.png){kind=icon}
.png){kind=icon}
Return
Dual-Tensor Collaborative Facilitation for Multi-view Clustering
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2025.129883.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A new Weighted Arctangent Coupled tensor nuclear norm is proposed. The WAC is capable of weighted arctangent coupling for singular values of different sizes, which fully fits the true rank of the singular value decomposition of the representation tensor in the frequency domain. This is also consistent with the practical situation that there are large differences between the multi-view features of the tensor representation. • A tensor FLP-norm is designed. The norm is able to scale the columns of the projection tensor according to the projection dimension of the original data to further remove the useless redundant information in the projection tensor. • In addition, the optimal representation matrix is obtained by adaptive weighted fusion based on Laplace rank constraints for the multi-view representation tensor to ensure that it has a strong block diagonal structure.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W