SHADOWCAST: Controllable Graph Generation

Abstract

We introduce the controllable graph generation problem, formulated as controlling graph attributes during the generative process to produce desired graphs with understandable structures. Using a transparent and straightforward Markov model to guide this generative process, practitioners can shape and understand the generated graphs. We propose $\mathrm{S}\mathrm{HADOW}\mathrm{C}\mathrm{AST}$, a generative model capable of controlling graph generation while retaining the original graph's intrinsic properties. The proposed model is based on a conditional generative adversarial network. Given an observed graph and some user-specified Markov model parameters, $\mathrm{S}\mathrm{HADOW}\mathrm{C}\mathrm{AST}$ controls the conditions to generate desired graphs. Comprehensive experiments on three real-world network datasets demonstrate our model's competitive performance in the graph generation task. Furthermore, we show its effective controllability by directing $\mathrm{S}\mathrm{HADOW}\mathrm{C}\mathrm{AST}$ to generate hypothetical scenarios with different graph structures.

fix title

References in corpus (9)

• Conditional Generative Adversarial Nets
• Variational Graph Auto-Encoders
• Interaction Networks for Learning about Objects, Relations and Physics
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• MolGAN: An implicit generative model for small molecular graphs
• Learning Deep Generative Models of Graphs
• NetGAN: Generating Graphs via Random Walks
• Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders
• Can GAN Learn Topological Features of a Graph?

Cited by in corpus (1)

• Poisoning Online Learning Filters: DDoS Attacks and Countermeasures