These are notes that I use to prepare for class. These are not intended to be official lecture notes. In particular, I don't proof-read them carefully. However, if you do find typos, please let me know and I will correct them. You are expected to come to class and take notes. Please do not ask me to put up the notes before lectures, I may not always be able to do this.
4. Value Iteration/Policy Iteration/LP Solution
5. Q-Learning
6. Function Approximation Using Neural Nets
9. Stochastic Shortest Path Problem
10. Linear Function Approximation
11. The ODE Method
13. Q-Learning for Average-cost-per-stage MDPs