Skip to content

ALjone/DeepQSnake

Repository files navigation

DeepQSnake

An implementation of many different methods for improving the performance of Q-learning tested on the game Snake.

Developed over a few years, achieving very good performance on 7x7 maps, where it averages around 45 out of 49 possible apples.

Note that in general, Q-learning is suboptimal for most RL scenarios and PPO is often the go to.

About

Using Deep Q learning to make an agent learn how to play snake

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages