HEAT: Hyperbolic Embedding of Attributed Networks

Finding a low dimensional representation of hierarchical, structured data described by a network remains a challenging problem in the machine learning community. An emerging approach is embedding networks into hyperbolic space because it can naturally represent a network's hierarchical structure. However, existing hyperbolic embedding approaches cannot deal with attributed networks, in which nodes are annotated with additional attributes. These attributes might provide additional proximity information to constrain the representations of the nodes, which is important to learn high quality hyperbolic embeddings. To overcome this gap we propose HEAT (Hyperbolic Embedding of Attributed Networks). HEAT first extracts training samples from the original graph capturing both topological and attribute similarity and then learns a hyperboloid embedding using full Riemannian Stochastic Gradient Descent. We show that HEAT can outperform other network embedding algorithms on several downstream tasks. As a general embedding method, HEAT opens the door to hyperbolic manifold learning on a wide range of attributed and unattributed networks.