Editing a classifier by rewriting its prediction rules
Shibani Santurkar ⋅ Dimitris Tsipras ⋅ Mahalaxmi Elango ⋅ David Bau ⋅ Antonio Torralba ⋅ Aleksander Madry
Keywords:
Robustness
2021 Poster
Abstract
We propose a methodology for modifying the behavior of a classifier by directly rewriting its prediction rules. Our method requires virtually no additional data collection and can be applied to a variety of settings, including adapting a model to new environments, and modifying it to ignore spurious features.
Video
Chat is not available.
Successful Page Load