–
Coffee & Check In (Gather.town)
–
Opening Remarks
–
Online Learning with A Lot of Batch Data
Shie Mannor (Technion & NVIDIA)
–
Learning Multi-Agent Collaborations With Decomposition
Yuandong Tian (Facebook AI Research)
–
Beyond Worst-Case: Instance-Dependent Optimality in Reinforcement Learning
Martin Wainwright (UC Berkeley)
–
Break
–
An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay
Doina Precup (McGill University)
–
Confident Off-policy Evaluation and Selection through Self-Normalized Importance Weighting
Ilja Kuzborskij (DeepMind)
–
Coffee & Check In (Gather.town)
–
Robust Learning of Stochastic Dynamical Systems
Munther Dahleh (MIT)
–
Monte Carlo Sampling Approach to Solving Stochastic Multistage Programs
Alex Shapiro (Georgia Tech)
–
Reinforcement Learning using Generative Models for Continuous State and Action Space Systems
Rahul Jain (USC)
–
Break
–
Batch Value-function Approximation with Only Realizability
Nan Jiang (University of Illinois at Urbana-Champaign)
–
Uniform Offline Policy Evaluation and Offline Learning in Tabular RL
Yu-Xiang Wang (UC Santa Barbara)
–
Coffee & Check In (Gather.town)
–
Zap Q-learning with Nonlinear Function Approximation
Sean Meyn (University of Florida)
–
Q-learning with Uniformly Bounded Variance
Adithya Devraj (Stanford)
–
The Mean-Squared Error of Double Q-Learning
R. Srikant (University of Illinois at Urbana-Champaign)
–
Break
–
Panel Discussion
–
Coffee & Check In (Gather.town)
–
Batch Policy Learning in Average Reward Markov Decision Processes
Peng Liao (Harvard)
–
Multiagent Reinforcement Learning: Rollout and Policy Iteration
Dimitri Bertsekas (ASU & MIT)
–
Statistical Efficiency in Offline Reinforcement Learning
Nathan Kallus (Cornell)
–
Break
–
Nearly Minimax Optimal Reward-Free Reinforcement Learning
Simon Du (University of Washington)
–
Sample Efficient Reinforcement Learning via Low-Rank Matrix Estimation
Devavrat Shah (MIT)
–
Coffee & Check In (Gather.town)
–
Stable Reinforcement Learning with Unbounded State Space
Qiaomin Xie (Cornell)
–
Policy Evaluation under Interference
Stefan Wager (Stanford)
–
Convergence and Sample Complexity of Gradient Methods for the Model-Free Linear Quadratic Regulator Problem
Mihailo Jovanovic (USC)
–
Break
–
Offline Reinforcement Learning and Model-Based Optimization
Sergey Levine (UC Berkeley)
–
Panel Discussion