Deep Q Network and Double Deep Q Network on Atari

This project applies ideas from research literature to solve Atari OpenAI gym environments.

Getting Started

1. Activate conda environment with dependencies installed
2. Run atari.py

Prerequisites

Project requires: Pytorch v1.4.0 installed Other dependencies include:

• Numpy
• gym
• cv2

Built With

• Pytorch - Deep learning Framework used along with Numpy to build Deep Q Networks.
• OpenAI Gym - Provides environments to test Agent's performance

Acknowledgments

This project was built referencing research papers on applying Q-learning with deep neural networks

https://deepmind.com/research/publications/human-level-control-through-deep-reinforcement-learning

https://arxiv.org/abs/1509.06461