Interpretable Graph Neural Networks for Tabular Data
It addresses the need for interpretability in GNNs for tabular data, which is common in real-world applications, but is incremental as it builds on existing GNN extensions.
The paper tackles the problem of black-box models in Graph Neural Networks (GNNs) for tabular data by proposing IGNNet, an interpretable approach that shows how predictions are computed from input features, and it performs on par with state-of-the-art methods like XGBoost and TabNet while providing explanations aligned with Shapley values.
Data in tabular format is frequently occurring in real-world applications. Graph Neural Networks (GNNs) have recently been extended to effectively handle such data, allowing feature interactions to be captured through representation learning. However, these approaches essentially produce black-box models, in the form of deep neural networks, precluding users from following the logic behind the model predictions. We propose an approach, called IGNNet (Interpretable Graph Neural Network for tabular data), which constrains the learning algorithm to produce an interpretable model, where the model shows how the predictions are exactly computed from the original input features. A large-scale empirical investigation is presented, showing that IGNNet is performing on par with state-of-the-art machine-learning algorithms that target tabular data, including XGBoost, Random Forests, and TabNet. At the same time, the results show that the explanations obtained from IGNNet are aligned with the true Shapley values of the features without incurring any additional computational overhead.