We are organizing an informal and intense reading group on reinforcement learning:

Zoom:  https://uwmadison.zoom.us/j/92888607205?pwd=dGVNUWJ2bGF3R0p1aFFoeG9oK3FSUT09 

Date: Every week day starting 6/6/22

Time: 9-10am, could spill to 11am depending on presentation

We will attempt to cover three topics, in this order:

1. The RL theory book https://rltheorybook.github.io
2. This spring 2022 seminar https://sites.google.com/view/rltheoryseminars/past-seminars/spring-2022
3. Adversarial attacks and defenses in RL/MARL

Sign up: (for the RL theory book version Jan 31 2022, let’s each lead one chapter.  No need to specify the dates beforehand, to give ourselves some flexibility.  Some chapters can be presented out of order to fit the presenter schedule.)

Chapter 1 Markov Decision Processes: Jerry 6/6, 6/7?

Chapter 2 Sample Complexity with a Generative Model: Yiding (6/8, 6/9)

Chapter 3 Linear Bellman Completeness: Andrew (6/10)

Chapter 4 Fitted Dynamic Programming Methods: Jeremy (6/16)

Chapter 5 Statistical Limits of Generalization: Yixuan Zhang (6/17, 6/20, 6/21)

Chapter 6 Multi-Armed and Linear Bandits: Ransheng Guan (6/22, 6/23)

Chapter 7 Strategic Exploration in Tabular MDPs: Stephen Jasina (6/24, 6/27)

Chapter 8 Linearly Parameterized MDPs: Shubham Bharti (6/28, 6/29)

Chapter 9 Generalization with Bounded Bilinear Rank: Yiding (6/30, 7/1)

Chapter 10 Deterministic MDPs with Linearly Parameterized Q*

Chapter 11 Policy Gradient Methods and Non-Convex Optimization: Matthew Zurek (date tbd)

Chapter 12 Optimality

Chapter 13 Function Approximation and the NPG

Chapter 14 CPI, TRPO, and More

Chapter 15 Imitation Learning

Chapter 16 Linear Quadratic Regulators

Chapter 17 Partially Observable Markov Decision Processes