Year: 2023
Type: preprint
Abstract: Diffusion-based graph generative models have recently obtained promising results for graph generation. However, existing diffusion-based graph generative models are mostly one-shot generative models that apply Gaussian diffusion in the dequantized adjacency matrix space. Such a strategy can suffer from difficulty in model training, slow sampling speed, and incapability of incorporating constraints. We propose an \emph{autoregressive diffusion} model for graph generation. Unlike existing methods, we define a node-absorbing diffusion process that operates directly in the discrete graph space. For forward diffusion, we design a \emph{diffusion ordering network}, which learns a data-dependent node absorbing ordering from graph topology. For reverse generation, we design a \emph{denoising network} that uses the reverse node ordering to efficiently reconstruct the graph by predicting the node type of the new node and its edges with previously denoised nodes at a time. Based on the permutation invariance of graph, we show that the two networks can be jointly trained by optimizing a simple lower bound of data likelihood. Our experiments on six diverse generic graph datasets and two molecule datasets show that our model achieves better or comparable generation performance with previous state-of-the-art, and meanwhile enjoys fast generation speed.
Source: arXiv (Cornell University)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: III: Small: Go Beyond Short-term Dependency and Homogeneity: A General-Purpose Transformer Recipe for Multi-Domain Heterogeneous Sequential Data Analysis, PIPP Phase I: BEHIVE - BEHavioral Interaction and Viral Evolution for Pandemic Prevention and Prediction, RAPID: Collaborative Research: Using Phylodynamics and Line Lists for Adaptive COVID-19 Monitoring, III: Medium: Collaborative Research: Detecting and Controlling Network-based Spread of Hospital Acquired Infections, CAREER: Accelerating Spatial Network Design: An Uncertainty-Driven Predict-and-Optimize Learning Framework, CAREER: Bridging the Data-Model Gap -- Leveraging Surveillance for Propagation Mining over Networks, III: Medium: Collaborative Research: Principled Uncertainty Quantification in Deep Learning Models for Time Series Analysis, G6671297155, Expeditions: Collaborative Research: Global Pervasive Computational Epidemiology
OpenAlex ID: W4384807693
