Feature Synergy, Redundancy, and Independence in Global Model Explanations using SHAP Vector Decomposition
Machine Learning
2021-07-28 v1 Artificial Intelligence
Abstract
We offer a new formalism for global explanations of pairwise feature dependencies and interactions in supervised models. Building upon SHAP values and SHAP interaction values, our approach decomposes feature contributions into synergistic, redundant and independent components (S-R-I decomposition of SHAP vectors). We propose a geometric interpretation of the components and formally prove its basic properties. Finally, we demonstrate the utility of synergy, redundancy and independence by applying them to a constructed data set and model.
Cite
@article{arxiv.2107.12436,
title = {Feature Synergy, Redundancy, and Independence in Global Model Explanations using SHAP Vector Decomposition},
author = {Jan Ittner and Lukasz Bolikowski and Konstantin Hemker and Ricardo Kennedy},
journal= {arXiv preprint arXiv:2107.12436},
year = {2021}
}
Comments
7 pages, 2 figures