Back to the program

Topic-Contributed Paper Session

Statistical Theory and Methods for Reinforcement Learning

Sun, Aug 6, 4:00 PM

Yuxin ChenChair

IMS

5 Presentations

Near-optimal inference in adaptive regimes.

4:05 PM - 4:25 PM

Koulik Khamaru (Department of Statistics, Rutgers University)

STEEL: Singularity-aware Reinforcement Learning

4:25 PM - 4:45 PM

Settling the Sample Complexity of Model-Based Offline Reinforcement Learning

4:45 PM - 5:05 PM

Yuting Wei (University of Pennsylvania)

Online Statistical Inference for Matrix Contextual Bandit

5:05 PM - 5:25 PM

Will Wei Sun (Purdue University)

Off-Policy Evaluation in Partially Observed Markov Decision Processes under Sequential Ignorability

5:25 PM - 5:45 PM

↑