GraphMaker: Can Diffusion Models Generate Large Attributed Graphs?

TL;DR

GraphMaker is a novel diffusion model that effectively generates large attributed graphs, capturing complex attribute-structure relationships and enabling data sharing for graph machine learning tasks.

Contribution

The paper introduces GraphMaker, a diffusion-based approach tailored for large attributed graph generation, addressing scalability and complex attribute-structure correlations.

Findings

Synthetic graphs enable competitive ML models without access to original data

Asynchronous generation captures attribute-structure correlations more effectively

Edge mini-batching improves scalability

Abstract

Large-scale graphs with node attributes are increasingly common in various real-world applications. Creating synthetic, attribute-rich graphs that mirror real-world examples is crucial, especially for sharing graph data for analysis and developing learning models when original data is restricted to be shared. Traditional graph generation methods are limited in their capacity to handle these complex structures. Recent advances in diffusion models have shown potential in generating graph structures without attributes and smaller molecular graphs. However, these models face challenges in generating large attributed graphs due to the complex attribute-structure correlations and the large size of these graphs. This paper introduces a novel diffusion model, GraphMaker, specifically designed for generating large attributed graphs. We explore various combinations of node attribute and graph structure generation processes, finding that an asynchronous approach more effectively captures the intricate attribute-structure correlations. We also address scalability issues through edge mini-batching generation. To demonstrate the practicality of our approach in graph data dissemination, we introduce a new evaluation pipeline. The evaluation demonstrates that synthetic graphs generated by GraphMaker can be used to develop competitive graph machine learning models for the tasks defined over the original graphs without actually accessing these graphs, while many leading graph generation methods fall short in this evaluation.