LGD-GCN: Local and Global Disentangled Graph Convolutional Networks

Abstract

Disentangled Graph Convolutional Network (DisenGCN) is an encouraging framework to disentangle the latent factors arising in a real-world graph. However, it relies on disentangling information heavily from a local range (i.e., a node and its 1-hop neighbors), while the local information in many cases can be uneven and incomplete, hindering the interpretabiliy power and model performance of DisenGCN. In this paper\footnote{This paper is a lighter version of \href{https://jingweio.github.io/assets/pdf/tnnls22.pdf}{"Learning Disentangled Graph Convolutional Networks Locally and Globally"} where the results and analysis have been reworked substantially. Digital Object Identifier \url{https://doi.org/10.1109/TNNLS.2022.3195336}.}, we introduce a novel Local and Global Disentangled Graph Convolutional Network (LGD-GCN) to capture both local and global information for graph disentanglement. LGD-GCN performs a statistical mixture modeling to derive a factor-aware latent continuous space, and then constructs different structures w.r.t. different factors from the revealed space. In this way, the global factor-specific information can be efficiently and selectively encoded via a message passing along these built structures, strengthening the intra-factor consistency. We also propose a novel diversity promoting regularizer employed with the latent space modeling, to encourage inter-factor diversity. Evaluations of the proposed LGD-GCN on the synthetic and real-world datasets show a better interpretability and improved performance in node classification over the existing competitive models. Code is available at \url{https://github.com/jingweio/LGD-GCN}.