Sparse density estimation with l1 penalties

被引：25

作者：

Bunea, Florentina ^{[1
]}

Tsybakov, Alexandre B. ^{[2
]}

Wegkamp, Marten H. ^{[1
]}

机构：

[1] Florida State Univ, Tallahassee, FL 32306 USA

[2] Univ Paris 04, Lab Probabil Modeles Aeatoires, Paris, France

来源：

LEARNING THEORY, PROCEEDINGS | 2007年 / 4539卷

关键词：

D O I：

10.1007/978-3-540-72927-3_38

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper studies oracle properties of l(1)-penalized estimators of a probability density. We show that the penalized least squares estimator satisfies sparsity oracle inequalities, i.e., bounds in terms of the number of non-zero components of the oracle vector. The results are valid even when the dimension of the model is (much) larger than the sample size. They are applied to estimation in sparse high-dimensional mixture models, to nonparametric adaptive density estimation and to the problem of aggregation of density estimators.

引用

页码：530 / +

页数：3

共 30 条

[21]

NEMIROVSKI A, 2000, LECT NOTES MATH, V28

[22]

RIGOLLET P, 2006, THESIS U PARIS 6

[23]

RIGOLLET PH, 2004, LINEAR CONVEX AGGREG

[24]

RUDEMO M, 1982, SCAND J STAT, V9, P65

[25] Regression shrinkage and selection via the Lasso [J].

Tibshirani, R .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 1996, 58 (01) :267-288

[26]

TSYBAKOV AB, 2003, LNCS LNAI, V2777

[27]

VANDEGEER SA, 2006, SEM STAT ETH ZUR

[28]

Vapnik V., 1998, STAT LEARNING THEORY, V1, P2

[29] Quasi-universal bandwidth selection for kernel density estimators [J].

Wegkamp, MH .

CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, 1999, 27 (02) :409-420

[30]

[No title captured]

← 1 2 3 →