Chapter11: Graph Neural Networks: Graph Generation

Renjie Liao, University of Toronto, rjliao@cs.toronto.edu

Description PDF

Abstract

In this chapter, we first review a few classic probabilistic models for graph generation including the Erd˝os–R´enyi model and the stochastic block model. Then we introduce several representative modern graph generative models that leverage deep learning techniques like graph neural networks, variational auto-encoders, deep auto-regressive models, and generative adversarial networks. At last, we conclude the chapter with a discussion on potential future directions.

Contents

  • Introduction
  • Classic Graph Generative Models
    • Erdos-Renyi Model
    • Stochastic Block Model
  • Deep Graph Generative Models
    • Representing Graphs
    • Variational Auto-Encoder Methods
    • Deep Autoregressive Methods
    • Generative Adversarial Methods
  • Summary

Citation

@incollection{GNNBook-ch11-liao,
author = "Liao, Renjie",
editor = "Wu, Lingfei and Cui, Peng and Pei, Jian and Zhao, Liang",
title = "Graph Neural Networks: Graph Generation",
booktitle = "Graph Neural Networks: Foundations, Frontiers, and Applications",
year = "2022",
publisher = "Springer Singapore",
address = "Singapore",
pages = "225--250",
}

R. Liao, “Graph neural networks: Graph generation,” in Graph Neural Networks: Foundations, Frontiers, and Applications, L. Wu, P. Cui, J. Pei, and L. Zhao, Eds. Springer Singapore, 2022, pp. 225–250.