Showing stat.MLShow all
2 papers · 1 filter
stat.ML2026
Model-based Bootstrap for Offline Policy Evaluation in Tabular Reinforcement Learning
Weiwei Wang, Yuqiang Li, Xianyi Wu +1
Offline policy evaluation (OPE) is crucial in high-stakes reinforcement learning applications, where new policies must be assessed reliably before deployment. In such settings, poi…
stat.ML2026
Pessimistic Risk-Aware Policy Learning in Contextual Bandits
Yilong Wan, Yuqiang Li, Xianyi Wu
We study risk-aware offline policy learning, aiming to learn a decision rule from logged data that is optimal under general risk criteria. This problem is crucial in high-stakes do…