Recent Advances in Reliable Deep Graph Learning: Inherent Noise, Distribution Shift, and Adversarial Attack

TL;DR

This survey reviews recent progress in making deep graph learning more reliable by addressing inherent noise, distribution shifts, and adversarial attacks, emphasizing broader reliability issues beyond just adversarial defenses.

Contribution

It provides a comprehensive overview of recent methods tackling reliability challenges in DGL, including noise and distribution shift, which are less covered in prior surveys.

Findings

Summarizes recent techniques for noise robustness in DGL

Highlights methods addressing distribution shift in graph learning

Discusses the interplay between different reliability threats

Abstract

Deep graph learning (DGL) has achieved remarkable progress in both business and scientific areas ranging from finance and e-commerce to drug and advanced material discovery. Despite the progress, applying DGL to real-world applications faces a series of reliability threats including inherent noise, distribution shift, and adversarial attacks. This survey aims to provide a comprehensive review of recent advances for improving the reliability of DGL algorithms against the above threats. In contrast to prior related surveys which mainly focus on adversarial attacks and defense, our survey covers more reliability-related aspects of DGL, i.e., inherent noise and distribution shift. Additionally, we discuss the relationships among above aspects and highlight some important issues to be explored in future research.