Graph Data Augmentation via Contrastive Generator Inversion (exttt{DCBA})

arXiv:2610.05653v1 Announce Type: new Abstract: Graphs provide a natural representation of many complex systems, ranging from social platforms to ecosystems. However, the development of graph-based machine learning methods is often constrained by the limited availability of large and diverse graph datasets. In this paper, we introduce exttt{DCBA}, a model-based approach to graph data augmentation that infers the configuration of a synthetic graph generator from an observed network. We instantiate the proposed framework using the exttt{ABCD} generator, which produces scale-free networks with community structure. Our model learns a joint representation of graphs and generator parametrisations using a multi-positive contrastive objective with soft negative weighting. The learned representation enables the prediction of an exttt{ABCD} configuration whose stochastic realisations preserve the macrostructural properties encoded by the generator. Experiments show that exttt{DCBA} recovers generator parameters more accurately and robustly than an algorithmic inverse-modelling baseline. Its downstream utility is further demonstrated in community detection, where inferred configurations used to fine-tune exttt{PRoCD} improve AMI on average by 161% on synthetic and 273% on real-world networks.

Related

• Learning the Latent Structure: A Feature-Centric Approach to Graph Data Augmentation
• Deep Neural Sheaf Diffusion
• Network Denoising Revisited: A Ricci-Flow-Inspired Graph Diffusion Method

Source: arXiv cs.LG | 2026-10-06