This repository provides comprehensive materials for a Reinforcement Learning course, including lecture notes, video lectures, and Python-based tutorial exercises with solutions. It is designed for both students seeking self-learning resources and lecturers looking to establish their own courses, offering a structured curriculum covering foundational RL concepts to contemporary algorithms.

How It Works

The course material is structured into distinct modules, each with accompanying video lectures and slide decks. The practical component involves Python 3.9 tutorials that implement various RL algorithms, often leveraging the Gymnasium library (a fork of OpenAI Gym) for environment simulation. This hands-on approach allows learners to directly apply theoretical concepts.

Quick Start & Requirements

• Install dependencies using pip install -r requirements.txt.
• Requires Python 3.9.
• Links to lecture videos, slides, and tutorial templates are provided within the README.

Highlighted Details

• Covers a broad spectrum of RL topics from MDPs and dynamic programming to function approximation and policy gradient methods.
• Includes contemporary algorithms like TRPO and PPO.
• Tutorials are based on Gymnasium environments, facilitating practical implementation.
• Course materials are inspired by seminal works like Sutton & Barto's "Reinforcement Learning: An Introduction" and David Silver's UCL course.

Maintenance & Community

The project welcomes contributions via issues for feedback and pull requests for new content. Direct contact is encouraged for larger contributions.

Licensing & Compatibility

The source code is open for use and adaptation. Specific license details are not explicitly stated in the README, but the intent is for broad, non-commercial and educational use.

Limitations & Caveats

One tutorial video from the 2022 edition is unavailable due to a technical outage. The README does not specify a formal license, which may require clarification for commercial or closed-source integration.