New approaches to statistical learning theory

被引：42

作者：

Bousquet, O ^{[1
]}

机构：

[1] Max Planck Inst Biol Cybernet, D-72076 Tubingen, Germany

来源：

ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS | 2003年 / 55卷 / 02期

关键词：

statistical learning theory; concentration inequalities; Rademacher averages; error bounds;

D O I：

10.1007/BF02530506

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We present new tools from probability theory that can be applied to the analysis of learning algorithms. These tools allow to derive new bounds on the generalization performance of learning algorithms and to propose alternative measures of the complexity of the learning task, which in turn can be used to derive new learning algorithms.

引用

页码：371 / 389

页数：19

共 19 条

[1] A RESULT OF VAPNIK WITH APPLICATIONS [J].

ANTHONY, M ;

SHAWETAYLOR, J .

DISCRETE APPLIED MATHEMATICS, 1993, 47 (03) :207-217

[2]

BARTLETT P, 2002, LECT NOTES COMPUT SC, P44

[3]

BARTLETT P, 2002, LOCAL RADEMACHER COM

[4]

Bartlett P. L., 2003, Journal of Machine Learning Research, V3, P463, DOI 10.1162/153244303321897690

[5] Model selection and error estimation [J].

Bartlett, PL ;

Boucheron, S ;

Lugosi, G .

MACHINE LEARNING, 2002, 48 (1-3) :85-113

[6]

Boucheron S, 2000, RANDOM STRUCT ALGOR, V16, P277, DOI 10.1002/(SICI)1098-2418(200005)16:3<277::AID-RSA4>3.0.CO

[7]

2-1

[8]

BOUCHERON S, 2002, IN PRESS ANN PROBAB

[9] Stability and generalization [J].

Bousquet, O ;

Elisseeff, A .

JOURNAL OF MACHINE LEARNING RESEARCH, 2002, 2 (03) :499-526

[10] A Bennett concentration inequality and its application to suprema of empirical processes [J].

Bousquet, O .

COMPTES RENDUS MATHEMATIQUE, 2002, 334 (06) :495-500

← 1 2 →