Decoupled Graph Neural Networks for Large Dynamic Graphs

TL;DR

This paper introduces a decoupled graph neural network that efficiently handles large-scale continuous and discrete dynamic graphs, achieving state-of-the-art results and high scalability across diverse real-world datasets.

Contribution

The paper presents a unified dynamic propagation method that supports both continuous and discrete dynamic graphs, enabling scalable and flexible graph neural network applications.

Findings

Achieves state-of-the-art performance on seven real-world datasets.

Handles graphs with over a billion edges and 100 million nodes.

Supports efficient computation by decoupling structure-related tasks from prediction.

Abstract

Real-world graphs, such as social networks, financial transactions, and recommendation systems, often demonstrate dynamic behavior. This phenomenon, known as graph stream, involves the dynamic changes of nodes and the emergence and disappearance of edges. To effectively capture both the structural and temporal aspects of these dynamic graphs, dynamic graph neural networks have been developed. However, existing methods are usually tailored to process either continuous-time or discrete-time dynamic graphs, and cannot be generalized from one to the other. In this paper, we propose a decoupled graph neural network for large dynamic graphs, including a unified dynamic propagation that supports efficient computation for both continuous and discrete dynamic graphs. Since graph structure-related computations are only performed during the propagation process, the prediction process for the downstream task can be trained separately without expensive graph computations, and therefore any sequence model can be plugged-in and used. As a result, our algorithm achieves exceptional scalability and expressiveness. We evaluate our algorithm on seven real-world datasets of both continuous-time and discrete-time dynamic graphs. The experimental results demonstrate that our algorithm achieves state-of-the-art performance in both kinds of dynamic graphs. Most notably, the scalability of our algorithm is well illustrated by its successful application to large graphs with up to over a billion temporal edges and over a hundred million nodes.