NACLab/self-revising-active-inference
Official Implementation for the paper "SR-AIF: Solving Sparse-Reward Robotic Tasks from Pixels with Active Inference and World Models"
This project helps robotics engineers develop and train AI agents that can solve complex robotic tasks, especially when the robot receives very infrequent feedback (sparse rewards) and has to learn from raw visual information. It takes in visual input (pixels) and previous attempts from an "expert" agent, then outputs a more robust and effective control policy for the robot. Robotics researchers and engineers working on autonomous systems or robot control would find this useful.
No commits in the last 6 months.
Use this if you are developing AI for robots that need to learn intricate physical tasks from visual input and sparse rewards, aiming for better performance than initial expert demonstrations.
Not ideal if your robotic tasks have dense, frequent reward signals or if your agents don't rely on visual learning from pixels.
Stars
64
Forks
3
Language
Python
License
—
Category
Last pushed
Sep 06, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NACLab/self-revising-active-inference"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
enactic/openarm
A fully open-source humanoid arm for physical AI research and deployment in contact-rich environments.
Shuijing725/awesome-robot-social-navigation
A curated list of robot social navigation.
thomashiemstra/fred
This my 3d printed robot arm project
jstmn/ikflow
Open source implementation to the paper "IKFlow: Generating Diverse Inverse Kinematics Solutions"
sizhe-li/neural-jacobian-field
Controlling diverse robots by inferring jacobian fields with deep networks! Let's make robots...