Skip to content

AlexVanMechelen/Q-learning-tetris

Repository files navigation

Project Reinforcement Learning - Tetris

This project expands on this project by S. Melax. We documented their original code here.

How to run?

The Q learning implementation is made in the c++ programming language. To execute the program it needs to be compiled first. Various compilers exist, one example is g++. The program can be compiled with the following command:

g++ -o run.exe tetris-Qlearning.cpp

It can now be executed by calling ./run.exe

Results

Various experiments have been performed like hyperparameter tuning, playing with the cost function and exploration method. The results of these experiments are processed in a Matlab Live Script and further discussed in a paper.

About

Q-learning applied to a simplified Tetris game

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Languages