no code implementations • 5 Mar 2024 • Yanbei Liu, Yu Zhao, Xiao Wang, Lei Geng, Zhitao Xiao
By an experimental analysis, we discover the semantic information of an augmented graph structure may be not consistent as original graph structure, and whether two augmented graphs are positive or negative pairs is highly related with the multi-scale structures.
no code implementations • 26 Nov 2019 • Yanbei Liu, Xiao Wang, Shu Wu, Zhitao Xiao
In this paper, we propose a novel Independence Promoted Graph Disentangled Networks (IPGDN) to learn disentangled node representation while enhancing the independence among node representations.