Exercise 9: Deep Reinforcement Learning

Paper Exercises

In this exercise we will derive the pseudoinverse of a matrix, complementary to the computer exercises you have started solving. We will explore an application of the Monte Carlo Search Tree to Tetris and take a closer look at approximate value iteration.

Update: paper exercises and solutions have now been added to the compendium Download compendium. 

Computer Exercises

The computer exercise is found in the notebook E10 on the jupyterhub server. It is strongly recommended that you complete the notebooks for exercises 6-10 before moving on to the laboratory assignment.