Rectifying Group Irregularities in Explanations for Distribution Shift

Adam Stein, Yinjun Wu, Eric Wong, Mayur Naik

Primary Area: visualization or interpretation of learned representations

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: explainability, distribution shift, group robust

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Abstract:

It is well-known that real-world changes constituting distribution shift adversely affect model performance.
How to characterize those changes in an interpretable manner is poorly understood.
Existing techniques take the form of shift explanations that elucidate how samples map from the original distribution toward the shifted one by reducing the disparity between the two distributions.
However, these methods can introduce group irregularities, leading to explanations that are less feasible and robust.
To address these issues, we propose Group-aware Shift Explanations (GSE), an explanation method that leverages worst-group optimization to rectify group irregularities.
We demonstrate that GSE not only maintains group structures, but can improve feasibility and robustness over a variety of domains by up to 20% and 25% respectively.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 8390