NetGAN: Generating Graphs via Random Walks

TL;DR

NetGAN is an innovative generative model that learns to produce realistic graphs by modeling the distribution of biased random walks, capturing complex network patterns without explicit specifications.

Contribution

It introduces the first implicit graph generative model based on random walks trained with Wasserstein GAN, demonstrating strong pattern replication and generalization capabilities.

Findings

Produces graphs with known network patterns

Achieves competitive link prediction performance

First to combine pattern generation and generalization in graph models

Abstract

We propose NetGAN - the first implicit generative model for graphs able to mimic real-world networks. We pose the problem of graph generation as learning the distribution of biased random walks over the input graph. The proposed model is based on a stochastic neural network that generates discrete output samples and is trained using the Wasserstein GAN objective. NetGAN is able to produce graphs that exhibit well-known network patterns without explicitly specifying them in the model definition. At the same time, our model exhibits strong generalization properties, as highlighted by its competitive link prediction performance, despite not being trained specifically for this task. Being the first approach to combine both of these desirable properties, NetGAN opens exciting avenues for further research.