SAGES: Scalable Attributed Graph Embedding With Sampling for Unsupervised Learning

Unsupervised graph embedding method generates node embeddings to preserve structural and content features in a graph without human labeling burden. However, most unsupervised graph representation learning methods suffer issues like poor scalability or limited utilization of content/structural relationships, especially on attributed graphs. In this paper, we propose SAGES, a graph sampling based autoencoder framework, which can promote both the performance and scalability of unsupervised learning on attributed graphs. Specifically, we propose a graph sampler that considers both the node connections and node attributes, thus nodes having a high influence on each other will be sampled in the same subgraph. After that, an unbiased Graph Autoencoder (GAE) with structure-level, content-level, and community-level reconstruction loss is built on the properly-sampled subgraphs in each epoch. The time and space complexity analysis is carried out to show the scalability of SAGES. We conducted experiments on three medium-size attributed graphs and three large attributed graphs. Experimental results illustrate that SAGES achieves the competitive performance in unsupervised attributed graph learning on a variety of node classification benchmarks and node clustering benchmarks.