Visual Causal Feature Learning

Krzysztof Chalupka Caltech, Pietro Perona Caltech, Frederick Eberhardt Caltech
Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, PMLR R13:328-337, 2015.

Abstract

We provide a rigorous definition of the visual cause of a behavior that is broadly applicable to the visually driven behavior in humans, animals, neurons, robots and other perceiving systems. Our framework generalizes standard accounts of causal learning to settings in which the causal variables need to be constructed from micro-variables. We prove the Causal Coarsening Theorem, which allows us to gain causal knowledge from observational data with minimal experimental effort. The theorem provides a connection to standard inference techniques in machine learning that identify features of an image that correlate with, but may not cause, the target behavior. Finally, we propose an active learning scheme to learn a manipulator function that performs optimal manipulations on the image to automatically identify the visual cause of a target behavior. We illustrate our inference and learning algorithms in experiments based on both synthetic and real data.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR13-caltech15a, title = {Visual Causal Feature Learning}, author = {Caltech, Krzysztof Chalupka and Caltech, Pietro Perona and Caltech, Frederick Eberhardt}, booktitle = {Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence}, pages = {328--337}, year = {2015}, editor = {Meila, Marina and Heskes, Tom}, volume = {R13}, series = {Proceedings of Machine Learning Research}, month = {12--16 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r13/main/assets/caltech15a/caltech15a.pdf}, url = {https://proceedings.mlr.press/r13/caltech15a.html}, abstract = {We provide a rigorous definition of the visual cause of a behavior that is broadly applicable to the visually driven behavior in humans, animals, neurons, robots and other perceiving systems. Our framework generalizes standard accounts of causal learning to settings in which the causal variables need to be constructed from micro-variables. We prove the Causal Coarsening Theorem, which allows us to gain causal knowledge from observational data with minimal experimental effort. The theorem provides a connection to standard inference techniques in machine learning that identify features of an image that correlate with, but may not cause, the target behavior. Finally, we propose an active learning scheme to learn a manipulator function that performs optimal manipulations on the image to automatically identify the visual cause of a target behavior. We illustrate our inference and learning algorithms in experiments based on both synthetic and real data.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Visual Causal Feature Learning %A Krzysztof Chalupka Caltech %A Pietro Perona Caltech %A Frederick Eberhardt Caltech %B Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2015 %E Marina Meila %E Tom Heskes %F pmlr-vR13-caltech15a %I PMLR %P 328--337 %U https://proceedings.mlr.press/r13/caltech15a.html %V R13 %X We provide a rigorous definition of the visual cause of a behavior that is broadly applicable to the visually driven behavior in humans, animals, neurons, robots and other perceiving systems. Our framework generalizes standard accounts of causal learning to settings in which the causal variables need to be constructed from micro-variables. We prove the Causal Coarsening Theorem, which allows us to gain causal knowledge from observational data with minimal experimental effort. The theorem provides a connection to standard inference techniques in machine learning that identify features of an image that correlate with, but may not cause, the target behavior. Finally, we propose an active learning scheme to learn a manipulator function that performs optimal manipulations on the image to automatically identify the visual cause of a target behavior. We illustrate our inference and learning algorithms in experiments based on both synthetic and real data. %Z Reissued by PMLR on 04 October 2026.
APA
Caltech, K.C., Caltech, P.P. & Caltech, F.E.. (2015). Visual Causal Feature Learning. Proceedings of the 31st Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R13:328-337 Available from https://proceedings.mlr.press/r13/caltech15a.html. Reissued by PMLR on 04 October 2026.

Related Material