---
title: "Pairwise Half-graph Discrimination: A Simple Graph-level Self-supervised Strategy for Pre-training Graph Neural Networks"
authors:
  - "Pengyong Li"
  - "Jun Wang"
  - "Ziliang Li"
  - "Yixuan Qiao"
  - "Xianggen Liu"
  - "Fei Ma"
  - "Peng Gao"
  - "Sen Song"
  - "Guotong Xie"
doi: "10.24963/ijcai.2021/371"
url: "https://ijcai.org/proceedings/2021/371"
pdf: "https://ijcai.org/proceedings/2021/0371.pdf"
venue: "Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence"
year: 2021
pages: "2694-2700"
track: "Main Track"
keywords:
  - "Machine Learning: Unsupervised Learning"
  - "Machine Learning: Deep Learning"
  - "Data Mining: Mining Graphs, Semi Structured Data, Complex Data"
---

# Pairwise Half-graph Discrimination: A Simple Graph-level Self-supervised Strategy for Pre-training Graph Neural Networks

Pengyong Li, Jun Wang, Ziliang Li, Yixuan Qiao, Xianggen Liu, Fei Ma, Peng Gao, Sen Song, Guotong Xie

*Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence*, pages 2694-2700. DOI: [10.24963/ijcai.2021/371](https://doi.org/10.24963/ijcai.2021/371)

## Abstract

Self-supervised learning has gradually emerged as a powerful technique for graph representation learning. However, transferable, generalizable, and robust representation learning on graph data still remains a challenge for pre-training graph neural networks. In this paper, we propose a simple and effective self-supervised pre-training strategy, named Pairwise Half-graph Discrimination (PHD), that explicitly pre-trains a graph neural network at graph-level. PHD is designed as a simple binary classification task to discriminate whether two half-graphs come from the same source. Experiments demonstrate that the PHD is an effective pre-training strategy that offers comparable or superior performance on 13 graph classification tasks compared with state-of-the-art strategies, and achieves notable improvements when combined with node-level strategies. Moreover, the visualization of learned representation revealed that PHD strategy indeed empowers the model to learn graph-level knowledge like the molecular scaffold. These results have established PHD as a powerful and effective self-supervised learning strategy in graph-level representation learning.