Video URL

https://simons.berkeley.edu/talks/online-reinforcement-learning-help-confounded-offline-data

Online Reinforcement Learning With The Help Of Confounded Offline Data

(2022). Online Reinforcement Learning With The Help Of Confounded Offline Data. The Simons Institute for the Theory of Computing. https://simons.berkeley.edu/talks/online-reinforcement-learning-help-confounded-offline-data

Online Reinforcement Learning With The Help Of Confounded Offline Data. The Simons Institute for the Theory of Computing, Feb. 14, 2022, https://simons.berkeley.edu/talks/online-reinforcement-learning-help-confounded-offline-data 

          @misc{ scivideos_19668,
            doi = {},
            url = {https://simons.berkeley.edu/talks/online-reinforcement-learning-help-confounded-offline-data},
            author = {},
            keywords = {},
            language = {en},
            title = {Online Reinforcement Learning With The Help Of Confounded Offline Data},
            publisher = {The Simons Institute for the Theory of Computing},
            year = {2022},
            month = {feb},
            note = {19668 see, \url{https://scivideos.org/Simons-Institute/19668}}
          }
Uri Shalit (Technion - Israel Institute of Technology)
Source Repository Simons Institute
Subject

Abstract

I will present recent work exploring how and when can confounded offline data be used to improve online reinforcement learning. We will explore conditions of partial observability and distribution shifts between the offline and online environments, and present results for contextual bandits, imitation learning and reinforcement learning.