Toward Multiple Specialty Learners for Explaining GNNs via Online Knowledge Distillation

TL;DR

SCALE is a novel, general, and fast framework for explaining GNN predictions by training multiple specialty learners guided by online knowledge distillation, providing structural and feature attributions for graph and node predictions.

Contribution

The paper introduces SCALE, a new GNN explanation framework that employs multiple learners and online knowledge distillation for efficient, general, and accurate explanations.

Findings

SCALE outperforms state-of-the-art baselines in explanation correctness.

SCALE demonstrates fast explanation generation with competitive accuracy.

Ablation studies reveal the effectiveness of multiple learners and distillation approach.

Abstract

Graph Neural Networks (GNNs) have become increasingly ubiquitous in numerous applications and systems, necessitating explanations of their predictions, especially when making critical decisions. However, explaining GNNs is challenging due to the complexity of graph data and model execution. Despite additional computational costs, post-hoc explanation approaches have been widely adopted due to the generality of their architectures. Intrinsically interpretable models provide instant explanations but are usually model-specific, which can only explain particular GNNs. Therefore, we propose a novel GNN explanation framework named SCALE, which is general and fast for explaining predictions. SCALE trains multiple specialty learners to explain GNNs since constructing one powerful explainer to examine attributions of interactions in input graphs is complicated. In training, a black-box GNN model guides learners based on an online knowledge distillation paradigm. In the explanation phase, explanations of predictions are provided by multiple explainers corresponding to trained learners. Specifically, edge masking and random walk with restart procedures are executed to provide structural explanations for graph-level and node-level predictions, respectively. A feature attribution module provides overall summaries and instance-level feature contributions. We compare SCALE with state-of-the-art baselines via quantitative and qualitative experiments to prove its explanation correctness and execution performance. We also conduct a series of ablation studies to understand the strengths and weaknesses of the proposed framework.