Week 14 Reinforcement learning

The exercise we have used in previous years has been in policy gradients. Since policy gradients is not curriculum this year, I suggest you have a look at the Q-learning tutorial.

Lecture:

Slides (pdf)

Exercise: Q-learning

https://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html

Exercise (Optional): Policy gradients

assignment/solution (zip)

You may need to install gym: pip install gym