Graph Condensation: A Survey

2024-01-22

arXiv.org

AI Summary

The explosive growth of graph data has imposed severe storage, transmission, and computational bottlenecks on Graph Neural Network (GNN) training, necessitating compact yet high-fidelity condensed graphs for efficient learning. This paper presents a systematic survey of Graph Condensation (GC) techniques. We propose, for the first time, a five-dimensional evaluation framework—encompassing effectiveness, generalizability, efficiency, fairness, and robustness—and unify GC methodologies into two core components: optimization strategies (e.g., gradient matching, feature distillation, meta-learning, adversarial generation) and condensed graph generation (e.g., spectral analysis, topological modeling, differentiable graph synthesis). Through empirical benchmarking of state-of-the-art methods, we analyze open-source ecosystems and cross-domain applications, identifying critical challenges—including limited scalability and lack of support for dynamic graphs—and outline promising future research directions.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Constraint Satisfaction and Optimization: Other Foundations of Constraint Satisfaction

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 evaluation methodologies and metrics

Abstract

The rapid growth of graph data poses significant challenges in storage, transmission, and particularly the training of graph neural networks (GNNs). To address these challenges, graph condensation (GC) has emerged as an innovative solution. GC focuses on synthesizing a compact yet highly representative graph, enabling GNNs trained on it to achieve performance comparable to those trained on the original large graph. The notable efficacy of GC and its broad prospects have garnered significant attention and spurred extensive research. This survey paper provides an up-to-date and systematic overview of GC, organizing existing research into five categories aligned with critical GC evaluation criteria: effectiveness, generalization, efficiency, fairness, and robustness. To facilitate an in-depth and comprehensive understanding of GC, this paper examines various methods under each category and thoroughly discusses two essential components within GC: optimization strategies and condensed graph generation. We also empirically compare and analyze representative GC methods with diverse optimization strategies based on the five proposed GC evaluation criteria. Finally, we explore the applications of GC in various fields, outline the related open-source libraries, and highlight the present challenges and novel insights, with the aim of promoting advancements in future research. The related resources can be found at https://github.com/XYGaoG/Graph-Condensation-Papers.

Problem

• Graph Data
• Graph Neural Networks
• Representation Learning

Innovation

• Graph Condensation
• Systematic Evaluation
• Multi-Domain Applications