Online learning of approximate maximum p-norm margin classifiers with bias

Kosuke Ishibashi, Kohei Hatano, Masayuki Takeda

Research output: Contribution to conferencePaper

9 Citations (Scopus)

Abstract

We propose a new online learning algorithm which provably approximates maximum margin classifiers with bias, where the margin is defined in terms of p-norm distance. Although learning of linear classifiers with bias can be reduced to learning of those without bias, the known reduction might lose the margin and slow down the convergence of online learning algorithms. Our algorithm, unlike previous online learning algorithms, implicitly uses a new reduction which preserves the margin and avoids such possible deficiencies. Our preliminary experiments show that our algorithm runs much faster than previous algorithms especially when the underlying linear classifier has large bias.

Original languageEnglish
Pages69-80
Number of pages12
Publication statusPublished - Dec 1 2008
Event21st Annual Conference on Learning Theory, COLT 2008 - Helsinki, Finland
Duration: Jul 9 2008Jul 12 2008

Other

Other21st Annual Conference on Learning Theory, COLT 2008
CountryFinland
CityHelsinki
Period7/9/087/12/08

Fingerprint

trend
learning
experiment

All Science Journal Classification (ASJC) codes

  • Education

Cite this

Ishibashi, K., Hatano, K., & Takeda, M. (2008). Online learning of approximate maximum p-norm margin classifiers with bias. 69-80. Paper presented at 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland.

Online learning of approximate maximum p-norm margin classifiers with bias. / Ishibashi, Kosuke; Hatano, Kohei; Takeda, Masayuki.

2008. 69-80 Paper presented at 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland.

Research output: Contribution to conferencePaper

Ishibashi, K, Hatano, K & Takeda, M 2008, 'Online learning of approximate maximum p-norm margin classifiers with bias', Paper presented at 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland, 7/9/08 - 7/12/08 pp. 69-80.
Ishibashi K, Hatano K, Takeda M. Online learning of approximate maximum p-norm margin classifiers with bias. 2008. Paper presented at 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland.
Ishibashi, Kosuke ; Hatano, Kohei ; Takeda, Masayuki. / Online learning of approximate maximum p-norm margin classifiers with bias. Paper presented at 21st Annual Conference on Learning Theory, COLT 2008, Helsinki, Finland.12 p.
@conference{80e4735649074b5691233cc791999be9,
title = "Online learning of approximate maximum p-norm margin classifiers with bias",
abstract = "We propose a new online learning algorithm which provably approximates maximum margin classifiers with bias, where the margin is defined in terms of p-norm distance. Although learning of linear classifiers with bias can be reduced to learning of those without bias, the known reduction might lose the margin and slow down the convergence of online learning algorithms. Our algorithm, unlike previous online learning algorithms, implicitly uses a new reduction which preserves the margin and avoids such possible deficiencies. Our preliminary experiments show that our algorithm runs much faster than previous algorithms especially when the underlying linear classifier has large bias.",
author = "Kosuke Ishibashi and Kohei Hatano and Masayuki Takeda",
year = "2008",
month = "12",
day = "1",
language = "English",
pages = "69--80",
note = "21st Annual Conference on Learning Theory, COLT 2008 ; Conference date: 09-07-2008 Through 12-07-2008",

}

TY - CONF

T1 - Online learning of approximate maximum p-norm margin classifiers with bias

AU - Ishibashi, Kosuke

AU - Hatano, Kohei

AU - Takeda, Masayuki

PY - 2008/12/1

Y1 - 2008/12/1

N2 - We propose a new online learning algorithm which provably approximates maximum margin classifiers with bias, where the margin is defined in terms of p-norm distance. Although learning of linear classifiers with bias can be reduced to learning of those without bias, the known reduction might lose the margin and slow down the convergence of online learning algorithms. Our algorithm, unlike previous online learning algorithms, implicitly uses a new reduction which preserves the margin and avoids such possible deficiencies. Our preliminary experiments show that our algorithm runs much faster than previous algorithms especially when the underlying linear classifier has large bias.

AB - We propose a new online learning algorithm which provably approximates maximum margin classifiers with bias, where the margin is defined in terms of p-norm distance. Although learning of linear classifiers with bias can be reduced to learning of those without bias, the known reduction might lose the margin and slow down the convergence of online learning algorithms. Our algorithm, unlike previous online learning algorithms, implicitly uses a new reduction which preserves the margin and avoids such possible deficiencies. Our preliminary experiments show that our algorithm runs much faster than previous algorithms especially when the underlying linear classifier has large bias.

UR - http://www.scopus.com/inward/record.url?scp=80054116173&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80054116173&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:80054116173

SP - 69

EP - 80

ER -