Self-supervised Auxiliary Learning with Meta-paths for Heterogeneous Graphs

TL;DR

This paper introduces a self-supervised auxiliary learning approach using meta-paths to enhance graph neural networks on heterogeneous graphs, automatically balancing auxiliary tasks to improve primary task performance.

Contribution

It proposes a novel meta-path based self-supervised auxiliary learning method that automatically balances auxiliary tasks for heterogeneous graph neural networks.

Findings

Improves link prediction accuracy on heterogeneous graphs

Enhances node classification performance

Applicable to various GNN architectures without extra data

Abstract

Graph neural networks have shown superior performance in a wide range of applications providing a powerful representation of graph-structured data. Recent works show that the representation can be further improved by auxiliary tasks. However, the auxiliary tasks for heterogeneous graphs, which contain rich semantic information with various types of nodes and edges, have less explored in the literature. In this paper, to learn graph neural networks on heterogeneous graphs we propose a novel self-supervised auxiliary learning method using meta-paths, which are composite relations of multiple edge types. Our proposed method is learning to learn a primary task by predicting meta-paths as auxiliary tasks. This can be viewed as a type of meta-learning. The proposed method can identify an effective combination of auxiliary tasks and automatically balance them to improve the primary task. Our methods can be applied to any graph neural networks in a plug-in manner without manual labeling or additional data. The experiments demonstrate that the proposed method consistently improves the performance of link prediction and node classification on heterogeneous graphs.