Time-aware Dynamic Graph Embedding for Asynchronous Structural Evolution

TL;DR

This paper introduces a novel time-aware Transformer model for dynamic graph embedding that captures asynchronous structural evolution, outperforming existing methods in accuracy and scalability.

Contribution

It formulates dynamic graphs as temporal edge sequences with joining times and durations, and develops a Transformer-based approach to embed these asynchronous dynamics.

Findings

Outperforms state-of-the-art in multiple graph mining tasks

Efficient and scalable for large-scale dynamic graphs

Effectively captures asynchronous structural evolution

Abstract

Dynamic graphs refer to graphs whose structure dynamically changes over time. Despite the benefits of learning vertex representations (i.e., embeddings) for dynamic graphs, existing works merely view a dynamic graph as a sequence of changes within the vertex connections, neglecting the crucial asynchronous nature of such dynamics where the evolution of each local structure starts at different times and lasts for various durations. To maintain asynchronous structural evolutions within the graph, we innovatively formulate dynamic graphs as temporal edge sequences associated with joining time of vertices (ToV) and timespan of edges (ToE). Then, a time-aware Transformer is proposed to embed vertices' dynamic connections and ToEs into the learned vertex representations. Meanwhile, we treat each edge sequence as a whole and embed its ToV of the first vertex to further encode the time-sensitive information. Extensive evaluations on several datasets show that our approach outperforms the state-of-the-art in a wide range of graph mining tasks. At the same time, it is very efficient and scalable for embedding large-scale dynamic graphs.