所属单位：

理学院

发表刊物：

Mathematical Problems in Engineering

关键字：

中文关键字：低秩矩阵恢复，英文关键字：Low-rank matrix recovery

摘要：

Low-rank matrix recovery (LRMR) has been becoming an increasingly popular technique for analyzing data with missing entries, grosscorruptions,andoutliers.AsasignificantcomponentofLRMR,themodeloflow-rankrepresentation(LRR)seeksthelowest- rankrepresentationamongallsamples anditisrobustforrecovering subspacestructures.Thispaper attemptstosolvetheproblem ofLRRwithpartiallyobservedentries.Firstly,weconstructanonconvexminimizationbytakingthelowrankness,robustness,and incompletionintoconsideration.ThenweemploythetechniqueofaugmentedLagrangemultiplierstosolvetheproposedprogram. Finally, experimental results on synthetic and real-world datasets validate the feasibility and effectiveness of the proposed method.

备注：

SCI

合写作者：

雍龙泉

第一作者：

郑秀云,杨威,史加荣

论文类型：

期刊论文

卷号：

卷：2014

期号：

期：

页面范围：

页：

是否译文：

否

发表时间：

2014-12-01