Summary

Coded the basic game play for my Tetris multiplayer AI game.

Started learning directly about reinforcement learning.

Project - Tetris AI

I wrote the basic code for my Tetris project using Python and Pygame.

If you don’t know about Pygame is a set of python libraries which have ton of useful methods for making a simple game. It is very fun to play around with and there are many tutorials online, you could make simple game very quickly if you are interested.

My project isn’t about making a Tetris game it wasn’t made from scratch, I’m not going to pretend I wrote it all, but the code I used I heavily modified.

I followed a guide by freecodecamp and from this I was able to make a simple single player Tetris game.

I then implemented code so that two games of Tetris would run at the same time and that those games could send lines at each other. The implementation is messy at the moment but I can and will refine it.

Tetris Game


At the moment there is no functioning AI this is just a display of my progress so far.

If you want to check it out it is on my github.

Reinforcement Learning

I’ve moved away from the previous course I was doing Machine Learning by Andrew Ng as I’m running low on time for my project. My plan is to learn just about the topic I need at the moment (reinforcement learning), and come back to cover Machine learning deeper afterwards.

I don’t like jumping around from topic to topic or course to course, but I think it is necessary here.

The course I’ve switch to watching on youtube is Reinforcement Learning Course by David Silver who is one of the leading researchers in deep reinforcement learning. I’ve found his course so far very easy to listen to and well explained. I haven’t gotten to far into it yet but plan to complete it as the next step towards implementing my Tetris AI.

Road Ahead -

Over the next week I plan to:

  • Complete the Tetris game play to a standard I’m happy with
  • Write a reinforcement learning programme for a simple issue

Thanks for reading and happy coding!