Efficient Dynamic Attributed Graph Generation
Fan Li, Xiaoyang Wang, Dawei Cheng, Cong Chen, Ying Zhang, Xuemin Lin
TL;DR
VRDAG tackles the challenge of generating dynamic attributed graphs by introducing a variational recurrent framework that jointly models evolving topology and node attributes. It combines a bi-flow graph encoder, a GRU-based recurrence updater, and a MixBernoulli-based structure decoder with an attention-based attribute decoder, enabling one-shot generation of each snapshot without heavy path sampling. Empirical results on six real-world datasets show VRDAG achieves state-of-the-art or competitive structure and attribute quality while delivering orders-of-magnitude speedups over path-based dynamic generators. The work demonstrates practical impact for data augmentation, benchmarking, and privacy-preserving graph synthesis in dynamic domains.
Abstract
Data generation is a fundamental research problem in data management due to its diverse use cases, ranging from testing database engines to data-specific applications. However, real-world entities often involve complex interactions that cannot be effectively modeled by traditional tabular data. Therefore, graph data generation has attracted increasing attention recently. Although various graph generators have been proposed in the literature, there are three limitations: i) They cannot capture the co-evolution pattern of graph structure and node attributes. ii) Few of them consider edge direction, leading to substantial information loss. iii) Current state-of-the-art dynamic graph generators are based on the temporal random walk, making the simulation process time-consuming. To fill the research gap, we introduce VRDAG, a novel variational recurrent framework for efficient dynamic attributed graph generation. Specifically, we design a bidirectional message-passing mechanism to encode both directed structural knowledge and attribute information of a snapshot. Then, the temporal dependency in the graph sequence is captured by a recurrence state updater, generating embeddings that can preserve the evolution pattern of early graphs. Based on the hidden node embeddings, a conditional variational Bayesian method is developed to sample latent random variables at the neighboring timestep for new snapshot generation. The proposed generation paradigm avoids the time-consuming path sampling and merging process in existing random walk-based methods, significantly reducing the synthesis time. Finally, comprehensive experiments on real-world datasets are conducted to demonstrate the effectiveness and efficiency of the proposed model.