期刊文献+

不完全非负矩阵分解的加速算法 被引量:13

Accelerated Algorithm to Incomplete Nonnegative Matrix Factorization
下载PDF
导出
摘要 非负矩阵分解(NMF)已成为数据分析与处理的一种日益流行的方法.当数据矩阵不完全时,可用加权非负矩阵分解(WNMF)来分解矩阵.但是在WNMF算法中,对于给定的搜索方向,步长的选取一般来说不是最优的.本文研究了不完全非负矩阵分解(INMF)问题,提出了加速算法(AINMF).首先,将INMF问题转化为交替地求解两个非负最小二乘(NNLS)问题.对于每个NNLS问题,在搜索方向上采用精确的步长.接着,分析了NNLS问题的算法复杂度.最后,试验结果证实了AINMF优于WNMF. Nonnegative matrix factorization(NMF) is an increasingly popular technique for data processing and analysis.For an incomplete data matrix,the weighted nonnegative matrix factorization(WNMF) is employed to decompose it.But the searching step size in WNMF is not optimal along the given searching direction.This paper studies the incomplete nonnegative matrix factorization(INMF) and proposes an accelerated algorithm.First,INMF is transformed into solving alternatively two nonnegative least squares(NNLS) problems.For each NNLS problem,the exact step size is chosen along the searching direction.Then,the complexity of NNLS problems is analyzed.Finally,experimental results show that the proposed method outperforms WNMF.
出处 《电子学报》 EI CAS CSCD 北大核心 2011年第2期291-295,共5页 Acta Electronica Sinica
基金 国家973重点基础研究发展计划(No.2006CB705707) 国家863高技术研究发展计划(No.2007AA12Z223 No.2007AA12Z136) 国家自然科学基金(No.60603019 No.60602064 No.60702062) 长江学者和创新团队发展计划(No.IRT0645)
关键词 非负矩阵分解 不完全非负矩阵分解 数据丢失问题 加权非负矩阵分解 非负最小二乘 nonnegative matrix factorization incomplete nonnegative matrix factorization missing data problem weighted nonnegative matrix factorization nonnegative least squares
  • 相关文献

参考文献17

  • 1LlU Weixiang ZHENG Nanning YOU Qubo.Nonnegative matrix factorization and its applications in pattern recognition[J].Chinese Science Bulletin,2006,51(1):7-18. 被引量:22
  • 2D D Lee, H S Seung. Learning the parts of objects by nonnega five matrix factorization [ J ]. Nature, 1999,401 (6755) : 788 - 791. 被引量:1
  • 3A Pascual-Montano, J M Carazo, K Kochik, et al. Nonsmooth nonnegative matrix factorization (nsNMF) [ J]. IEEE. Transac tions on Pattern Analysis and Machine Intelligence, 2006, 28 (3) :403 - 415. 被引量:1
  • 4A Cichocki, R Zdunek, S Amari. Nonnegative matrix and tensor Factorization[ J]. IEEE Signal Processing Magazine, 2008, 25 (1) : 142 - 145. 被引量:1
  • 5C Fevotte, N Bertin, J L Dun'ieu. Nonnegafive matrix factoriza tion with the Itakura-Saito divergence with application to music analysis[ J ]. Neural Computation, 2009,21 (3) : 793 - 830. 被引量:1
  • 6C Ding, X He, H D Simon. On the equivalence of nonnegative matrix factorization and spectral clustering[ A]. Proc. SIAM Intemafional Conference on Data Mining [ C ]. Newport Beach, California, 2005. 606 - 610. 被引量:1
  • 7W Xu, X Liu, Y Gong. Document-clustering based on non-neg ative matrix factorization[A]. Proc ACM SIGIRE C]. Toronto, Canada, 2003. 267 - 273. 被引量:1
  • 8H Kim, H Park. Sparse non-negative matrix factorizations via alternating non-negativity-conslrained least squares for microar ray data analysis [ J]. Bioinformatics, 2(107, 23 (12) : 1495 - 1502. 被引量:1
  • 9E J Candes,B Recht. Exact matrix completion via conve opti mization[ J]. Foundations of Computational Mathematics, 2009, 9(6) :717 - 772. 被引量:1
  • 10R H Keshavan, Oh Sewoong. A Montanari. Matrix completion from a few entries[ A ]. Proc IEEE International Symposium on Information Theory[ C]. Seoul, Korea, 2009. 324 - 328. 被引量:1

二级参考文献37

共引文献122

同被引文献131

  • 1王芳,盛卫星,马晓峰,王昊.基于B(2D)~2PGNMF的ISAR像目标识别[J].南京理工大学学报,2013,37(6):863-868. 被引量:2
  • 2LlU Weixiang ZHENG Nanning YOU Qubo.Nonnegative matrix factorization and its applications in pattern recognition[J].Chinese Science Bulletin,2006,51(1):7-18. 被引量:22
  • 3Yan S,Xu D,Zhang B,et al.Graph embedding and extensions:A general framework for dimensionality reduction[J].IEEE Transac tions on Pattern Analysis and Machine Intelligence,2007,29( 1 ) : 40-51. 被引量:1
  • 4Cheng B, Yang J, Yan S, et al.Learning with l'-graph for image analysis[J].IEEE Transactions on Image Processing, 2010, 19 (4) : 858-866. 被引量:1
  • 5Wright J, Yang A, Sastry S, et al.Robust face recognition via sparse representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31 (2) :210-227. 被引量:1
  • 6Qiao L,Chen S, Tan X.Sparsity preserving projections with applications to face recognition[J].Pattem Recognition, 2010,43 ( 1 ) : 331-341. 被引量:1
  • 7Sha F,Lin Y,Saul L K,et al.Multiplicative updates for nonnegative quadratic programming[J].Neural Computation,2007, 19(8): 2004-2031. 被引量:1
  • 8Lee D, Seung H S.Learning the parts of objects by non-negative matrix factorization[J].Nature, 1999,401 : 788-791. 被引量:1
  • 9Hoyer P O.Non-negative sparse coding[C]//Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing, 2002 : 557-565. 被引量:1
  • 10Ding C H Q, Tao L, Jordan M I. Convex and semi-nonnegative matrix factorizations[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(1) :45-55. 被引量:1

引证文献13

二级引证文献33

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部