Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty

arXiv:2604.25508v1 Announce Type: new Abstract: Safety remains an open problem in reinforcement learning (RL), especially during training. While safety filters are promising to address safe exploration, they are generally poorly suited for high-dimensional systems with unknown dynamics. We propose Dyna-style Safety Augmented Reinforcement Learning (Dyna-SAuR), a novel algorithm that learns both a scalable safety filter and a control policy using a learned uncertainty-aware dynamics model, while requiring minimal domain knowledge. The filter avoids failures and high uncertainty regions. Thus, better models expand the set of safe and certain states, reducing filter conservatism. We present the effectiveness of Dyna-SAuR on goal-reaching CartPole as well as MuJoCo Walker, reducing failures compared to state-of-the-art methods by 2 orders of magnitude.

Safe-Support Q-Learning: Learning without Unsafe Exploration
Safe Reinforcement Learning using Action Projection: Safeguard the Policy or the Environment?
Safety Training Modulates Harmful Misalignment Under On-Policy RL, But Direction Depends on Environment Design
Policy Gradient Primal-Dual Method for Safe Reinforcement Learning from Human Feedback
RL-STPA: Adapting System-Theoretic Hazard Analysis for Safety-Critical Reinforcement Learning

Source: arXiv cs.LG | 2026-04-29

Related