Deep Graph Translation

Xiaojie Guo, Lingfei Wu, Liang Zhao

Keywords: Graph translation, graph generation, deep neural network

TL;DR: a novel Graph-Translation-Generative-Adversarial-Networks (GT-GAN) that transforms the input graphs into their target output graphs

Abstract: Deep graph generation models have achieved great successes recently, among which, however, are typically unconditioned generative models that have no control over the target graphs are given an input graph. In this paper, we propose a novel Graph-Translation-Generative-Adversarial-Networks (GT-GAN) that transforms the input graphs into their target output graphs. GT-GAN consists of a graph translator equipped with innovative graph convolution and deconvolution layers to learn the translation mapping considering both global and local features, and a new conditional graph discriminator to classify target graphs by conditioning on input graphs. Extensive experiments on multiple synthetic and real-world datasets demonstrate that our proposed GT-GAN significantly outperforms other baseline methods in terms of both effectiveness and scalability. For instance, GT-GAN achieves at least 10X and 15X faster runtimes than GraphRNN and RandomVAE, respectively, when the size of the graph is around 50.

Code: https://github.com/anonymous1025/Deep-Graph-Translation-

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:1805.09980/code)