wtf is 2021-hyu-hai-rlbootcamp?

This repository is a teaching toolkit for an intermediate-level course on reinforcement learning, originally run in 2021, 2022 for the HAI club at Hanyang University. It brings together lecture slides, example code, and assignments in one place so that a student can follow a structured path from the basics of reinforcement learning through to fairly advanced topics. All code examples use PyTorch. Reinforcement learning is a branch of machine learning where an agent learns to make good decisions by trial and error, getting rewards or penalties for its actions, similar to how a player learns a game through practice. The course assumes you already know the fundamentals, since Week 1 is a review of core ideas like Q-learning and policy gradients. From there, each week layers on a new family of techniques: improved versions of Deep Q-Networks, actor-critic methods, algorithms that handle continuous actions, approaches that keep policy updates stable (like PPO), and finally multi-agent scenarios where several agents learn to interact at once. The audience is people who already have a foothold in machine learning or reinforcement learning and want to go deeper. A self-taught developer who has watched intro videos and wants to understand how modern game-playing AI or robotics control actually works would benefit. The layout, one folder of PDF lectures, one folder of code, makes it a self-study resource as much as a classroom one. The course follows the book "Deep Reinforcement Learning Hands-On, Second Edition" and lists several optional readings, including Korean-language books, suggesting it was designed for a Korean-speaking student audience. The README does not go into detail about prerequisites beyond that assumed baseline knowledge. The project is open source under the MIT license and welcomes contributions.