This is here temporarily for an application, I'm still working on the program and a larger project I intend for it to be a part of.
This is a tkinter program designed to help understand the bandits problem intuitively by allowing you to manually play "bandits" and run your own algorithms to see them solve the problem step by step. Additionally, to better understand more complex algorithms like UCB, you can run algorithms against each other to see differences in their deceision making at each time step as well as their performance.
There are two modes: a simplified 3 arm dice mode to more intuively understand the math that's going on, and a more complex Gaussian distributtion 10-armed mode. Still lots of work to do on this, but I'm very happy with it so far.