Visual Causal Feature Learning
- Others:
- Mella, Marina
- Heskes, Tom
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.
Additional Information
© 2015 AUAI Press. KC's work was funded by the Qualcomm Innovation Fellowship 2014. KC's and PP's work was supported by the ONR MURI grant N00014-10-1-0933. FE would like to thank Cosma Shalizi for pointers to many relevant results this paper builds on.Attached Files
Accepted Version - 1412.2309.pdf
Files
Name | Size | Download all |
---|---|---|
md5:d37f30b56a7d3c04e483eb196245934b
|
1.1 MB | Preview Download |
Additional details
- Eprint ID
- 94199
- Resolver ID
- CaltechAUTHORS:20190327-085913684
- Qualcomm Inc.
- Office of Naval Research (ONR)
- N00014-10-1-0933
- Created
-
2019-03-27Created from EPrint's datestamp field
- Updated
-
2023-06-02Created from EPrint's last_modified field