DyTed: Disentangled Representation Learning for Discrete-time Dynamic Graph
This addresses the need for more interpretable and robust dynamic graph representations for applications like social network analysis, though it is incremental as it builds on existing contrastive learning methods.
The paper tackles the problem of mixing time-invariant and time-varying information in dynamic graph representation learning by proposing DyTed, a disentangled framework that achieves state-of-the-art performance on various downstream tasks and shows improved robustness against noise.
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.