arXiv：cs.AI· Luca Vignola, Bruce D. Lee, Manish Prajapat, Manuel Wendl, Melanie Zeilinger, Andreas Krause, Yarden As·· 13 小时前
SBSRL：基于采样的安全强化学习
Sampling-Based Safe Reinforcement Learning
AI 导读
研究者提出 SBSRL（Sampling-Based Safe Reinforcement Learning），一种基于模型的强化学习算法，通过在有限动态样本集合上联合施加约束，在整个学习过程中维持安全性。
正文
Abstract:Safe exploration remains a fundamental challenge in reinforcement learning (RL), limiting the deployment of RL agents in the real world. We propose Sampling-Based Safe Reinforcement Learning (SBSRL), a model-based RL algorithm that maintains safety throughout the learning process by enforcing constraints jointly across a finite set of dynamics samples. This formulation approximates an intractable worst-case optimization over uncertain dynamics and enables practical safety guarantees in continuous domains. We further introduce an exploration strategy based on constraining epistemic uncertainty, eliminating the need for explicit exploration bonuses. Under regularity conditions, we derive high-probability guarantees of safety throughout learning and a finite-time sample complexity bound for recovering a near-optimal policy. Empirically, SBSRL achieves safe and efficient exploration both in simulation and in real robotic hardware, and readily extends to practical deep-ensemble implementations that scale to high-dimensional continuous control problems.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO) |
| Cite as: | arXiv:2605.19469 [cs.LG] |
| (or arXiv:2605.19469v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.19469 arXiv-issued DOI via DataCite |
Submission history
From: Luca Vignola [view email]
[v1]
Tue, 19 May 2026 07:21:53 UTC (6,055 KB)
[v2]
Fri, 9 Oct 2026 15:01:32 UTC (6,907 KB)
来源：arXiv：cs.AI · arxiv.org