The Geometry of Uniqueness, Sparsity and Clustering in Penalized Estimation

被引:0
|
作者
Schneider, Ulrike [1 ]
Tardivel, Patrick [2 ,3 ]
机构
[1] TU Wien, Inst Stat & Math Methods Econ, Vienna, Austria
[2] Univ Wroclaw, Inst Math, Wroclaw, Poland
[3] Univ Burgundy, Dijon, France
关键词
Penalized Estimation; SLOPE; Uniqueness; Sparsity; Clustering; Regulariza-tion; Geometry; Polytope; VARIABLE SELECTION; REGRESSION SHRINKAGE; LASSO PROBLEM; RECOVERY; POLYTOPES; L1;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We provide a necessary and sufficient condition for the uniqueness of penalized least-squares estimators whose penalty term is given by a norm with a polytope unit ball, covering a wide range of methods including SLOPE, PACS, fused, clustered and classical LASSO as well as the related method of basis pursuit. We consider a strong type of uniqueness that is relevant for statistical problems. The uniqueness condition is geometric and involves how the row span of the design matrix intersects the faces of the dual norm unit ball, which for SLOPE is given by the signed permutahedron. Further considerations based this condition also allow to derive results on sparsity and clustering features. In particular, we define the notion of a SLOPE pattern to describe both sparsity and clustering properties of this method and also provide a geometric characterization of accessible SLOPE patterns.
引用
收藏
页数:36
相关论文
共 50 条
  • [1] The Geometry of Uniqueness, Sparsity and Clustering in Penalized Estimation
    Schneider, Ulrike
    Tardivel, Patrick
    Journal of Machine Learning Research, 2022, 23
  • [2] Performance Analysis of Sparsity-Penalized LMS Algorithms in Channel Estimation
    Yang, Jie
    Huang, Hao
    Wang, Jie
    Hong, Sheng
    Hua, Zijian
    Zhang, Jian
    Gui, Guan
    ADVANCED HYBRID INFORMATION PROCESSING, 2018, 219 : 407 - 416
  • [3] PENALIZED ESTIMATION OF HIGH-DIMENSIONAL MODELS UNDER A GENERALIZED SPARSITY CONDITION
    Horowitz, Joel L.
    Huang, Jian
    STATISTICA SINICA, 2013, 23 (02) : 725 - 748
  • [4] A penalized likelihood estimation on transcriptional module-based clustering
    Yoshida, R
    Imoto, S
    Higuchi, T
    COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2005, PT 3, 2005, 3482 : 389 - 401
  • [5] Sparsity in penalized empirical risk minimization
    Koltchinskii, Vladimir
    ANNALES DE L INSTITUT HENRI POINCARE-PROBABILITES ET STATISTIQUES, 2009, 45 (01): : 7 - 57
  • [6] Complex Interbank Network Estimation: Sparsity-Clustering Threshold
    Bundi, Nils
    Khashanah, Khaldoun
    COMPLEX NETWORKS AND THEIR APPLICATIONS VII, VOL 2, 2019, 813 : 487 - 498
  • [7] Imposing uniqueness to achieve sparsity
    Dillon, Keith
    Wang, Yu-Ping
    SIGNAL PROCESSING, 2016, 123 : 1 - 8
  • [8] Penalized probabilistic clustering
    Lu, Zhengdong
    Leen, Todd K.
    NEURAL COMPUTATION, 2007, 19 (06) : 1528 - 1567
  • [9] Density Estimation by Total Variation Penalized Likelihood Driven by the Sparsity l1 Information Criterion
    Sardy, Sylvain
    Tseng, Paul
    SCANDINAVIAN JOURNAL OF STATISTICS, 2010, 37 (02) : 321 - 337
  • [10] Penalized Model-Based Clustering with Group-Dependent Shrinkage Estimation
    Casa, Alessandro
    Cappozzo, Andrea
    Fop, Michael
    BUILDING BRIDGES BETWEEN SOFT AND STATISTICAL METHODOLOGIES FOR DATA SCIENCE, 2023, 1433 : 73 - 78