RL READING & DISCUSSION SESSIONS

In this RL series we will cover “Reinforcement learning: An introduction” by Richard Sutton and Andrew Barto.

📌 Session leads:
Pierre Wüthrich, Emil Vatai, Anugraha Sinha (APAC)
Raymond Chua, Mrityunjay Bhardwaj (Americas, EMEA)

📌 Session structure
● 15 minutes of recap (if second part of a chapter)
● 60 minutes silent reading
● 45 min discussion

📌 Join Meetup
To get the most out of the sessions make sure to get the book, prepare for the session chapters and read a bit ahead if possible. That will serve as a good basis for an interactive and productive discussion.
Join us on Slack for discussions #rl_book

📌 Book Info
Book : Reinforcement learning, An introduction
Author : Richard Sutton and Andrew Barto
Publication : MIT Press

A physical copy of the book can be purchased e.g. on Amazon
Link to book. Alternatively, the book is available as a pdf from the authors website: http://incompleteideas.net/index.html


📌 Part 1: Tabular Solution Methods
Session #1 Introduction
Session #2: Multi-armed Bandits
Session #3: Finite Markov Decision Processes
Session #4: Dynamic Programming (1)
Session #5: Dynamic Programming (2)
Session #6: Monte Carlo Methods (1)
Session #7: Monte Carlo Methods (2)
Session #8: Temporal-difference Learning (1)
Session #9: Temporal-difference Learning (2)
Session #10: n-step Bootstrapping (1)
Session #11: n-step Bootstrapping (2)
Session #12: Planning and Learning with Tabular methods (1)
Session #13: Planning and Learning with Tabular methods (1)

📌 Part 2: TBA

RL tic-tac-toe player built by MLT member Mrityunjay Bhardwaj.

📌 Join RL sessions on Meetup



📌MLT PATRON
Become a MLT Patron and help us to keep MLT meetups like this inclusive and for free. https://www.patreon.com/MLTOKYO

📌 SUBSCRIBE
Subscribe to our monthly newsletter: https://machinelearningtokyo.com/

📌 FIND MLT RESOURCES
Github: https://github.com/Machine-Learning-Tokyo
Youtube: https://www.youtube.com/MLTOKYO
Slack: https://bit.ly/2Yb0uXI

📌 CODE OF CONDUCT
MLT promotes an inclusive environment that values integrity, openness and respect. https://github.com/Machine-Learning-Tokyo/MLT_starterkit

Leave a Reply