|Table of Contents|

Double-layer Multi-view Subspace Clustering with Joint Low-rankRepresentation and Sparsity Constraint(PDF)

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2022年01期
Page:
59-67
Research Field:
机器学习
Publishing date:

Info

Title:
Double-layer Multi-view Subspace Clustering with Joint Low-rankRepresentation and Sparsity Constraint
Author(s):
Zhang Fukang1; Yin Hongwei1; Cheng Xinmin1; Du Wenhui2; Xu Huangzhen2
(1.School of Information Engineering,Huzhou University,Huzhou 313000,China)(2.Huzhou Special Equipment Inspection Center,Huzhou 313000,China)
Keywords:
multi-view clustering; multi-view subspace clustering; low-rank representation; sparse constraint
PACS:
TP391
DOI:
10.3969/j.issn.1672-1292.2022.01.009
Abstract:
Multi-view clustering is the use of multiple different description of the data set to the same class of similar data together as far as possible. Among them,multi-view subspace clustering is a kind of clustering method to deal with high-dimensional data. The current approach is to solve the multi-view subspace clustering problem by constructing the affinity matrix on each view separately. The importance of the combination of low-rank representation and sparse constraints in constructing the adjacency matrix is not considered. In order to solve this problem,this paper proposes a double-layer multi-view subspace clustering with joint low-rank representation and sparsity constraint,so that it can describe the data itself more comprehensively,so as to achieve more effective clustering. For the low-rank representation and sparsity constraint optimization problems related to each view,we use ADMM method to solve them. Finally,experiments are conducted on multiple data sets,and it is found that the clustering performance of the proposed algorithm is better than the existing multi-view subspace clustering algorithm. It is proved that the combination of low rank representation and sparse constraint can improve the accuracy of clustering.

References:

[1] 夏菁,丁世飞. 基于低秩稀疏约束的自权重多视角子空间聚类[J]. 南京大学学报(自然科学),2020,56(6):862-869.
[2]唐启凡,张玉龙,何士豪,等. 自适应多视角子空间聚类[J]. 西安交通大学学报,2021,55(5):102-112.
[3]李凯鑫. 基于低秩的子空间聚类算法[D]. 广州:华南理工大学,2018.
[4]侯成浩. 面向多视数据的稀疏子空间聚类方法研究[D]. 北京:北京工业大学,2017.
[5]LIU G C,LIN Z C,YU Y. Robust subspace segmentation by low-rank representation[C]//Proceedings of the 27th International Conference on Machine Learning. Haifa,Israel:Omnipress,2010:663-670.
[6]江楠,张国明,王俊淑,等. 融合低秩和形态学的高光谱影像特征提取[J]. 南京师范大学学报(工程技术版),2020,20(2):52-58.
[7]ELHAMIFAR E,VIDAL R. Sparse subspace clustering[C]//Proceedings of the 2009 IEEE Conference on Computer Vision & Pattern Recognition. Miami,USA:IEEE,2009:2790-2797.
[8]XUE Z,DU J P,DU D W,et al. Deep low-rank subspace ensemble for multi-view clustering[J]. Information Sciences,2019,482:210-227.
[9]BRBIAC’1 M,KOPRIVA I. Multi-view low-rank sparse subspace clustering[J]. Pattern Recognition,2018,73:247-258.
[10]DING C H Q,LI T,JORDAN M I. Convex and semi-nonnegative matrix factorizations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(1):45-55.
[11]BOYD S,PARIKH N,CHU E,et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trends in Machine Learning,2011,3(1):1-122.
[12]CAI J F,CANDèS E J,SHEN Z W. A singular value thresholding algorithm for matrix completion[J]. SIAM Journal on Optimization,2010,20(4):1956-1982.
[13]DAUBECHIES I,DEFRISE M,DE MOL C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint[J]. Communications on Pure and Applied Mathematics,2004,57(11):1413-1457.
[14]YU S X,SHI J B. Multiclass spectral clustering[C]//Proceedings of the 9th IEEE International Conference on Computer Vision. Nice,France:IEEE,2003.
[15]ZHAN K,NIE F P,WANG J,et al. Multiview consensus graph clustering[J]. IEEE Transactions on Image Processing,2019,28(3):1261-1270.
[16]ZHAN K,ZHANG C Q,GUAN J P,et al. Graph learning for multiview clustering[J]. IEEE Transactions on Cybernetics,2018,48(10):2887-2895.
[17]XIA R K,PAN Y,DU L,et al. Robust multi-view spectral clustering via low-rank and sparse decomposition[C]//Proceedings of the 28th AAAI Conference on Artificial Intelligence. Quebec,Canada:AAAI Press,2014:2149-2155.
[18]ZONG L L,ZHANG X C,LIU X Y,et al. Weighted multi-view spectral clustering based on spectral perturbation[C]//Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence. New Orleans,USA:AAAI Press,2018.

Memo

Memo:
-
Last Update: 2022-03-15