.00755/) (2022)
- [Efficient Learning of Safe Driving Policy via Human-AI Copilot Optimization](https://mlsanity.com/paper/2202.10341/) (2022)
- [Safe Reinforcement Learning Using Advantage-Based Intervention](https://mlsanity.com/paper/2106.09110/) (2021)
- [A Survey of Constraint Formulations in Safe Reinforcement Learning](https://mlsanity.com/paper/2402.02025/) (2024)
- [Provable Repair of Deep Neural Networks](https://mlsanity.com/paper/2104.04413/) (2021)
- [Safe Reinforcement Learning via Probabilistic Logic Shields](https://mlsanity.com/paper/2303.03226/) (2023)
- [Learning Barrier Certificates: Towards Safe Reinforcement Learning with Zero Training-time Violations](https://mlsanity.com/paper/2108.01846/) (2021)
- [On the Robustness of Safe Reinforcement Learning under Observational Perturbations](https://mlsanity.com/paper/2205.14691/) (2022)
- [A Survey of Sim-to-Real Methods in RL: Progress, Prospects and Challenges with Foundation Models](https://mlsanity.com/paper/2502.13187/) (2025)
- [Provably Safe Reinforcement Learning: Conceptual Analysis, Survey, and Benchmarking](https://mlsanity.com/paper/2205.06750/) (2022)
- [Verifying Learning-Based Robotic Navigation Systems](https://mlsanity.com/paper/2205.13536/) (2022)
- [ISAACS: Iterative Soft Adversarial Actor-Critic for Safety](https://mlsanity.com/paper/2212.03228/) (2022)
- [Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithms](https://mlsanity.com/paper/2310.03225/) (2023)
- [ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Prediction](https://mlsanity.com/paper/2508.00500/) (2025)
- [Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees](https://mlsanity.com/paper/2201.08355/) (2022)
- [Provably Safe Reinforcement Learning via Action Projection using Reachability Analysis and Polynomial Zonotopes](https://mlsanity.com/paper/2210.10691/) (2022)
- [Learning Control Policies for Stochastic Systems with Reach-avoid Guarantees](https://mlsanity.com/paper/2210.05308/) (2022)
- [CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learning](https://mlsanity.com/paper/2403.18765/) (2024)
- [Online Shielding for Reinforcement Learning](https://mlsanity.com/paper/2212.01861/) (2022)
- [When to Ask for Help: Proactive Interventions in Autonomous Reinforcement Learning](https://mlsanity.com/paper/2210.10765/) (2022)
- [SoK: Explainable Machine Learning for Computer Security Applications](https://mlsanity.com/paper/2208.10605/) (2022)
- [Neural Network Repair with Reachability Analysis](https://mlsanity.com/paper/2108.04214/) (2021)
- [COOL-MC: A Comprehensive Tool for Reinforcement Learning and Model Checking](https://mlsanity.com/paper/2209.07133/) (2022)

## Abstract (arXiv)

Reinforcement learning algorithms discover policies that maximize reward, but do not necessarily guarantee safety during learning or execution phases. We introduce a new approach to learn optimal policies while enforcing properties expressed in temporal logic. To this end, given the temporal logic specification that is to be obeyed by the learning system, we propose to synthesize a reactive system called a shield. The shield is introduced in the traditional learning process in two alternative ways, depending on the location at which the shield is implemented. In the first one, the shield acts each time the learning agent is about to make a decision and provides a list of safe actions. In the second way, the shield is introduced after the learning agent. The shield monitors the actions from the learner and corrects them only if the chosen action causes a violation of the specification. We discuss which requirements a shield must meet to preserve the convergence guarantees of the learner. Finally, we demonstrate the versatility of our approach on several challenging reinforcement learning scenarios.