/Deep-Arcade

Library of experimental deep learning reinforcement game simulations.

Primary LanguagePython

The Deep Arcade

A Reinforcement Learning (Deep RL) Game Library

Welcome to the Deep Reinforcement Learning Game Library!

This is a collection of games and playful simulations designed with two goals in mind:

  • To pit users against custom-trained deep learning bots.
  • To introduce fellow gamers and developers to a profound application of programming in game development!

Description

Deep Reinforcement Learning is a powerful tool both blessed and cursed by its complexity and creative uses.

With these projects, I aim to learn more and dig deeper into the mathematical and algorithmic mechanics behind deep learning and data science, so to better understand its usefulness in today's world.

Please star this repo and follow my progress into the world of deep learning!

:-)

Progression

CURRENT GAMES:

  • Tic Tac Toe (In Progress.)
  • Snake (v0 Complete!)

FUTURE GAMES:

  • Super Mario Brothers
  • The Binding of Isaac
  • Grand Theft Auto 5

Technologies Used:

  • Python 3.x
    • PyGame
    • Keras (TensorFlow)
    • NumPy, Pandas
    • Seaborn, MatPlotLib

Credits

If you would like to contact me to provide feedback, ask any questions, or otherwise outreach, please email me at aakash.sudhakar@gmail.com.

Created and maintained by Aakash Sudhakar.

Special thanks to Mauro Comi's brilliant Towards Data Science article that inspired these projects.