This repository is a curated list of resources for Deep Reinforcement Learning (DRL), serving as a comprehensive guide for researchers and practitioners. It aims to consolidate libraries, benchmarks, environments, competitions, books, tutorials, and blogs related to DRL, providing a centralized hub for learning and development in the field.

How It Works

The repository functions as a directory, categorizing and linking to a vast array of DRL-related projects and materials. It highlights key libraries, simulators, and research frameworks, offering insights into their design and capabilities. The curated nature of the list allows users to quickly identify relevant tools and learning resources without extensive searching.

Quick Start & Requirements

This is a curated list, not a runnable project. No installation or specific requirements are applicable.

Highlighted Details

• Extensive catalog of DRL libraries, including major frameworks like Ray RLLib, DeepMind Acme, TF-Agents, and Stable Baselines.
• Comprehensive list of DRL environments, spanning simulators for autonomous driving (Carla, AirSim), robotics (MuJoCo, Holodeck), and game-based platforms (OpenAI Gym, DeepMind Lab, Unity ML-Agents).
• Historical timeline of key DRL milestones, algorithms, and influential papers from 1947 to the present.
• Curated list of books, tutorials, and blogs from leading researchers and institutions like DeepMind, OpenAI, and Berkeley.

Maintenance & Community

The list is maintained by kengz and includes contributions from various prominent organizations and researchers in the AI community, such as DeepMind, OpenAI, Facebook AI, and Google AI. Links to relevant community resources and learning platforms are provided.

Licensing & Compatibility

The repository itself is licensed under the MIT License, promoting broad use and contribution. However, the licensing of the individual projects linked within the list varies significantly, and users must consult the specific licenses of each referenced library or environment for compatibility and usage terms.

Limitations & Caveats

As a curated list, this repository does not provide runnable code or direct support for any of the listed projects. The rapidly evolving nature of DRL means that some links or project statuses may become outdated over time. Users are responsible for verifying the current state and compatibility of individual resources.