My final course (I took 2 the last semester but the other class wasn’t as cool even though it had the same instructor) in the GT OMS CS program was 6460 Education Technology. The purpose of the course was to contribute to education technology. You could do research, create content, or write code with the hopes that you continue after the semester is over.
Going into the course I know I wanted to create something that would teach Reinforcement Learning. I got the idea from the head TA in the RL course I took in the summer of 2017. He created a GitHub that would compliment the course’s book. He has since gone on to write a book based on his work.
Since his work was so good I wanted to go a slightly different direction so I decided to just focus on Q-Learning and its various iterations.
Here is the result: https://github.com/ehennis/ReinforcementLearning
I created 6 Jupyter Notebooks that will take the user through each algorithm and hopefully introduce them to RL and get them hooked like I was when I first saw it.
Feel free to fork/copy/etc. the repo and see how it goes. If I get motivated enough I might try and turn this into a video series with someone.
Other notes about the course: The first section of the course was doing some research on general education technologies as well as different types of learning types. I fought this at first but after the first few weeks I just accepted that I needed to do the leg work and found out some pretty cool stuff.