Skip to content

Playing the Snake game with reinforcement learning.

Notifications You must be signed in to change notification settings

bwubrian/snakeai

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Using Reinforcement Learning to play the Snake game

Trained with Unity's implentation of PPO on various board sizes.

Shown are nine independent parallel randomized instantiations on an 8x8 board.

Alt text

Block Color Game Object
Red Snake Head
White Snake Body
Yellow Food Pellet

Steps are incremented each time the agent takes an action. Score is incremented each time the snake eats a food pellet and increases in length by one. The current model is on average able to consistently acheive a score around half the total size of the board (e.g. length of 32 for board size of 8x8) but rarely gets far above that.

About

Playing the Snake game with reinforcement learning.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published