Generative-Contrastive Heterogeneous Graph Neural Network

TL;DR

This paper introduces GC-HGNN, a novel heterogeneous graph neural network that combines generative and contrastive learning to improve data augmentation, sampling, and local-global information capture, outperforming existing models.

Contribution

The paper proposes a generative-contrastive framework for HGNNs, including a masked autoencoder, advanced sampling strategies, and hierarchical contrastive learning, addressing limitations of previous contrastive methods.

Findings

Outperforms 17 baselines on node classification tasks

Achieves superior results on link prediction across 8 datasets

Enhances local and global information capture in heterogeneous graphs

Abstract

Heterogeneous Graphs (HGs) effectively model complex relationships in the real world through multi-type nodes and edges. In recent years, inspired by self-supervised learning (SSL), contrastive learning (CL)-based Heterogeneous Graphs Neural Networks (HGNNs) have shown great potential in utilizing data augmentation and contrastive discriminators for downstream tasks. However, data augmentation remains limited due to the graph data's integrity. Furthermore, the contrastive discriminators suffer from sampling bias and lack local heterogeneous information. To tackle the above limitations, we propose a novel Generative-Contrastive Heterogeneous Graph Neural Network (GC-HGNN). Specifically, we propose a heterogeneous graph generative learning method that enhances CL-based paradigm. This paradigm includes: 1) A contrastive view augmentation strategy using a masked autoencoder. 2) Position-aware and semantics-aware positive sample sampling strategy for generating hard negative samples. 3) A hierarchical contrastive learning strategy aimed at capturing local and global information. Furthermore, the hierarchical contrastive learning and sampling strategies aim to constitute an enhanced contrastive discriminator under the generative-contrastive perspective. Finally, we compare our model with seventeen baselines on eight real-world datasets. Our model outperforms the latest baselines on node classification and link prediction tasks.