| 查看: 424 | 回复: 3 | ||
| 本帖产生 1 个 LS-EPI ，点击这里进行查看 | ||
| 当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖 | ||
hopfliking铁杆木虫 (小有名气)
|
[求助]
帮忙查一下这篇文章的检索信息，谢谢
|
|
|
A fast tri-factorization method for low-rank matrix recovery and completion [ 发自手机版 http://muchong.com/3g ] |
回复此楼» 猜你喜欢
国家基金申请书模板内插入图片不可调整大小?
已经有9人回复
-
退学或坚持读
已经有20人回复
-
免疫学博士有名额，速联系
已经有14人回复
-
面上基金申报没有其他的参与者成吗
已经有4人回复
-
多组分精馏求助
已经有6人回复
-
国家级人才课题组招收2026年入学博士
已经有6人回复
muse
捐助贵宾 (知名作家)
- LS-EPI: 147
- 应助: 314 (大学生)
- 贵宾: 0.549
- 金币: 20674
- 散金: 2783
- 红花: 81
- 沙发: 156
- 帖子: 6098
- 在线: 1468.2小时
- 虫号: 1207111
- 注册: 2011-02-19
- 专业: 保险学
3楼2014-09-21 19:05:07
muse
捐助贵宾 (知名作家)
- LS-EPI: 147
- 应助: 314 (大学生)
- 贵宾: 0.549
- 金币: 20674
- 散金: 2783
- 红花: 81
- 沙发: 156
- 帖子: 6098
- 在线: 1468.2小时
- 虫号: 1207111
- 注册: 2011-02-19
- 专业: 保险学
【答案】应助回帖
★ ★ ★ ★ ★
感谢参与，应助指数 +1
hopfliking: 金币+5, ★★★★★最佳答案 2014-09-21 19:15:07
sunshan4379: LS-EPI+1, 感谢应助！ 2014-09-21 19:36:50
感谢参与，应助指数 +1
hopfliking: 金币+5, ★★★★★最佳答案 2014-09-21 19:15:07
sunshan4379: LS-EPI+1, 感谢应助！ 2014-09-21 19:36:50
|
A fast tri-factorization method for low-rank matrix recovery and completion 作者:Liu, YY (Liu, Yuanyuan)[ 1 ] ; Jiao, LC (Jiao, L. C.)[ 1 ] ; Shang, FH (Shang, Fanhua)[ 1 ] PATTERN RECOGNITION 卷: 46 期: 1 页: 163-173 DOI: 10.1016/j.patcog.2012.07.003 出版年: JAN 2013 查看期刊信息 摘要 In recent years, matrix rank minimization problems have received a significant amount of attention in machine learning, data mining and computer vision communities. And these problems can be solved by a convex relaxation of the rank minimization problem which minimizes the nuclear norm instead of the rank of the matrix, and has to be solved iteratively and involves singular value decomposition (SVD) at each iteration. Therefore, those algorithms for nuclear norm minimization problems suffer from high computation cost of multiple SVDs. In this paper, we propose a Fast Tri-Factorization (FTF) method to approximate the nuclear norm minimization problem and mitigate the computation cost of performing SVDs. The proposed FTF method can be used to reliably solve a wide range of low-rank matrix recovery and completion problems such as robust principal component analysis (RPCA), low-rank representation (LRR) and low-rank matrix completion (MC). We also present three specific models for RPCA, LRR and MC problems, respectively. Moreover, we develop two alternating direction method (ADM) based iterative algorithms for solving the above three problems. Experimental results on a variety of synthetic and real-world data sets validate the efficiency, robustness and effectiveness of our FTF method comparing with the state-of-the-art nuclear norm minimization algorithms. (C) 2012 Elsevier Ltd. All rights reserved. 关键词 作者关键词:Rank minimization; Nuclear norm minimization; Matrix completion; Low-rank and sparse decomposition; Low rank representation KeyWords Plus:LINEAR INVERSE PROBLEMS; THRESHOLDING ALGORITHM; FACE RECOGNITION; APPROXIMATION; SEGMENTATION; SUBSPACES 作者信息 通讯作者地址: Liu, YY (通讯作者) [显示增强组织信息的名称] Xidian Univ, Minist Educ China, Key Lab Intelligent Percept & Image Understanding, Mailbox 224,2 S TaiBai Rd, Xian 710071, Peoples R China. 地址: [显示增强组织信息的名称] [ 1 ] Xidian Univ, Minist Educ China, Key Lab Intelligent Percept & Image Understanding, Xian 710071, Peoples R China 电子邮件地址:yuanyuanliu0917@yahoo.com.cn; jlcxidian@163.com; shangfanhua@hotmail.com 基金资助致谢 基金资助机构 授权号 National Natural Science Foundation of China 60971112 60971128 60970067 61072108 Fund for Foreign Scholars in University Research and Teaching Programs (111 Project) B07048 Fundamental Research Funds for the Central Universities JY10000902001 JY10000902041 JY10000902043 查看基金资助信息 出版商 ELSEVIER SCI LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, OXON, ENGLAND 类别 / 分类 研究方向:Computer Science; Engineering Web of Science 类别:Computer Science, Artificial Intelligence; Engineering, Electrical & Electronic 文献信息 文献类型:Article 语种:English 入藏号: WOS:000309785000015 ISSN: 0031-3203 电子 ISSN: 1873-5142 期刊信息 目录： Current Contents Connect® Impact Factor (影响因子): Journal Citation Reports® 其他信息 IDS 号: 020CA Web of Science 核心合集中的 "引用的参考文献": 52 Web of Science 核心合集中的 "被引频次": 1 |
