Work done as a part of E1 254: Game Theory course at IISc. The code tries to simulate an environment which mimics A/B testing. One needs to assign the unknown paramters of the environment and then interface it with an RL agent to get the solution.
The associated comments in the code along with the paper linked below is enough to understnad the working of the environment.
The model of the environment is roughly based on the paper: Experimental Design in Two-Sided Platforms: An Analysis of Bias