Leveraging Graph Diffusion Models for Network Refinement Tasks
This work addresses network refinement for applications dealing with incomplete or corrupted graph data, representing an incremental advancement by adapting diffusion models to graph-specific tasks.
The paper tackles the problem of refining noisy and incomplete real-world networks by proposing SGDM, a graph generative framework based on subgraph diffusion, which effectively supports tasks like denoising extraneous subgraphs, expanding existing subgraphs, and performing style transfer, as demonstrated through extensive empirical analysis and novel metrics.
Most real-world networks are noisy and incomplete samples from an unknown target distribution. Refining them by correcting corruptions or inferring unobserved regions typically improves downstream performance. Inspired by the impressive generative capabilities that have been used to correct corruptions in images, and the similarities between "in-painting" and filling in missing nodes and edges conditioned on the observed graph, we propose a novel graph generative framework, SGDM, which is based on subgraph diffusion. Our framework not only improves the scalability and fidelity of graph diffusion models, but also leverages the reverse process to perform novel, conditional generation tasks. In particular, through extensive empirical analysis and a set of novel metrics, we demonstrate that our proposed model effectively supports the following refinement tasks for partially observable networks: T1: denoising extraneous subgraphs, T2: expanding existing subgraphs and T3: performing "style" transfer by regenerating a particular subgraph to match the characteristics of a different node or subgraph.