Saved in:
Bibliographic Details
Main Authors: You, Weiqiu, Qu, Helen, Gatti, Marco, Jain, Bhuvnesh, Wong, Eric
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.16316
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Self-attributing neural networks (SANNs) present a potential path towards interpretable models for high-dimensional problems, but often face significant trade-offs in performance. In this work, we formally prove a lower bound on errors of per-feature SANNs, whereas group-based SANNs can achieve zero error and thus high performance. Motivated by these insights, we propose Sum-of-Parts (SOP), a framework that transforms any differentiable model into a group-based SANN, where feature groups are learned end-to-end without group supervision. SOP achieves state-of-the-art performance for SANNs on vision and language tasks, and we validate that the groups are interpretable on a range of quantitative and semantic metrics. We further validate the utility of SOP explanations in model debugging and cosmological scientific discovery. Our code is available at https://github.com/BrachioLab/sop