Deep Reinforcement Learning Algorithms

This repository contains my implementations of some of the popular Deep Reinforcement Learning algorithms. I have focussed on code readability and tried to keep each algorithm's code as self contained as possible. So there might be some code repetition and inefficiencies.

Algorithms

• Vanilla Policy Gradient (VPG) - Code, Pseudo Code
• Deep Q Network (DQN) - Code, Pseudo Code
• Asynchronous Advantage Actor Critic (A3C) Code Pseudo Code
• Proximal Policy Optimization (PPO) Code Pseudo Code
• Deep Deterministic Policy Gradients (DDPG)

Requirements

This code is written for Python 3.6 and PyTorch 1.1. Install the dependencies by the following command,

pip install -r requirements.txt