LightZero is an efficient, scalable, and open-source framework implementing MuZero, a powerful model-based reinforcement learning algorithm that learns to predict rewards and transitions without explicit environment models. Developed by OpenDILab, LightZero focuses on providing a highly optimized and user-friendly platform for both academic research and industrial applications of MuZero and similar algorithms.
Features
- Full implementation of MuZero for model-based RL
- Highly efficient and scalable for large-scale training
- Modular and extensible architecture for custom research
- Supports a variety of Gym and custom environments
- Includes pre-built pipelines for training, evaluation, and benchmarking
- Offers tools for visualizing agent learning and model predictions
Categories
Reinforcement Learning AlgorithmsLicense
Apache License V2.0Follow LightZero
Other Useful Business Software
Fully Managed MySQL, PostgreSQL, and SQL Server
Cloud SQL handles your database ops end to end, so you can focus on your app.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of LightZero!