Graph-Based Neural Network Models with Multiple Self-Supervised Auxiliary Tasks

TL;DR

This paper introduces three novel self-supervised auxiliary tasks for graph neural networks, enhancing their ability to learn robust representations through multi-task learning on structured data.

Contribution

It proposes new self-supervised auxiliary tasks for graph neural networks, improving multi-task learning and semi-supervised graph classification performance.

Findings

Achieved competitive results on standard graph classification benchmarks.

Demonstrated the effectiveness of multi-task self-supervised learning for graph models.

Enhanced representation learning in graph neural networks.

Abstract

Self-supervised learning is currently gaining a lot of attention, as it allows neural networks to learn robust representations from large quantities of unlabeled data. Additionally, multi-task learning can further improve representation learning by training networks simultaneously on related tasks, leading to significant performance improvements. In this paper, we propose three novel self-supervised auxiliary tasks to train graph-based neural network models in a multi-task fashion. Since Graph Convolutional Networks are among the most promising approaches for capturing relationships among structured data points, we use them as a building block to achieve competitive results on standard semi-supervised graph classification tasks.