Decomposing and Editing Predictions by Modeling Model Computation
Harshay Shah, Andrew Ilyas, Aleksander Madry
Abstract
*How does the internal computation of a machine learning model transform inputs into predictions?* To tackle this question, we introduce a framework called *component modeling* for decomposing a model prediction in terms of its components---architectural "building blocks" such as convolution filters or attention heads. We focus on a special case of this framework, *component attribution*, where the goal is to estimate the counterfactual impact of individual components on a given prediction. We then present COAR, a scalable algorithm for estimating component attributions, and demonstrate its effectiveness across models, datasets and modalities. Finally, we show that COAR directly enables effective model editing. Our code is available at [github.com/MadryLab/modelcomponents]([https://github.com/MadryLab/modelcomponents]).
BibTeX
@inproceedings{
shah2024decomposing,
title={Decomposing and Editing Predictions by Modeling Model Computation},
author={Harshay Shah and Andrew Ilyas and Aleksander Madry},
booktitle={Forty-first International Conference on Machine Learning},
year={2024},
url={https://openreview.net/forum?id=rTBR0eqE4G}
}