Tensor Graph Convolutional Network for Dynamic Graph Representation Learning
This addresses the limitation of hybrid designs in dynamic graph models for applications involving evolving entity interactions.
The paper tackles the problem of capturing spatial-temporal continuity in dynamic graph representation learning by proposing a tensor graph convolutional network that models both spatial and temporal features simultaneously in one convolution framework. Experiments on real-world datasets show the model achieves state-of-the-art performance.
Dynamic graphs (DG) describe dynamic interactions between entities in many practical scenarios. Most existing DG representation learning models combine graph convolutional network and sequence neural network, which model spatial-temporal dependencies through two different types of neural networks. However, this hybrid design cannot well capture the spatial-temporal continuity of a DG. In this paper, we propose a tensor graph convolutional network to learn DG representations in one convolution framework based on the tensor product with the following two-fold ideas: a) representing the information of DG by tensor form; b) adopting tensor product to design a tensor graph convolutional network modeling spatial-temporal feature simultaneously. Experiments on real-world DG datasets demonstrate that our model obtains state-of-the-art performance.