Statistical Test for Saliency Maps of Graph Neural Networks via Selective Inference

TL;DR

This paper introduces a statistical testing framework using Selective Inference to evaluate the significance of saliency maps in GNNs, ensuring reliable interpretation by controlling false positives.

Contribution

It presents a novel statistical method to validate GNN saliency maps, addressing data double-dipping and providing valid p-values for interpretability.

Findings

Validates the method on synthetic datasets

Demonstrates improved reliability of saliency maps

Controls Type I error rate effectively

Abstract

Graph Neural Networks (GNNs) have gained prominence for their ability to process graph-structured data across various domains. However, interpreting GNN decisions remains a significant challenge, leading to the adoption of saliency maps for identifying salient subgraphs composed of influential nodes and edges. Despite their utility, the reliability of GNN saliency maps has been questioned, particularly in terms of their robustness to input noise. In this study, we propose a statistical testing framework to rigorously evaluate the significance of saliency maps. Our main contribution lies in addressing the inflation of the Type I error rate caused by double-dipping of data, leveraging the framework of Selective Inference. Our method provides statistically valid $p$-values while controlling the Type I error rate, ensuring that identified salient subgraphs contain meaningful information rather than random artifacts. The method is applicable to a variety of saliency methods with piecewise linearity (e.g., Class Activation Mapping). We validate our method on synthetic and real-world datasets, demonstrating its capability in assessing the reliability of GNN interpretations.