Out-Of-Distribution Generalization on Graphs: A Survey
Haoyang Li, Xin Wang, Ziwei Zhang, Wenwu Zhu Haoyang Li, Xin Wang, Ziwei Zhang, and Wenwu Zhu are with the Department of Computer Science and Technology in Tsinghua University, Beijing, China. Haoyang Li and Xin Wang contribute equally. E-mail:
Abstract
Graph machine learning has been extensively studied in both academia and industry. Although booming with a vast number of emerging methods and techniques, most of the literature is built on the in-distribution hypothesis, i.e., testing and training graph data are identically distributed. However, this in-distribution hypothesis can hardly be satisfied in many real-world graph scenarios where the model performance substantially degrades when there exist distribution shifts between testing and training graph data. To solve this critical problem, out-of-distribution (OOD) generalization on graphs, which goes beyond the in-distribution hypothesis, has made great progress and attracted ever-increasing attention from the research community. In this paper, we comprehensively survey OOD generalization on graphs and present a detailed review of recent advances in this area. First, we provide a formal problem definition of OOD generalization on graphs. Second, we categorize existing methods into three classes from conceptually different perspectives, i.e., data, model, and learning strategy, based on their positions in the graph machine learning pipeline, followed by detailed discussions for
中文速览
图神经网络（GNN）在现实部署中常常遭遇"训练分布与测试分布不一致"的问题，导致模型性能大幅下滑，而现有文献大多仍建立在两者同分布的假设之上。这篇综述系统梳理了图上的分布外泛化（Out-of-Distribution Generalization on Graphs）研究，将现有方法按其在机器学习流程中的位置分为三大类：以图数据增强为核心的数据层方法、以解耦表示和因果建模为核心的模型层方法、以及涵盖不变学习、对抗训练和自监督学习的训练策略层方法。文章还整理了相关理论基础与常用评测数据集，并指出了未来值得深入探索的方向。这是目前该领域首篇系统性综述，对希望将图模型应用于药物发现、金融风控、自动驾驶等高风险真实场景的研究者具有重要参考价值。
原文 arXiv:2202.07987;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2202.07987v2