A Survey of Deep Graph Learning under Distribution Shifts: from Graph Out-of-Distribution Generalization to Adaptation

2024-10-25

arXiv.org

AI Summary

Graph machine learning often suffers from distributional shifts (out-of-distribution, OOD) between training and test data in real-world scenarios, leading to significant performance degradation. To address this, this work systematically investigates the robustness of graph deep learning under distribution shifts and proposes the first unified taxonomy comprising three task paradigms: graph OOD generalization, training-time adaptation, and test-time adaptation. Methodologically, it categorizes approaches into “model-centric” (e.g., invariant learning, causal reasoning) and “data-centric” (e.g., graph augmentation, reweighting, subgraph reconstruction) strategies. Contributions include formal modeling of covariate and concept drift on graphs, comprehensive curation of benchmark datasets and evaluation protocols, identification of key challenges—particularly structural-attribute coupling shifts—and maintenance of an open-source, dynamically updated reading list. Collectively, this work establishes foundational principles and directions for scalable, theoretically grounded OOD adaptation in graph learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Game Theory and Economic Paradigms: Adversarial Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Distribution shifts on graphs -- the discrepancies in data distribution between training and employing a graph machine learning model -- are ubiquitous and often unavoidable in real-world scenarios. These shifts may severely deteriorate model performance, posing significant challenges for reliable graph machine learning. Consequently, there has been a surge in research on graph machine learning under distribution shifts, aiming to train models to achieve satisfactory performance on out-of-distribution (OOD) test data. In our survey, we provide an up-to-date and forward-looking review of deep graph learning under distribution shifts. Specifically, we cover three primary scenarios: graph OOD generalization, training-time graph OOD adaptation, and test-time graph OOD adaptation. We begin by formally formulating the problems and discussing various types of distribution shifts that can affect graph learning, such as covariate shifts and concept shifts. To provide a better understanding of the literature, we introduce a systematic taxonomy that classifies existing methods into model-centric and data-centric approaches, investigating the techniques used in each category. We also summarize commonly used datasets in this research area to facilitate further investigation. Finally, we point out promising research directions and the corresponding challenges to encourage further study in this vital domain. We also provide a continuously updated reading list at https://github.com/kaize0409/Awesome-Graph-OOD.

Problem

• Addressing performance degradation in graph learning due to distribution shifts
• Surveying deep graph learning methods for OOD generalization and adaptation
• Classifying approaches to handle covariate and concept shifts in graph data

Innovation

• Deep graph learning under distribution shifts
• Model-centric and data-centric approaches
• Graph OOD generalization and adaptation