2楼2014-09-21 19:04:53
听雪飞天
新虫 (文坛精英)
- LS-EPI: 4
- 应助: 53 (初中生)
- 金币: 41505.3
- 散金: 110
- 红花: 8
- 沙发: 4512
- 帖子: 19545
- 在线: 677.9小时
- 虫号: 1899380
- 注册: 2012-07-18
- 专业: 高分子材料与环境
【答案】应助回帖
感谢参与，应助指数 +1
|
A fast tri-factorization method for low-rank matrix recovery and completion 作者:Yuanyuan Liu; Jiao, L.C.; Fanhua Shang Pattern Recognition 卷: 46 期: 1 页: 163-73 DOI: 10.1016/j.patcog.2012.07.003 出版年: Jan. 2013 摘要 In recent years, matrix rank minimization problems have received a significant amount of attention in machine learning, data mining and computer vision communities. And these problems can be solved by a convex relaxation of the rank minimization problem which minimizes the nuclear norm instead of the rank of the matrix, and has to be solved iteratively and involves singular value decomposition (SVD) at each iteration. Therefore, those algorithms for nuclear norm minimization problems suffer from high computation cost of multiple SVDs. In this paper, we propose a Fast Tri-Factorization (FTF) method to approximate the nuclear norm minimization problem and mitigate the computation cost of performing SVDs. The proposed FTF method can be used to reliably solve a wide range of low-rank matrix recovery and completion problems such as robust principal component analysis (RPCA), low-rank representation (LRR) and low-rank matrix completion (MC). We also present three specific models for RPCA, LRR and MC problems, respectively. Moreover, we develop two alternating direction method (ADM) based iterative algorithms for solving the above three problems. Experimental results on a variety of synthetic and real-world data sets validate the efficiency, robustness and effectiveness of our FTF method comparing with the state-of-the-art nuclear norm minimization algorithms. [All rights reserved Elsevier]. 作者信息 作者地址: Yuanyuan Liu; Jiao, L.C.; Fanhua Shang; Key Lab. of Intell. Perception & Image Understanding of Minist. of Educ. of China, Xidian Univ., Xi'an, China. 出版商 Elsevier Science Ltd., UK 类别 / 分类 研究方向:Mathematics; Business & Economics (由 Thomson Reuters 提供) 分类代码:A0210 Algebra, set theory, and graph theory; A0250 Probability theory, stochastic processes, and statistics; B0210 Algebra; B0240Z Other topics in statistics; B0260 Optimisation techniques; C1110 Algebra; C1140Z Other topics in statistics; C1180 Optimisation techniques; E0210A Algebra; E0210J Statistics; E0210G Optimisation CODEN  TNRA8受控索引:convex programming; matrix algebra; principal component analysis; singular value decomposition 非受控索引:fast trifactorization method; low rank matrix recovery; low rank matrix completion; matrix rank minimization problems; machine learning; data mining; computer vision communities; convex relaxation; singular value decomposition; SVD; nuclear norm minimization problems; fast tri factorization; FTF; robust principal component analysis; RPCA; low rank representation; LRR; MC; alternating direction method; ADM 文献信息 文献类型:Journal Paper 语种:English 入藏号:12981816 ISSN:0031-3203 参考文献数:52 期刊信息 Impact Factor (影响因子): Journal Citation Reports® 其他信息 处理类型:Bibliography, Theoretical or Mathematical 文献号:S0031-3203(12)00298-1 |
4楼2014-09-21 19:06:55