Constructing descriptive and discriminative nonlinear features:: Rayleigh coefficients in kernel feature spaces

被引：147

作者：

Mika, S

Rätsch, G

Weston, J

Schölkopf, B

Smola, A

Müller, KR

机构：

[1] Fraunhofer FIRST, D-12489 Berlin, Germany

[2] Univ Potsdam, D-14469 Potsdam, Germany

[3] Australian Natl Univ, Canberra, ACT 0200, Australia

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

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 2003年 / 25卷 / 05期

关键词：

Fisher's discriminant; nonlinear feature extraction; support vector machine; kernel functions; Rayleigh coefficient; oriented PCA;

D O I：

10.1109/TPAMI.2003.1195996

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We incorporate prior knowledge to construct nonlinear algorithms for invariant feature extraction and discrimination. Employing a unified framework in terms of a nonlinearized variant of the Rayleigh coefficient, we propose nonlinear generalizations of Fisher's discriminant and oriented PCA using support vector kernel functions. Extensive simulations show the utility of our approach.

引用

页码：623 / 628

页数：6

共 30 条

[1]

[Anonymous], [No title captured]

[2]

[Anonymous], 1998, Encyclopedia of Biostatistics

[3]

[Anonymous], P INT C COMP LEARN T

[4] Generalized discriminant analysis using a kernel approach [J].

Baudat, G ;

Anouar, FE .

NEURAL COMPUTATION, 2000, 12 (10) :2385-2404

[5]

Bishop C. M., 1995, NEURAL NETWORKS PATT

[6]

Blake C.L., 1998, UCI repository of machine learning databases

[7]

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

[8] Stability and generalization [J].

Bousquet, O ;

Elisseeff, A .

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

[9]

Diamantaras KI, 1996, Principal Component Neural Networks: Theory and Applications

[10] The use of multiple measurements in taxonomic problems [J].

Fisher, RA .

ANNALS OF EUGENICS, 1936, 7 :179-188

← 1 2 3 →