0

Deep Integrated Explanations

Deep Integrated Explanations (DIX) integrate intermediate representations and gradients to generate faithful and accurate explanation maps for vision models, outperforming existing methods.

Year
2023
Venue
arXiv 2023
Authors
6
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2310.15368v2ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

This paper presents Deep Integrated Explanations (DIX) - a universal method for explaining vision models. DIX generates explanation maps by integrating information from the intermediate representations of the model, coupled with their corresponding gradients. Through an extensive array of both objective and subjective evaluations spanning diverse tasks, datasets, and model configurations, we showcase the efficacy of DIX in generating faithful and accurate explanation maps, while surpassing current state-of-the-art methods.

Authors

6