Asymptotically Unbiased Off-Policy Policy Evaluation when Reusing Old Data in Nonstationary EnvironmentsJan 2, 2023·Vincent Liu,Yash Chandak,Philip Thomas,Martha White· 0 min read Cite URLTypeConference paperPublicationInternational Conference on Artificial Intelligence and StatisticsLast updated on Jan 2, 2023 AuthorsVincent LiuPostdoctoral Research and Teaching Fellow ← Exploiting action impact regularity and exogenous state variables for offline reinforcement learning Jan 3, 2023Measuring and Mitigating Interference in Reinforcement Learning Jan 1, 2023 →
