Statistical Foundations of Reinforcement Learning Through a Non-asymptotic Lens
Speaker: Yuejie Chi – Pittsburgh, PA, United StatesTopic(s): Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing , Computational Theory, Algorithms and Mathematics
Abstract
Reinforcement learning (RL) is garnering significant interest in recent years due to its success in a wide variety of modern applications. However, theoretical understandings on the non-asymptotic sample and computational efficiencies of RL algorithms still remain elusive, and are in imminent need to cope with the ever-increasing problem dimensions. In this talk, we discuss our recent progresses on understanding and improving the efficacy and resiliency of RL algorithms in Markov decision processes (MDPs), and demonstrate how the model-based approach using a plug-in model estimate achieves minimax-optimality across a variety of settings.
About this Lecture
Number of Slides: 40Duration: 45 minutes
Languages Available: English
Last Updated:
Request this Lecture
To request this particular lecture, please complete this online form.
Request a Tour
To request a tour with this speaker, please complete this online form.
All requests will be sent to ACM headquarters for review.