Efficient Graph Generation with Graph Recurrent Attention Networks

TL;DR

This paper introduces Graph Recurrent Attention Networks (GRANs), a novel deep generative model for graphs that improves efficiency and quality by generating graphs in blocks with attention mechanisms, capable of scaling to large graphs.

Contribution

The paper presents GRAN, a scalable, efficient graph generative model that captures complex dependencies using attention and block-wise generation, surpassing previous models in speed and size.

Findings

Achieves state-of-the-art efficiency and quality on benchmarks.

Capable of generating large graphs up to 5,000 nodes.

First deep model to scale to such large graph sizes.

Abstract

We propose a new family of efficient and expressive deep generative models of graphs, called Graph Recurrent Attention Networks (GRANs). Our model generates graphs one block of nodes and associated edges at a time. The block size and sampling stride allow us to trade off sample quality for efficiency. Compared to previous RNN-based graph generative models, our framework better captures the auto-regressive conditioning between the already-generated and to-be-generated parts of the graph using Graph Neural Networks (GNNs) with attention. This not only reduces the dependency on node ordering but also bypasses the long-term bottleneck caused by the sequential nature of RNNs. Moreover, we parameterize the output distribution per block using a mixture of Bernoulli, which captures the correlations among generated edges within the block. Finally, we propose to handle node orderings in generation by marginalizing over a family of canonical orderings. On standard benchmarks, we achieve state-of-the-art time efficiency and sample quality compared to previous models. Additionally, we show our model is capable of generating large graphs of up to 5K nodes with good quality. To the best of our knowledge, GRAN is the first deep graph generative model that can scale to this size. Our code is released at: https://github.com/lrjconan/GRAN.