Graph Learning

TL;DR

Graph Learning surveys the rapid expansion of learning on graphs, a non-Euclidean data domain, and provides a unifying framework across six core strands: scalability, temporal dynamics, multimodality, generative modeling, explainability, and responsibility. It catalogs state-of-the-art techniques for handling large-scale graphs, modeling time-evolving structures, fusing heterogeneous modalities, synthesizing new graphs, and enhancing interpretability and trust. The survey also discusses emerging topics such as graph foundation models, graph reinforcement learning, federated graph learning, knowledge-infused graph learning, and quantum graph learning, while highlighting challenges like privacy, fairness, robustness, and data heterogeneity. By offering taxonomy, representative methods, and future directions, the work serves as a comprehensive resource for researchers and practitioners navigating the rapidly evolving graph-learning landscape.

Abstract

Graph learning has rapidly evolved into a critical subfield of machine learning and artificial intelligence (AI). Its development began with early graph-theoretic methods, gaining significant momentum with the advent of graph neural networks (GNNs). Over the past decade, progress in scalable architectures, dynamic graph modeling, multimodal learning, generative AI, explainable AI (XAI), and responsible AI has broadened the applicability of graph learning to various challenging environments. Graph learning is significant due to its ability to model complex, non-Euclidean relationships that traditional machine learning struggles to capture, thus better supporting real-world applications ranging from drug discovery and fraud detection to recommender systems and scientific reasoning. However, challenges like scalability, generalization, heterogeneity, interpretability, and trustworthiness must be addressed to unlock its full potential. This survey provides a comprehensive introduction to graph learning, focusing on key dimensions including scalable, temporal, multimodal, generative, explainable, and responsible graph learning. We review state-of-the-art techniques for efficiently handling large-scale graphs, capturing dynamic temporal dependencies, integrating heterogeneous data modalities, generating novel graph samples, and enhancing interpretability to foster trust and transparency. We also explore ethical considerations, such as privacy and fairness, to ensure responsible deployment of graph learning models. Additionally, we identify and discuss emerging topics, highlighting recent integration of graph learning and other AI paradigms and offering insights into future directions. This survey serves as a valuable resource for researchers and practitioners seeking to navigate the rapidly evolving landscape of graph learning.

Figures (27)

• Figure 1: Examples of Euclidean Data and Graph Data. (a) Image of Euclidean data and (b) image of non-Euclidean data.
• Figure 2: Some Representative Graph Learning Methods.
• Figure 3: Comparison of Nodes vs. Edges in Large Graph Datasets.
• Figure 4: Strategy Taxonomy of Scalable Graph Learning.
• Figure 5: Data Summarization Techniques.