An asymptotically optimal kernel combined classifier

被引：1

作者：

Mojirsheibani, Majid ^{[1
]}

Kong, Jiajie ^{[1
]}

机构：

[1] Calif State Univ Northridge, Dept Math, Northridge, CA 91330 USA

来源：

STATISTICS & PROBABILITY LETTERS | 2016年 / 119卷

基金：

美国国家科学基金会;

关键词：

Kernel; Hamming distance; Smoothing parameter; Combined classifier; RANDOM FORESTS; REGRESSION; IMPROVE;

D O I：

10.1016/j.spl.2016.07.017

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：

A kernel ensemble classifier is developed for accurate classification based on several initial classifiers. A data-driven choice of the smoothing parameter of the kernel is considered and the resulting classifier is shown to be asymptotically optimal. Therefore, the proposed combined classifier asymptotically outperforms each individual classifier. (C) 2016 Elsevier B.V. All rights reserved.

引用

页码：91 / 100

页数：10

共 23 条

[1] Ensemble classification of paired data [J].

Adler, Werner ;

Brenning, Alexander ;

Potapov, Sergej ;

Schmid, Matthias ;

Lausen, Berthold .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2011, 55 (05) :1933-1941

[2] A simple method for combining estimates to improve the overall error rates in classification [J].

Balakrishnan, Narayanaswamy ;

Mojirsheibani, Majid .

COMPUTATIONAL STATISTICS, 2015, 30 (04) :1033-1049

[3] COBRA: A combined regression strategy [J].

Biau, Gerard ;

Fischer, Aurelie ;

Guedj, Benjamin ;

Malley, James D. .

JOURNAL OF MULTIVARIATE ANALYSIS, 2016, 146 :18-28

[4]

Biau G, 2008, J MACH LEARN RES, V9, P2015

[5]

Boser B. E., 1992, Proceedings of the Fifth Annual ACM Workshop on Computational Learning Theory, P144, DOI 10.1145/130385.130401

[6]

Breiman L, 1996, MACH LEARN, V24, P49

[7] Random forests [J].

Breiman, L .

MACHINE LEARNING, 2001, 45 (01) :5-32

[8] Random forests [J].

Breiman, L .

MACHINE LEARNING, 2001, 45 (01) :5-32

[9] Ensemble classification based on generalized additive models [J].

De Bock, Koen W. ;

Coussement, Kristof ;

Van den Poel, Dirk .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2010, 54 (06) :1535-1546

[10]

Devroye L., 2013, PROBABILISTIC THEORY, V31

← 1 2 3 →