AI-Playing-Atari-Game-BreakOut

A review over NN code creating an AI agent that can play the atari game BREAKOUT.

My last personal attempt at this problem before using C51 code is documented in file 'BREAKOUT_last_attempt.ipynb' with summary and explanations.

The solution is based on categorical C51 extension of DQN using Reinforcement Learning, original paper: https://arxiv.org/pdf/1707.06887

Name		Name	Last commit message	Last commit date
Latest commit History 24 Commits
Agent before training.mp4		Agent before training.mp4
Agent performance mid level training.mp4		Agent performance mid level training.mp4
Atari_Env_Wrappers_Explained.ipynb		Atari_Env_Wrappers_Explained.ipynb
BREAKOUT_last_attempt.ipynb		BREAKOUT_last_attempt.ipynb
Build_Neural_Network.ipynb		Build_Neural_Network.ipynb
CompleteCodeBREAKOUT.ipynb		CompleteCodeBREAKOUT.ipynb
High level training.mp4		High level training.mp4
README.md		README.md
ckpt-19.pk		ckpt-19.pk
main.py		main.py
requirements.txt		requirements.txt

Provide feedback