Learning to play Can't Stop with a Deep Q Network

Wed, 20 May 2020 00:00:00 +0000

What with the COVID19 related confinement, I have been playing a lot of board games on board game arena with friends, colleagues, you name it. When you first create an account, the site takes you through the tutorial of a simple probabilistic risk vs gain evaluation game called Can’t Stop. Having a little more time on my hands during weekends than usual, I decided to train a reinforcement learning agent to learn to play this game, for which the full code can be found here.

Machine Learning on Welcome

Learning to play Can't Stop with a Deep Q Network