Back to Search Start Over

NVDiff: Graph Generation through the Diffusion of Node Vectors

Authors :
Chen, Xiaohui
Li, Yukun
Zhang, Aonan
Liu, Li-Ping
Publication Year :
2022

Abstract

Learning to generate graphs is challenging as a graph is a set of pairwise connected, unordered nodes encoding complex combinatorial structures. Recently, several works have proposed graph generative models based on normalizing flows or score-based diffusion models. However, these models need to generate nodes and edges in parallel from the same process, whose dimensionality is unnecessarily high. We propose NVDiff, which takes the VGAE structure and uses a score-based generative model (SGM) as a flexible prior to sample node vectors. By modeling only node vectors in the latent space, NVDiff significantly reduces the dimension of the diffusion process and thus improves sampling speed. Built on the NVDiff framework, we introduce an attention-based score network capable of capturing both local and global contexts of graphs. Experiments indicate that NVDiff significantly reduces computations and can model much larger graphs than competing methods. At the same time, it achieves superior or competitive performances over various datasets compared to previous methods.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2211.10794
Document Type :
Working Paper