Materials for the reinforcement learning course at Data Science Retreat.
This course is aimed at students with a grasp of supervised learning - no prior understanding of reinforcement learning required.
The course materials are:
- slides for two days of lectures - pdf - GitPitch
- detailed notes to support lectures and for future study
This project is built and maintained by Adam Green - firstname.lastname@example.org.
- background statistical concepts
- Markov Decision processes
- value function methods (DQN and it's extensions)
- policy gradient methods
- practical advice for experiments
- current state of the art
Goals for the course
Introduction to concepts, ideas and terminology. Familiarity with important literature. Understanding of the state of the field today. Practical strategies to run reinforcement learning experiments.
Where to go next
- The Holy Book of reinforcement learning - Sutton & Barto - Reinforcement Learning: An Introduction - 2nd Edition
- UCL Lectures - David Silver (Head of Reinforcement Learning at DeepMind) - slides - lecture videos
- Li (2017) Deep Reinforcement Learning: An Overview
Further resources (video lectures, blog posts etc) are listed in dsr-rl/resources.md.