It’s been a few years since I was first exposed to reinforcement learning. What got me into it was seeing this video that had trained a neural network to play Mario. As someone who grew up playing Mario, seeing deep learning being applied to something I knew so well seemed to provided the perfect introduction to the topic. Sadly though, the project was written using Torch, and I was still a naive young programmer. I didn’t get too far along before the frustrations with learning lua lead me to give up, and just focus on other projects instead. Recently though, I began working with PyTorch that inspired me to try it again; this time in python.
Getting Mario up and running was far easier this time around. I was able to utilize OpenAI’s gym package, which made the process fairly simple, but it wasn’t exactly straight forward. It required some work quick work arounds and tweaks to the gym environment. After I finally got it up and running, my coworkers were fascinated, and I began trouble shooting their installations. Seeing their interest, I thought I could provide instructions for how to get it set up; so that anyone can start letting their computer learn how to play Super Mario Bros.