HetCAN: A Heterogeneous Graph Cascade Attention Network with Dual-Level Awareness
Zeyuan Zhao, Qingqing Ge, Anfeng Cheng, Yiding Liu, Xiang Li, Shuaiqiang Wang
TL;DR
HetCAN tackles representation learning on heterogeneous information networks by jointly modeling graph heterogeneity and feature-level high-order information. It introduces a cascade architecture with a type-aware encoder to preserve node-/edge-type signals and a dimension-aware transformer-style encoder to capture latent feature interactions, with outputs concatenated per cascade block. Across node classification and link prediction tasks on multiple datasets, HetCAN achieves state-of-the-art or competitive results while maintaining efficiency and robustness, particularly on large-scale and highly heterogeneous graphs. The work demonstrates that integrating node-type signals with feature-level interactions in a unified metapath-free framework yields expressive embeddings and practical advantages over existing metapath-based and metapath-free HGNNs.
Abstract
Heterogeneous graph neural networks(HGNNs) have recently shown impressive capability in modeling heterogeneous graphs that are ubiquitous in real-world applications. Most existing methods for heterogeneous graphs mainly learn node embeddings by stacking multiple convolutional or attentional layers, which can be considered as capturing the high-order information from node-level aspect. However, different types of nodes in heterogeneous graphs have diverse features, it is also necessary to capture interactions among node features, namely the high-order information from feature-level aspect. In addition, most methods first align node features by mapping them into one same low-dimensional space, while they may lose some type information of nodes in this way. To address these problems, in this paper, we propose a novel Heterogeneous graph Cascade Attention Network (HetCAN) composed of multiple cascade blocks. Each cascade block includes two components, the type-aware encoder and the dimension-aware encoder. Specifically, the type-aware encoder compensates for the loss of node type information and aims to make full use of graph heterogeneity. The dimension-aware encoder is able to learn the feature-level high-order information by capturing the interactions among node features. With the assistance of these components, HetCAN can comprehensively encode information of node features, graph heterogeneity and graph structure in node embeddings. Extensive experiments demonstrate the superiority of HetCAN over advanced competitors and also exhibit its efficiency and robustness.