Sailor: Structural Augmentation Based Tail Node Representation Learning

By J. Liao et al

Published on Jan. 1, 2023

Read the original document by opening this link in a new tab.

Table of Contents

1. Introduction
2. Related Work
3. Preliminaries
4. The Heterophily of Tail Nodes

Summary

The document discusses the SAILOR framework for improving representation learning for tail nodes in graph neural networks. It explores how the deficiency of structural information affects tail node representation learning and proposes a solution by adding pseudo-homophilic edges to enhance the graph structure. Extensive experiments demonstrate the effectiveness of SAILOR in outperforming existing methods for tail node representation learning.