In this course, we were asked to complete three implementation projects that address concepts of Reinforcement Learning. My language of choice was Python.

Here is a notebook that implements the Multi-Armed Bandit problem on some toy datasets.

Here is a notebook that compares the performance of three Reinforcement Learning Algorithms on the Grid World goal searching problem.

Here is a report that addresses the use of a neural network approach to attempt to solve the grid world goal searching problem. You will notice that I had some struggles with this project, which I now realize may be solved by one-hot encoding the state space.