Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Instance-Dependent Regret for CMDPs with Step-Wise Constraints
Qian Zuo, Francesco Emanuele Stradi, Leyang Xue +1
We study online learning in episodic tabular constrained Markov decision processes with step-wise safety constraints. In such a setting, the constraints induce a safe subgraph that…
cs.LG2026
Near-Constant Strong Violation and Last-Iterate Convergence for Online CMDPs via Decaying Safety Margins
Qian Zuo, Zhiyong Wang, Fengxiang He
We study safe online reinforcement learning in Constrained Markov Decision Processes (CMDPs) under strong regret and violation metrics, which forbid error cancellation over time. E…
cs.LG2025
Ensuring Safety in an Uncertain Environment: Constrained MDPs via Stochastic Thresholds
Qian Zuo, Fengxiang He
This paper studies constrained Markov decision processes (CMDPs) with constraints against stochastic thresholds, aiming at safety of reinforcement learning in unknown and uncertain…