Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces
Amirhossein Roknilamouki, Arnob Ghosh, Ming Shi, Fatemeh Nourzad, Eylem Ekici, Ness B. Shroff
2025年份
2顶会引用
摘要
In Reinforcement Learning (RL), tasks with instantaneous hard constraints present significant challenges, particularly when the decision space is non-convex or non-star-convex. This issue is especially relevant in domains like autonomous vehicles and robotics, where constraints such as collision avoidance often take a non-convex form, and the state-space may be large. In this paper, we establish a regret bound of Õ 1 +
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式，引用它的顶会 Paper 也一样。你提问，回答直接引用原文。
引用它的顶会 Paper2
- Provably Efficient RL under Episode-Wise Safety in Constrained MDPs with Linear Function ApproximationToshinori Kitamura, Arnob Ghosh, Tadashi Kozuno, Wataru Kumagai 等NeurIPS 2025 · 被引用 5 次
- Primal-Dual Policy Optimization for Linear CMDPs with Adversarial LossesKihyun Yu, Seoungbin Bae, Dabeen LeeICLR 2026 · 被引用 2 次
相关 Paper
- A Near-Optimal Algorithm for Safe Reinforcement Learning Under Instantaneous Hard ConstraintsMing Shi, Yingbin Liang, Ness B. ShroffICML 2023 · 被引用 18 次
- Constrained Markov Decision Processes via Backward Value FunctionsHarsh Satija, Philip Amortila, Joelle PineauICML 2020 · 被引用 58 次
- SafeMPO: Constrained Reinforcement Learning with Probabilistic Incremental ImprovementAlexander Mattick, Dominik Seuß, Christopher MutschlerICLR 2026
- First Order Constrained Optimization in Policy SpaceYiming Zhang, Quan Vuong, Keith W. RossNeurIPS 2020 · 被引用 238 次
- Provably Efficient Model-Free Constrained RL with Linear Function ApproximationArnob Ghosh, Xingyu Zhou, Ness B. ShroffNeurIPS 2022 · 被引用 41 次
