Edge-Level Explanations for Graph Neural Networks by Extending Explainability Methods for Convolutional Neural Networks

TL;DR

This paper extends CNN explainability methods like LIME and Grad-CAM to graph neural networks to identify important edges, demonstrating that LIME-based explanations are most effective across various tasks.

Contribution

It introduces a novel extension of CNN explainability techniques to GNNs, enabling edge-level interpretability of model decisions.

Findings

LIME-based explanations outperform other methods in efficiency.

The approach effectively identifies important edges in real-world graph tasks.

LIME-based method surpasses state-of-the-art GNN explainability techniques.

Abstract

Graph Neural Networks (GNNs) are deep learning models that take graph data as inputs, and they are applied to various tasks such as traffic prediction and molecular property prediction. However, owing to the complexity of the GNNs, it has been difficult to analyze which parts of inputs affect the GNN model's outputs. In this study, we extend explainability methods for Convolutional Neural Networks (CNNs), such as Local Interpretable Model-Agnostic Explanations (LIME), Gradient-Based Saliency Maps, and Gradient-Weighted Class Activation Mapping (Grad-CAM) to GNNs, and predict which edges in the input graphs are important for GNN decisions. The experimental results indicate that the LIME-based approach is the most efficient explainability method for multiple tasks in the real-world situation, outperforming even the state-of-the-art method in GNN explainability.