Machine learning
Reinforcement learning
Agents learning from reward. Markov decision processes, dynamic programming, Q-learning and policy gradients.
About 40 hours of study.
Where to learn it
Read
Reinforcement Learning: An Introduction
Richard Sutton, Andrew Barto
Part I (chapters 2–8), then chapter 13 on policy gradients
The standard textbook, free from the authors.
Watch
Introduction to Reinforcement Learning (DeepMind x UCL)
David Silver, Google DeepMind
The whole course
Practise
Understanding Deep Learning notebooks
Simon J. D. Prince
Notebooks for chapter 19
Sixty-eight Python notebooks with the key code left for you to fill in.
Wrong chapter, dead link, or know something better? Tell us on GitHub.
Which roles ask for it
- Deep learning engineer (vision / language) good to have
- Quantitative researcher / analyst good to have
Before and after
Reinforcement learning
Opens up
Nothing else on the map yet.