Dynamic Graph Representation Learning via Self-Attention Networks

TL;DR

This paper introduces DySAT, a self-attention based neural network that learns dynamic graph representations capturing both structure and temporal evolution, outperforming existing methods in link prediction tasks.

Contribution

The paper proposes DySAT, a novel architecture that jointly models structural and temporal information in dynamic graphs using self-attention mechanisms.

Findings

DySAT significantly outperforms state-of-the-art baselines in link prediction.

DySAT effectively captures both structural and temporal patterns.

Experimental results validate the superiority of DySAT on communication and bipartite rating networks.

Abstract

Learning latent representations of nodes in graphs is an important and ubiquitous task with widespread applications such as link prediction, node classification, and graph visualization. Previous methods on graph representation learning mainly focus on static graphs, however, many real-world graphs are dynamic and evolve over time. In this paper, we present Dynamic Self-Attention Network (DySAT), a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns. Specifically, DySAT computes node representations by jointly employing self-attention layers along two dimensions: structural neighborhood and temporal dynamics. We conduct link prediction experiments on two classes of graphs: communication networks and bipartite rating networks. Our experimental results show that DySAT has a significant performance gain over several different state-of-the-art graph embedding baselines.