Approximation bounds for sparse principal component analysis

被引:0
|
作者
Alexandre d’Aspremont
Francis Bach
Laurent El Ghaoui
机构
[1] CNRS & D.I.,École Normale Supérieure
[2] UMR 8548,École Normale Supérieure
[3] INRIA,EECS
[4] SIERRA Project-Team & D.I.,undefined
[5] U.C. Berkeley,undefined
来源
Mathematical Programming | 2014年 / 148卷
关键词
62H25; 90C22; 90C27;
D O I
暂无
中图分类号
学科分类号
摘要
We produce approximation bounds on a semidefinite programming relaxation for sparse principal component analysis. The sparse maximum eigenvalue problem cannot be efficiently approximated up to a constant approximation ratio, so our bounds depend on the optimum value of the semidefinite relaxation: the higher this value, the better the approximation. In particular, these bounds allow us to control approximation ratios for tractable statistics in hypothesis testing problems where data points are sampled from Gaussian models with a single sparse leading component.
引用
收藏
页码:89 / 110
页数:21
相关论文
共 50 条
  • [1] Approximation bounds for sparse principal component analysis
    d'Aspremont, Alexandre
    Bach, Francis
    El Ghaoui, Laurent
    MATHEMATICAL PROGRAMMING, 2014, 148 (1-2) : 89 - 110
  • [2] Exact and Approximation Algorithms for Sparse Principal Component Analysis
    Li, Yongchun
    Xie, Weijun
    INFORMS JOURNAL ON COMPUTING, 2024,
  • [3] Sparse principal component analysis
    Zou, Hui
    Hastie, Trevor
    Tibshirani, Robert
    JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2006, 15 (02) : 265 - 286
  • [4] Sparse principal component analysis via regularized low rank matrix approximation
    Shen, Haipeng
    Huang, Jianhua Z.
    JOURNAL OF MULTIVARIATE ANALYSIS, 2008, 99 (06) : 1015 - 1034
  • [5] Robust sparse principal component analysis
    ZHAO Qian
    MENG DeYu
    XU ZongBen
    Science China(Information Sciences), 2014, 57 (09) : 175 - 188
  • [6] Multilinear Sparse Principal Component Analysis
    Lai, Zhihui
    Xu, Yong
    Chen, Qingcai
    Yang, Jian
    Zhang, David
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2014, 25 (10) : 1942 - 1950
  • [7] Robust Sparse Principal Component Analysis
    Croux, Christophe
    Filzmoser, Peter
    Fritz, Heinrich
    TECHNOMETRICS, 2013, 55 (02) : 202 - 214
  • [8] Robust sparse principal component analysis
    Zhao Qian
    Meng DeYu
    Xu ZongBen
    SCIENCE CHINA-INFORMATION SCIENCES, 2014, 57 (09) : 1 - 14
  • [9] Weighted sparse principal component analysis
    Van Deun, Katrijn
    Thorrez, Lieven
    Coccia, Margherita
    Hasdemir, Dicle
    Westerhuis, Johan A.
    Smilde, Age K.
    Van Mechelen, Iven
    CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2019, 195
  • [10] Biobjective sparse principal component analysis
    Carrizosa, Emilio
    Guerrero, Vanesa
    JOURNAL OF MULTIVARIATE ANALYSIS, 2014, 132 : 151 - 159