phlippe/CITRIS

Code repository of the paper "CITRIS: Causal Identifiability from Temporal Intervened Sequences" and "iCITRIS: Causal Representation Learning for Instantaneous Temporal Effects"

40
/ 100
Emerging

This project helps researchers and scientists understand the underlying cause-and-effect relationships in dynamic visual systems, like how different factors influence an object's movement or changes in a game. You input sequences of images where you've made specific changes (interventions), and it outputs the hidden causal factors at play, even when effects are rapid. This is for researchers in AI, robotics, or causal inference who work with visual data.

No commits in the last 6 months.

Use this if you are a researcher studying complex systems from video or image sequences and need to identify the independent causal variables and their relationships, especially when interventions are possible.

Not ideal if you don't have temporal sequences of data, cannot perform interventions, or are not working with visual observations.

causal-inference robotics-control dynamic-systems-analysis computer-vision-research experimental-design
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

58

Forks

10

Language

Python

License

BSD-3-Clause-Clear

Last pushed

Jun 16, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/phlippe/CITRIS"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.