Self-Explainable Temporal Graph Networks based on Graph Information Bottleneck

TL;DR

This paper introduces TGIB, a novel self-explainable temporal graph neural network that integrates prediction and explanation in an end-to-end framework, improving both accuracy and interpretability for dynamic graph data.

Contribution

The work presents the first end-to-end model for simultaneous prediction and explanation in temporal graphs using the Graph Information Bottleneck theory.

Findings

TGIB outperforms state-of-the-art methods in link prediction accuracy.

TGIB provides more accurate and interpretable explanations.

Experimental results confirm the effectiveness of the integrated approach.

Abstract

Temporal Graph Neural Networks (TGNN) have the ability to capture both the graph topology and dynamic dependencies of interactions within a graph over time. There has been a growing need to explain the predictions of TGNN models due to the difficulty in identifying how past events influence their predictions. Since the explanation model for a static graph cannot be readily applied to temporal graphs due to its inability to capture temporal dependencies, recent studies proposed explanation models for temporal graphs. However, existing explanation models for temporal graphs rely on post-hoc explanations, requiring separate models for prediction and explanation, which is limited in two aspects: efficiency and accuracy of explanation. In this work, we propose a novel built-in explanation framework for temporal graphs, called Self-Explainable Temporal Graph Networks based on Graph Information Bottleneck (TGIB). TGIB provides explanations for event occurrences by introducing stochasticity in each temporal event based on the Information Bottleneck theory. Experimental results demonstrate the superiority of TGIB in terms of both the link prediction performance and explainability compared to state-of-the-art methods. This is the first work that simultaneously performs prediction and explanation for temporal graphs in an end-to-end manner.