Simple Multigraph Convolution Networks

Existing multigraph convolution methods either ignore the crossview interaction among multiple graphs, or induce extremely high computational cost due to standard cross-view polynomial operators. To alleviate this problem, this paper proposes a Simple Multi-Graph Convolution Networks (SMGCN) which first extracts consistent cross-view topology from multigraphs including edge-level and subgraph-level topology, then performs polynomial expansion based on raw multigraphs and consistent topologies. In theory, SMGCN utilizes the consistent topologies in polynomial expansion rather than standard cross-view polynomial expansion, which performs credible cross-view spatial message-passing, follows the spectral convolution paradigm, and effectively reduces the complexity of standard polynomial expansion. In the simulations, experimental results demonstrate that SMGCN achieves state-of-the-art performance on ACM and DBLP multigraph benchmark datasets. Our Figure 1: Overview of the proposed SMGCN.