Acme
A research framework for reinforcement learning providing modular building blocks and reference agent implementations.
What is Acme?
Acme is a research framework for reinforcement learning that provides a library of modular building blocks and reference agent implementations. It solves the problem of creating efficient, readable, and scalable RL agents by offering well-documented components that serve as both performance baselines and starting points for novel research.
Reinforcement learning researchers and practitioners who need reference implementations, strong baselines, and flexible building blocks for developing and testing new RL algorithms.
Developers choose Acme because it offers production-quality, scalable agent implementations from DeepMind, with a focus on readability and research flexibility that allows easy extension and modification for experimental work.
Overview
A library of reinforcement learning components and agents
Use Cases
Best For
- Implementing and benchmarking reinforcement learning algorithms
- Building novel RL agents using modular components
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
