Self-Explainable Multi-Label Graph Neural Network for Correlated Evidence Attribution

2026-08-27

AI Summary

本文提出了一种自解释多标签图神经网络(SEMGNN)，通过同时学习预测器和稀疏边掩码解释器来解决多标签图学习中的证据归属问题。

Abstract

Multi-label graph learning intends to capture the intrinsic complexity of real-world applications, where one sample is often related to multiple groups or consists of multiple objects. To date, a handful of multi-label graph learning methods exist, but none of them integrate training-time interpretation capability. While post-hoc graph explainers have been developed, they do not explicitly model label-dependent evidence sharing in multi-label graph learners, especially when label pairs are weakly or negatively associated. As a result, post-hoc approaches may miss how evidence should be shared or separated across different labels. This paper advances a new end-to-end self-explainable multi-label graph neural network (SEMGNN), which aims to simultaneously classify multi-labeled nodes and identify edges significantly contributing to each target node w.r.t. predicted labels. Different from post-hoc methods, SEMGNN jointly learns a predictor and a sparse edge-mask explainer within a unified framework and training objective. Label-label correlations are used to improve multi-label node classification and enhance individual label explanations, so that different labels of a node can be supported by distinct yet coherent structural and/or correlated evidence. Experiments and comparisons on synthetic and real-world multi-label networks, in social networking, entertainment, and life sciences, show that SEMGNN achieves competitive or improved predictive performance while providing more faithful and compact label-conditioned explanations.

Problem

• multi-label graph learning
• interpretation capability
• label-dependent evidence sharing
• label correlations
• edge-mask explainer

Innovation

• self-explainable
• multi-label graph neural network
• label correlations
• end-to-end learning
• sparse edge-mask explainer