Interpretable Graph Neural Networks for Heterogeneous Tabular Data

TL;DR

This paper introduces IGNH, an interpretable graph neural network designed for heterogeneous tabular data, providing accurate predictions with transparent feature attributions, outperforming some existing models.

Contribution

The paper presents IGNH, a novel GNN model that handles heterogeneous tabular data and offers exact feature attributions, addressing interpretability and performance limitations of prior methods.

Findings

IGHN's feature attributions align with Shapley values.

IGHN outperforms Random Forests and TabNet.

IGHN achieves similar performance to XGBoost.

Abstract

Many machine learning algorithms for tabular data produce black-box models, which prevent users from understanding the rationale behind the model predictions. In their unconstrained form, graph neural networks fall into this category, and they have further limited abilities to handle heterogeneous data. To overcome these limitations, an approach is proposed, called IGNH (Interpretable Graph Neural Network for Heterogeneous tabular data), which handles both categorical and numerical features, while constraining the learning process to generate exact feature attributions together with the predictions. A large-scale empirical investigation is presented, showing that the feature attributions provided by IGNH align with Shapley values that are computed post hoc. Furthermore, the results show that IGNH outperforms two powerful machine learning algorithms for tabular data, Random Forests and TabNet, while reaching a similar level of performance as XGBoost.