Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915963061927936 |
|---|---|
| author | Eisele, Artur Frauenknecht, Bernd Solowjow, Friedrich Trimpe, Sebastian |
| author_facet | Eisele, Artur Frauenknecht, Bernd Solowjow, Friedrich Trimpe, Sebastian |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_25508 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty Eisele, Artur Frauenknecht, Bernd Solowjow, Friedrich Trimpe, Sebastian Machine Learning 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. |
| title | Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.25508 |