Skip to content

Solving the Monty Hall problem with reinforcement learning

Notifications You must be signed in to change notification settings

irenenikk/lets-make-a-deal

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Using reinforcement learning to solve the Monty Hall problem

Here is a friendly Wikipedia link for those who are unfamiliar with the formulation.

This is a simple program to train an agent to switch doors in the Monty Hall problem using Q-learning.

Yes, this is a total overkill, absolutely no reinforcement learning is required to train an agent to play successfully. As there is only one state, the problem is effectively a repeated game, and the agent could just easily calculate how often it was a better idea to switch, because there is no need to take future states into account. I basically built this just to tinker around with Q learning a little in practice, and decided to start off easy.

About

Solving the Monty Hall problem with reinforcement learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages