Probing GNN Explainers: A Rigorous Theoretical and Empirical Analysis of GNN Explanation Methods
This work addresses the need for reliable explanations in critical GNN applications, providing a systematic analysis that is foundational for the field.
The paper tackled the problem of analyzing the reliability of Graph Neural Network (GNN) explanation methods by conducting the first theoretical analysis of state-of-the-art methods with respect to properties like faithfulness and stability, establishing upper bounds on violations, and empirically validating these findings using nine real-world datasets.
As Graph Neural Networks (GNNs) are increasingly being employed in critical real-world applications, several methods have been proposed in recent literature to explain the predictions of these models. However, there has been little to no work on systematically analyzing the reliability of these methods. Here, we introduce the first-ever theoretical analysis of the reliability of state-of-the-art GNN explanation methods. More specifically, we theoretically analyze the behavior of various state-of-the-art GNN explanation methods with respect to several desirable properties (e.g., faithfulness, stability, and fairness preservation) and establish upper bounds on the violation of these properties. We also empirically validate our theoretical results using extensive experimentation with nine real-world graph datasets. Our empirical results further shed light on several interesting insights about the behavior of state-of-the-art GNN explanation methods.