CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networks

TL;DR

CF-GNNExplainer introduces a novel approach for generating counterfactual explanations for GNNs by minimally perturbing input graphs through edge deletions to understand prediction changes.

Contribution

The paper presents the first method for counterfactual explanations in GNNs, focusing on minimal edge deletions to alter predictions.

Findings

Achieves over 94% accuracy in generating CF explanations.

Removes less than 3 edges on average per explanation.

Effectively identifies crucial edges influencing GNN predictions.

Abstract

Given the increasing promise of graph neural networks (GNNs) in real-world applications, several methods have been developed for explaining their predictions. Existing methods for interpreting predictions from GNNs have primarily focused on generating subgraphs that are especially relevant for a particular prediction. However, such methods are not counterfactual (CF) in nature: given a prediction, we want to understand how the prediction can be changed in order to achieve an alternative outcome. In this work, we propose a method for generating CF explanations for GNNs: the minimal perturbation to the input (graph) data such that the prediction changes. Using only edge deletions, we find that our method, CF-GNNExplainer, can generate CF explanations for the majority of instances across three widely used datasets for GNN explanations, while removing less than 3 edges on average, with at least 94\% accuracy. This indicates that CF-GNNExplainer primarily removes edges that are crucial for the original predictions, resulting in minimal CF explanations.