Adaptation in statistical pattern recognition using tangent vectors

被引：37

作者：

Keysers, D ^{[1
]}

Macherey, W ^{[1
]}

Ney, G ^{[1
]}

Dahmen, J ^{[1
]}

机构：

[1] Rhein Westfal TH Aachen, Aachen Tech Univ, Dept Comp Sci, Lehrstuhl Informat 6, D-52056 Aachen, Germany

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 2004年 / 26卷 / 02期

关键词：

statistical pattern recognition; adaptation; tangent vectors; linear models;

D O I：

10.1109/TPAMI.2004.1262198

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We integrate the tangent method into a statistical framework for classification analytically and practically. The resulting consistent framework for adaptation allows us to efficiently estimate the tangent vectors representing the variability. The framework improves classification results on two real-world pattern recognition tasks from the domains handwritten character recognition and automatic speech recognition.

引用

页码：269 / 274

页数：6

共 21 条

[1]

Bishop C. M., 1996, Neural networks for pattern recognition

[2]

Bishop CM, 1999, ADV NEUR IN, V11, P382

[3] Statistical image object recognition using mixture densities [J].

Dahmen, J ;

Keysers, D ;

Ney, H ;

Güld, MO .

JOURNAL OF MATHEMATICAL IMAGING AND VISION, 2001, 14 (03) :285-296

[4]

EISELE T, 1996, P INT C SPOK LANG PR, V1, P252

[5]

Friedman J., 2001, The elements of statistical learning, V1, DOI DOI 10.1007/978-0-387-21606-5

[6]

Fukunaga K., 1990, INTRO STAT PATTERN R

[7]

Hart, 2006, PATTERN CLASSIFICATI

[8]

Hastie T, 1998, STAT SCI, V13, P54

[9] Modeling the manifolds of images of handwritten digits [J].

Hinton, GE ;

Dayan, P ;

Revow, M .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (01) :65-74

[10] Dimension reduction by local principal component analysis [J].

Kambhatla, N ;

Leen, TK .

NEURAL COMPUTATION, 1997, 9 (07) :1493-1516

← 1 2 3 →