CORC  > 上海财经大学  > 上海财经大学
RECOVERING LOW CP/TUCKER RANKED TENSORS, WITH APPLICATIONS IN TENSOR COMPLETION
Gao, Xiang1; Jiang, Bo2; Tao, Shaozhe3
刊名PACIFIC JOURNAL OF OPTIMIZATION
2015-04
卷号11期号:2页码:385-402
关键词tensor low rank group sparsity tensor completion
ISSN号1348-9151
英文摘要In this paper, we propose a method to find tensor decompositions with low CP rank. In our approach, no prior knowledge is assumed regarding the rank of the tensor to be recovered. The key underlying idea is to take advantage of group sparsity structure. The L-1,L-2-norm regularization and the L-1,L-infinity-norm regularization are proposed to achieve the group sparse effect. Then the problem reduce to the multi-block optimization, which can be efficiently solved by the so-called Block Coordinate Decent (BCD) method. We also show that our approach can be used to solve the tensor low CP/Tucker rank completion problem. Simulation results demonstrate that our new method performs well numerically, especially when the tensor to be recovered indeed has a low rank structure.
WOS研究方向Operations Research & Management Science ; Mathematics
语种英语
出版者YOKOHAMA PUBL
WOS记录号WOS:000355241700011
内容类型期刊论文
源URL[http://10.2.47.112/handle/2XS4QKH4/1556]  
专题上海财经大学
通讯作者Jiang, Bo
作者单位1.Univ Minnesota, Dept Ind & Syst Engn, Minneapolis, MN 55455 USA;
2.Shanghai Univ Finance & Econ, Sch Informat Management & Engn, Res Ctr Management Sci & Data Analyt, Shanghai 200433, Peoples R China;
3.Univ Minnesota, Dept Ind & Syst Engn, Minneapolis, MN 55455 USA
推荐引用方式
GB/T 7714
Gao, Xiang,Jiang, Bo,Tao, Shaozhe. RECOVERING LOW CP/TUCKER RANKED TENSORS, WITH APPLICATIONS IN TENSOR COMPLETION[J]. PACIFIC JOURNAL OF OPTIMIZATION,2015,11(2):385-402.
APA Gao, Xiang,Jiang, Bo,&Tao, Shaozhe.(2015).RECOVERING LOW CP/TUCKER RANKED TENSORS, WITH APPLICATIONS IN TENSOR COMPLETION.PACIFIC JOURNAL OF OPTIMIZATION,11(2),385-402.
MLA Gao, Xiang,et al."RECOVERING LOW CP/TUCKER RANKED TENSORS, WITH APPLICATIONS IN TENSOR COMPLETION".PACIFIC JOURNAL OF OPTIMIZATION 11.2(2015):385-402.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace