Overview

This repository contains the projects that I completed as part of the Deep Reinforcement Learning course offered by HuggingFace. The course covered various aspects of deep reinforcement learning, including Q-learning, policy gradients, actor-critic methods, and deep deterministic policy gradients (DDPG).

The projects in this repository demonstrate the application of these techniques to different environments. Each project includes README.md file that explains the project detailes.

Projects

• Vizdoom Health Gathering Supreme: This project uses a Proximal Policy Optimization (PPO) algorithm to train an agent to gather health in the Vizdoom environment.
• Reinforce Pixelcopter PLE v0: This project uses a policy gradient algorithm to train an agent to fly a pixelcopter in the PLE environment.
• A2C PandaReachDense-v2: This project uses an Advantage Actor-Critic (A2C) algorithm to train an agent to reach a goal with a panda arm in the MuJoCo environment.
• DQN SpaceInvadersNoFrameskip-v4: This project uses a Deep Q-Network (DQN) algorithm to train an agent to play Space Invaders in the Atari environment.
• Q-Taxi-v3: This project uses a Q-learning algorithm to train an agent to navigate a taxi in the OpenAI Gym environment.
• A2C AntBulletEnv-v0: This project uses an A2C algorithm to train an agent to control an ant in the Bullet environment.
• PPO Pyramids: This project uses a PPO algorithm to train an agent to navigate a pyramid in the Gym environment.
• PPO SnowballTarget: This project uses a PPO algorithm to train an agent to throw snowballs at a target in the Gym environment.
• CartPole-v1: This project uses a Q-learning algorithm to train an agent to balance a pole on a moving cart.
• Q-FrozenLake-v1-4x4-noSlippery: This project uses a Q-learning algorithm to train an agent to navigate a frozen lake.
• PPO Huggy: This project uses a PPO algorithm to train an agent to hug a teddy bear in the Gym environment.
• PPO LunarLander-v2: This project uses a PPO algorithm to train an agent to land a lunar lander safely.

Requirements

• Python 3
• NumPy
• PyTorch

You can find the original repos here: https://huggingface.co/arhamk