Efficient Graph Generation with Graph Recurrent Attention Networks

Renjie Liao, Yujia Li, Yang Song, Shenlong Wang, Will Hamilton, David Duvenaud, Raquel Urtasun, Richard Zemel

Abstract: We propose a new family of scalable and expressive generative models of graphs, called Graph Recurrent Attention Networks (GRANs). These models (1) better capture the auto-regressive conditioning between the already generated and to be generated parts of the graph using Graph Neural Networks, and (2) improve model quality by integrating over an adaptive mixture of node orderings. Our model generates graphs one block of nodes and associated edges at a time, independently conditioned on the context. The block size allows us to favorably trade-off model quality for efficiency. On standard benchmarks, our model generates graphs comparable in quality with the previous state-of-the-art, and is at least an order of magnitude faster.

Code Link: https://github.com/lrjconan/GRAN

CMT Num: 2394