Reinforcement Learning - Paper Reading

Reading list for the Grad Topics on Reinforcement Learning (ECE 750T4/ECE 457C)

Introduction

This course will have two components, which shift focus over the term. The first component will be lectures and work problems about the fundamentals of Reinforcement Learning (RL).

The second component will be communal presentation and discussion of research papers on advanced topics in RL. The papers will be read by all, presented briefly by a student, and then discussed by everyone, led by the presenting student, for half of the class period.

This second component is structured around the common grad school practice of Reading Groups, see more information below or see the pages for some previous reading groups here.

General Reading Groups Tips

In a reading group everyone takes turns leading discussion of a paper each week. Leading discussion can be as simple as having your own annotated notes on Hypothes.is to share and start discussion as we go through it together. Or it could be more involved, including making slides to present your overview of the paper’s contributions, highlights and weak points.


(2024: What you need to do)

Note, this process may be updated in future years.

If you are a student in this course you need to do the following:

  • Open the RL Reading List page.
  • Create a Hypothes.is account and sign up for the course group on Hypothes.is so you and everyone in the class can see our shared annotations
  • Pick the papers you will be reading, presenting, and leading discussion of and sign up (sign-up process TBD)
    • PhD Students: choose two papers, one of them near the start to set a good example!
    • Master’s Students: choose at least one paper
  • Then read the paper in detail, use Hypothes.is to make annotations for yourself and to guide others. Use the Hypothes.is course group you were all invited to do to this.
  • Prepare to present the main points of the paper, and guide discussion through the parts that are surprising, challening, interesting, or that you don’t understand.
  • The class discussion of the paper should help everyone, including you and the prof come away with a better understanding and evaluation of this publication