paper-with-me

홈 › Papers

Robust Probabilistic Shielding for Safe Offline Reinforcement Learning

2026-05-11 · Maris F. L. Galesloot, Thomas Rhemrev, Nils Jansen arxiv

In offline reinforcement learning (RL), we learn policies from fixed datasets without environment interaction. The major challenges are to provide guarantees on the (1) performance and (2) safety of the resulting policy. A technique called safe policy improvement (SPI) provides a performance guarantee: with high probability, the new policy outperforms a given baseline policy, which is assumed to be safe. Orthogonally, in the context of safe RL, a shield provides a safety guarantee by restricting the action space to those actions that are provably safe with respect to a given safety-relevant model. We integrate these paradigms by extending shielding to offline RL, relying solely on the available dataset and knowledge of safe and unsafe states. Then, we shield the policy improvement steps, guaranteeing, with high probability, a safe policy. Experimental results demonstrate that shielded SPI outperforms its unshielded counterpart, improving both average and worst-case performance, particularly in low-data regimes.

📄 PDF Abstract BibTeX arXiv:2605.10293

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningOffline RL

Similar Papers 제목 키워드 기반

Shields to Guarantee Probabilistic Safety in MDPs

2026-05-11 · Linus Heck, Filip Macák, Roman Andriushchenko, Milan Češka 외 arxiv

Shielding is a prominent model-based technique to ensure safety of autonomous agents. Classical shielding aims to ensure that nothing bad ever happens and comes with strong guarantees about safety and maximal permissiven…

Safe Reinforcement Learning via Probabilistic Logic Shields

2023-03-06 · Wen-Chi Yang, Giuseppe Marra, Gavin Rens, Luc De Raedt

Safe Reinforcement learning (Safe RL) aims at learning optimal policies while staying safe. A popular solution to Safe RL is shielding, which uses a logical safety specification to prevent an RL agent from taking unsafe …

reinforcement-learningReinforcement LearningReinforcement Learning (RL)Safe Reinforcement Learning

Verification-Guided Shielding for Deep Reinforcement Learning

2024-06-10 · Davide Corsi, Guy Amir, Andoni Rodriguez, Cesar Sanchez 외

In recent years, Deep Reinforcement Learning (DRL) has emerged as an effective approach to solving real-world tasks. However, despite their successes, DRL-based policies suffer from poor reliability, which limits their d…

Deep Reinforcement Learningreinforcement-learningReinforcement Learning

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

2024-02-01 · Alexander W. Goodall, Francesco Belardinelli

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, pa…

Reinforcement Learning (RL)Safe Reinforcement Learning

Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learning

2026-08-20 · Astrid Horn Brorholt, Maris F. L. Galesloot, Nils Jansen, Kim Guldstrand Larsen 외 arxiv

Probabilistic shielding is a technique for safe reinforcement learning (RL). Typically, a static observer -- called the shield -- constrains the learning agent's actions to those for which acting safely remains feasible.…

Reinforcement Learning