mlsanityData to 7 Oct 2026

Safe Reinforcement Learning via Shielding under Partial Observability

arXiv 2204.00755 ↗First posted 2 Apr 2022cs.AIcited by 49

Not on the map. Why?

Abstract (arXiv)

Safe exploration is a common problem in reinforcement learning (RL) that aims to prevent agents from making disastrous decisions while exploring their environment. A family of approaches to this problem assume domain knowledge in the form of a (partial) model of this environment to decide upon the safety of an action. A so-called shield forces the RL agent to select only safe actions. However, for adoption in various applications, one must look beyond enforcing safety and also ensure the applicability of RL with good performance. We extend the applicability of shields via tight integration with state-of-the-art deep RL, and provide an extensive, empirical study in challenging, sparse-reward environments under partial observability. We show that a carefully integrated shield ensures safety and can improve the convergence rate and final performance of RL agents. We furthermore show that a shield can be used to bootstrap state-of-the-art RL agents: they remain safe after initial learning in a shielded setting, allowing us to disable a potentially too conservative shield eventually.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). Builds on 18 of its 44 references that are in the map. First posted on arXiv 2 Apr 2022.

Builds on

18 of 44 references.

References · 18

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
  3. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,704
  4. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
  5. 2015Deep Recurrent Q-Learning for Partially Observable MDPscited by 457
  6. 2017Safe Reinforcement Learning via Shieldingcited by 427
  7. 2017Deep Reinforcement Learning framework for Autonomous Drivingcited by 224
  8. 2019Soft Actor-Critic for Discrete Action Settingscited by 161
  9. 2020The Probabilistic Model Checker Stormcited by 64
  10. 2020Cautious Reinforcement Learning with Logical Constraintscited by 44
  11. 2015Safety-Constrained Reinforcement Learning for MDPscited by 31
  12. 2019Reinforcement Learning with Probabilistic Guarantees for Autonomous Drivingcited by 27
6 more
  1. 2020Robust Finite-State Controllers for Uncertain POMDPscited by 23
  2. 2015Permissive Controller Synthesis for Probabilistic Systemscited by 16
  3. 2020Enforcing Almost-Sure Reachability in POMDPscited by 13
  4. 2015A Symbolic SAT-based Algorithm for Almost-sure Reachability with Small Strategies in POMDPscited by 6
  5. 2022Learning a Shield from Catastrophic Action Effects: Never Repeat the Same Mistakecited by 5
  6. 2021Reward prediction for representation learning and reward shapingcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 12; 2024: 8; 2025: 10; 2026: 18.

49 papers.

Sort:

By area:

  1. 2023Safe Reinforcement Learning via Probabilistic Logic Shieldscited by 41
  2. 2023Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithmscited by 27
  3. 2024Can a Bayesian Oracle Prevent Harm from an Agent?cited by 12
  4. 2022Safe Reinforcement Learning From Pixels Using a Stochastic Latent Representationcited by 11
  5. 2024Dynamic Model Predictive Shielding for Provably Safe Reinforcement Learningcited by 11
  6. 2023Decision-Making Under Uncertainty: Beyond Probabilitiescited by 8
  7. 2023Shielded Reinforcement Learning for Hybrid Systemscited by 8
  8. 2023Safe POMDP Online Planning via Shieldingcited by 8
  9. 2023Risk-Aware Reinforcement Learning through Optimal Transport Theorycited by 7
  10. 2024Pessimistic Iterative Planning with RNNs for Robust POMDPscited by 7
  11. 2024Safe Reinforcement Learning in Black-Box Environments via Adaptive Shieldingcited by 6
  12. 2024Enhancing RL Safety with Counterfactual LLM Reasoningcited by 6
37 more
  1. 2023Reinforcement Learning with Knowledge Representation and Reasoning: A Brief Surveycited by 5
  2. 2023Reinforcement Learning by Guided Safe Explorationcited by 5
  3. 2024Leveraging Approximate Model-based Shielding for Probabilistic Safety Guarantees in Continuous Environmentscited by 5
  4. 2024Probabilistic Model Checking of Stochastic Reinforcement Learning Policiescited by 5
  5. 2025Rectified Robust Policy Optimization for Model-Uncertain Constrained Reinforcement Learning without Strong Dualitycited by 4
  6. 2023Safety Shielding under Delayed Observationcited by 3
  7. 2025Conformal Safety Shielding for Imperfect-Perception Agentscited by 3
  8. 2025Optimistic Exploration for Risk-Averse Constrained Reinforcement Learningcited by 3
  9. 2026Formally Verifying and Explaining Sepsis Treatment Policies with COOL-MCcited by 3
  10. 2023Approximate Shielding of Atari Agents for Safe Explorationcited by 2
  11. 2025Adaptive Shielding for Safe Reinforcement Learning under Hidden-Parameter Dynamics Shiftscited by 2
  12. 2025Provably Optimal Reinforcement Learning under Safety Filteringcited by 2
  13. 2026Shields to Guarantee Probabilistic Safety in MDPscited by 2
  14. 2026Robust Shielding for Safe Reinforcement Learningcited by 2
  15. 2024Think Smart, Act SMARL! Analyzing Probabilistic Logic Shields for Multi-Agent Reinforcement Learningcited by 1
  16. 2025Predictive Safety Shield for Dyna-Q Reinforcement Learningcited by 1
  17. 2026Safe Continual Reinforcement Learning Methods for Nonstationary Environments. Towards a Survey of the State of the Artcited by 1
  18. 2026Safety Generalization Under Distribution Shift in Safe Reinforcement Learning: A Diabetes Testbedcited by 1
  19. 2026Easy-to-Use Shielding for Reinforcement Learningcited by 1
  20. 2023Experiments with Detecting and Mitigating AI Deceptioncited by 0
  21. 2023Searching for Optimal Runtime Assurance via Reachability and Reinforcement Learningcited by 0
  22. 2025Bi-directional Recurrence Improves Transformer in Partially Observable Markov Decision Processescited by 0
  23. 2025Efficient Dynamic Shielding for Parametric Safety Specificationscited by 0
  24. 2025Learning to Navigate Under Imperfect Perception: Conformalised Segmentation for Safe Reinforcement Learningcited by 0
  25. 2025Shielded Controller Units for RL with Operational Constraints Applied to Remote Microgridscited by 0
  26. 2026COOL-MC: Verifying and Explaining RL Policies for Platelet Inventory Managementcited by 0
  27. 2026Interval POMDP Shielding for Imperfect-Perception Agentscited by 0
  28. 2026Action-Conditioned Risk Gating for Safety-Critical Control under Partial Observabilitycited by 0
  29. 2026Safe Continual Reinforcement Learning under Nonstationarity via Adaptive Safety Constraintscited by 0
  30. 2026Energy Shields for Fairnesscited by 0
  31. 2026Latent Q-Barrier Shielding for Safe In-Context Reinforcement Learningcited by 0
  32. 2026Explainably Safe Reinforcement Learningcited by 0
  33. 2026Shielded Analysis: Certification and Characterization of Defensibility in Systems under Adversarial Interactioncited by 0
  34. 2026Shielding for Higher-Order Safetycited by 0
  35. 2026Legislating World-Model-Based Planning with Legal Reasoningcited by 0
  36. 2026Managing Action Preconditions in Neuro-Symbolic RL: Three Placement Strategies for Embodied Agentscited by 0
  37. 2026AI Harness: Certification under Proposal-Conditioned Information for Foundation-Model Agentscited by 0