DyTed: Disentangled Representation Learning for Discrete-time Dynamic Graph

TL;DR

DyTed is a novel framework for disentangled representation learning in discrete-time dynamic graphs, effectively separating stable and dynamic features, leading to improved performance and robustness across multiple tasks.

Contribution

It introduces a disentanglement-aware adversarial framework with contrastive learning for dynamic graphs, enhancing interpretability and robustness compared to existing methods.

Findings

Achieves state-of-the-art results on multiple datasets

Demonstrates robustness against noise

Effectively separates time-invariant and time-varying features

Abstract

Unsupervised representation learning for dynamic graphs has attracted a lot of research attention in recent years. Compared with static graph, the dynamic graph is a comprehensive embodiment of both the intrinsic stable characteristics of nodes and the time-related dynamic preference. However, existing methods generally mix these two types of information into a single representation space, which may lead to poor explanation, less robustness, and a limited ability when applied to different downstream tasks. To solve the above problems, in this paper, we propose a novel disenTangled representation learning framework for discrete-time Dynamic graphs, namely DyTed. We specially design a temporal-clips contrastive learning task together with a structure contrastive learning to effectively identify the time-invariant and time-varying representations respectively. To further enhance the disentanglement of these two types of representation, we propose a disentanglement-aware discriminator under an adversarial learning framework from the perspective of information theory. Extensive experiments on Tencent and five commonly used public datasets demonstrate that DyTed, as a general framework that can be applied to existing methods, achieves state-of-the-art performance on various downstream tasks, as well as be more robust against noise.