Deep graph clustering with multi-level subspace fusion
作者：
Highlights：
• Graph Convolutional Network is bothered by over-smoothness problem.
• Over-smoothness may decrease the distinction between dissimilar nodes.
• Self-expressive learning makes robust representations.
• The multi-level self-expressive learning captures multi-scaled information.
• The fusing of structure information from different scales increases distinction between nodes.
摘要
•Graph Convolutional Network is bothered by over-smoothness problem.•Over-smoothness may decrease the distinction between dissimilar nodes.•Self-expressive learning makes robust representations.•The multi-level self-expressive learning captures multi-scaled information.•The fusing of structure information from different scales increases distinction between nodes.
论文关键词：Graph clustering,Subspace,Self-expressive learning,Fusion
论文评审过程：Received 9 February 2022, Revised 22 August 2022, Accepted 25 September 2022, Available online 27 September 2022, Version of Record 4 October 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.109077
