mlsanityData to 7 Oct 2026

Leveraging Approximate Model-based Shielding for Probabilistic Safety Guarantees in Continuous Environments

arXiv 2402.00816 ↗First posted 1 Feb 2024cs.LG cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Shielding is a popular technique for achieving safe reinforcement learning (RL). However, classical shielding approaches come with quite restrictive assumptions making them difficult to deploy in complex environments, particularly those with continuous state or action spaces. In this paper we extend the more versatile approximate model-based shielding (AMBS) framework to the continuous setting. In particular we use Safety Gym as our test-bed, allowing for a more direct comparison of AMBS with popular constrained RL algorithms. We also provide strong probabilistic safety guarantees for the continuous setting. In addition, we propose two novel penalty techniques that directly modify the policy gradient, which empirically provide more stable convergence in our experiments.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 24 of its 49 references that are in the map. First posted on arXiv 1 Feb 2024.

Builds on

24 of 49 references.

References · 24

  1. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,302
  2. 2023Mastering Diverse Domains through World Modelscited by 1,193
  3. 2017Constrained Policy Optimizationcited by 928
  4. 2018Recurrent World Models Facilitate Policy Evolutioncited by 761
  5. 2018Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewcited by 466
  6. 2017Safe Reinforcement Learning via Shieldingcited by 427
  7. 2016Benchmarking Deep Reinforcement Learning for Continuous Controlcited by 361
  8. 2017Safe Model-based Reinforcement Learning with Stability Guaranteescited by 347
  9. 2020Information Theoretic Regret Bounds for Online Nonlinear Controlcited by 87
  10. 2020A Game Theoretic Framework for Model Based Reinforcement Learningcited by 72
  11. 2022Safe Reinforcement Learning by Imagining the Near Futurecited by 67
  12. 2022Constrained Policy Optimization via Bayesian World Modelscited by 55
12 more
  1. 2022Safe Reinforcement Learning via Shielding under Partial Observabilitycited by 49
  2. 2015Shield Synthesis: Runtime Enforcement for Reactive Systemscited by 41
  3. 2023Safe Reinforcement Learning via Probabilistic Logic Shieldscited by 41
  4. 2023Learning in POMDPs is Sample-Efficient with Hindsight Observabilitycited by 17
  5. 2021Shielding Atari Games with Bounded Presciencecited by 14
  6. 2021PC-MLP: Model-based Reinforcement Learning with Policy Cover Guided Explorationcited by 14
  7. 2023Model-based Dynamic Shielding for Safe and Efficient Multi-Agent Reinforcement Learningcited by 11
  8. 2023Provably Safe Reinforcement Learning with Step-wise Violation Constraintscited by 10
  9. 2021Do Androids Dream of Electric Fences? Safety-Aware Reinforcement Learning with Latent Shieldingcited by 9
  10. 2022Learning a Shield from Catastrophic Action Effects: Never Repeat the Same Mistakecited by 5
  11. 2022Shielding in Resource-Constrained Goal POMDPscited by 4
  12. 2023Approximate Shielding of Atari Agents for Safe Explorationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 2.

5 papers.

  1. 2025Conformal Safety Shielding for Imperfect-Perception Agentscited by 3
  2. 2026Robust Shielding for Safe Reinforcement Learningcited by 2
  3. 2026Safe Reinforcement Learning via Recovery-based Shielding with Gaussian Process Dynamics Modelscited by 1
  4. 2025Constraint-Aware Reinforcement Learning via Adaptive Action Scalingcited by 0
  5. 2025ProSh: Probabilistic Shielding for Model-free Reinforcement Learningcited by 0