Saved in:
Bibliographic Details
Main Authors: Kamalov, Firuz, Falasi, Mohmad Al, Thabtah, Fadi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.17491
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909799520665600
author Kamalov, Firuz
Falasi, Mohmad Al
Thabtah, Fadi
author_facet Kamalov, Firuz
Falasi, Mohmad Al
Thabtah, Fadi
contents Integrated Gradients (IG) is a widely used attribution method in explainable artificial intelligence (XAI). In this paper, we introduce Path-Weighted Integrated Gradients (PWIG), a generalization of IG that incorporates a customizable weighting function into the attribution integral. This modification allows for targeted emphasis along different segments of the path between a baseline and the input, enabling improved interpretability, noise mitigation, and the detection of path-dependent feature relevance. We establish its theoretical properties and illustrate its utility through experiments on a dementia classification task using the OASIS-1 MRI dataset. Attribution maps generated by PWIG highlight clinically meaningful brain regions associated with various stages of dementia, providing users with sharp and stable explanations. The results suggest that PWIG offers a flexible and theoretically grounded approach for enhancing attribution quality in complex predictive models.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17491
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Path-Weighted Integrated Gradients for Interpretable Dementia Classification
Kamalov, Firuz
Falasi, Mohmad Al
Thabtah, Fadi
Machine Learning
Integrated Gradients (IG) is a widely used attribution method in explainable artificial intelligence (XAI). In this paper, we introduce Path-Weighted Integrated Gradients (PWIG), a generalization of IG that incorporates a customizable weighting function into the attribution integral. This modification allows for targeted emphasis along different segments of the path between a baseline and the input, enabling improved interpretability, noise mitigation, and the detection of path-dependent feature relevance. We establish its theoretical properties and illustrate its utility through experiments on a dementia classification task using the OASIS-1 MRI dataset. Attribution maps generated by PWIG highlight clinically meaningful brain regions associated with various stages of dementia, providing users with sharp and stable explanations. The results suggest that PWIG offers a flexible and theoretically grounded approach for enhancing attribution quality in complex predictive models.
title Path-Weighted Integrated Gradients for Interpretable Dementia Classification
topic Machine Learning
url https://arxiv.org/abs/2509.17491