Provably fast training algorithms for support vector machines

Author

Balcázar, José L. ; Dai, Yang ; Watanabe, Osamu

Author_Institution

Dept. de Llenguatges i Sistemes Inf., Univ. Politecnica de Catalunya, Barcelona, Spain

fYear

2001

fDate

2001

Firstpage

43

Lastpage

50

Abstract

Support vector machines are a family of data analysis algorithms based on convex quadratic programming. We focus on their use for classification: in that case, the SVM algorithms work by maximizing the margin of a classifying hyperplane in a feature space. The feature space is handled by means of kernels if the problems are formulated in dual form. Random sampling techniques successfully used for similar problems are studied. The main contribution is a randomized algorithm for training SVMs, for which we can formally prove an upper bound on the expected running time that is quasilinear on the number of data points. To our knowledge, this is the first algorithm for training SVMs in dual formulation and with kernels for which such a quasilinear time bound has been formally proved

Keywords

convex programming; data analysis; learning (artificial intelligence); learning automata; quadratic programming; randomised algorithms; sampling methods; SVM algorithms; classifying hyperplane; convex quadratic programming; data analysis algorithms; data points; dual formulation; expected running time; feature space; kernels; provably fast training algorithms; quasilinear; quasilinear time bound; random sampling techniques; randomized algorithm; support vector machines; Biomedical engineering; Data analysis; Data mining; Kernel; Quadratic programming; Sampling methods; Space technology; Support vector machine classification; Support vector machines; Upper bound;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Mining, 2001. ICDM 2001, Proceedings IEEE International Conference on

Conference_Location

San Jose, CA

Print_ISBN

0-7695-1119-8

Type

conf

DOI

10.1109/ICDM.2001.989499

Filename

989499