Neural Graph Generation from Graph Statistics

Kiarash Zahirnia, Yaochen Hu, Mark Coates, Oliver Schulte

Keywords: Graph Generation, Local Differential Privacy, Graph Statistics, Latent Adjacency Matrix

TL;DR: This work highlights the potential of learning deep GGMs from aggregate graph statistics as a privacy-preserving alternative to traditional approaches that rely on sensitive graph data.

Abstract:

We describe a new setting for learning a deep graph generative model (GGM) from aggregate graph statistics, rather than from the graph adjacency matrix. Matching the statistics of observed training graphs is the main approach for learning traditional GGMs (e.g, BTER, Chung-Lu, and Erdos-Renyi models). Privacy researchers have proposed learning from graph statistics as a way to protect privacy.
We develop an architecture for training a deep GGM to match statistics while preserving local differential privacy guarantees. Empirical evaluation on 8 datasets indicates that our deep GGM model generates more realistic graphs than the traditional GGMs when both are learned from graph statistics only. We also benchmark our deep GGM trained on statistics only, against state-of-the-art deep GGM models that are trained on the entire adjacency matrix. The results show that graph statistics are often sufficient to build a competitive deep GGM that generates realistic graphs while protecting local privacy.

Submission Number: 5566