Deep Reinforcment Learning Sandbox

This repo is a sandbox for trying out various Deep Reinforcement Learning algorithms on OpenAi Gym environments.

Algorithms:

• Deep Q-Learning
• Double Deep Q-Learning
• Q-Learning with prioritized experience Replay

Environments:

• cartpole
• breakout

Setup

install pipenv install requirements via pipenv start experiment

pip install pipenv
pipenv install 
pipenv shell 
cd cartpole
python run_experiment.py

Results