Decomposing and Editing Predictions by Modeling Model Computation
Shah, Harshay, Ilyas, Andrew, Madry, Aleksander
How does the internal computation of a machine learning model transform inputs into predictions? In this paper, we introduce a task called component modeling that aims to address this question. The goal of component modeling is to decompose an ML model's prediction in terms of its components -- simple functions (e.g., convolution filters, attention heads) that are the "building blocks" of model computation. We focus on a special case of this task, 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; we demonstrate its effectiveness across models, datasets, and modalities. Finally, we show that component attributions estimated with COAR directly enable model editing across five tasks, namely: fixing model errors, ``forgetting'' specific classes, boosting subpopulation robustness, localizing backdoor attacks, and improving robustness to typographic attacks. We provide code for COAR at https://github.com/MadryLab/modelcomponents .
Apr-17-2024
- Country:
- North America > United States (0.92)
- Genre:
- Research Report > New Finding (0.46)
- Industry:
- Technology:
- Information Technology
- Artificial Intelligence
- Machine Learning > Neural Networks
- Deep Learning (0.93)
- Natural Language (1.00)
- Vision (1.00)
- Machine Learning > Neural Networks
- Data Science (1.00)
- Artificial Intelligence
- Information Technology