UniG-Encoder: A universal feature encoder for graph and hypergraph node classification

Minhao Zou, Zhongxue Gan, Yutong Wang, Junheng Zhang, Dongyan Sui, Chun Guan, Siyang Leng

Abstract: Highlights•UniG-Encoder is proposed towards representation learning for graphs and hypergraphs.•Heterophilic and homophilic graphs can both be addressed.•The architecture is realized via an intuitive and interpretable projection matrix.•The architecture involves minor consumption but achieves superior performance.•A variant version, UniG-Encoder II, is devised to leverage multi-hop information.