GNN-MultiFix: Addressing the pitfalls for GNNs for multi-label node classification

TL;DR

This paper critically examines the limitations of GNNs in multi-label node classification, revealing their failure to learn effectively without node attributes or label info, and proposes GNN-MultiFix to address these issues.

Contribution

The paper introduces GNN-MultiFix, a simple yet effective method that incorporates feature, label, and positional information to improve multi-label node classification with GNNs.

Findings

GNNs fail to learn well on multi-label datasets even with abundant data.

Expressive GNNs struggle without node attributes or label inputs in transductive settings.

GNN-MultiFix significantly improves performance across multiple datasets.

Abstract

Graph neural networks (GNNs) have emerged as powerful models for learning representations of graph data showing state of the art results in various tasks. Nevertheless, the superiority of these methods is usually supported by either evaluating their performance on small subset of benchmark datasets or by reasoning about their expressive power in terms of certain graph isomorphism tests. In this paper we critically analyse both these aspects through a transductive setting for the task of node classification. First, we delve deeper into the case of multi-label node classification which offers a more realistic scenario and has been ignored in most of the related works. Through analysing the training dynamics for GNN methods we highlight the failure of GNNs to learn over multi-label graph datasets even for the case of abundant training data. Second, we show that specifically for transductive node classification, even the most expressive GNN may fail to learn in absence of node attributes and without using explicit label information as input. To overcome this deficit, we propose a straightforward approach, referred to as GNN-MultiFix, that integrates the feature, label, and positional information of a node. GNN-MultiFix demonstrates significant improvement across all the multi-label datasets. We release our code at https://anonymous.4open.science/r/Graph-MultiFix-4121.