lucidrains/metacontroller
Implementation of the MetaController proposed in "Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning" from the Paradigms of Intelligence team at Google
This is an implementation for developers working on advanced AI agents in environments requiring complex, sequential decision-making. It helps in training agents that can learn both low-level actions and higher-level, more abstract goals by processing raw state observations and desired actions. AI researchers and machine learning engineers developing reinforcement learning systems would find this useful for creating more efficient and capable agents.
Use this if you are developing AI agents for environments where emergent temporal abstractions can improve learning efficiency and task performance.
Not ideal if you are looking for a pre-trained, ready-to-deploy agent or a solution for simple, single-step decision-making tasks.
Stars
93
Forks
11
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 26, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/lucidrains/metacontroller"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
DLR-RM/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
google-deepmind/dm_control
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning...
Denys88/rl_games
RL implementations
pytorch/rl
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
yandexdataschool/Practical_RL
A course in reinforcement learning in the wild