A new discriminative kernel from probabilistic models

K. Tsuda; M. Kawanabe; G. Ratsch; S. Sonnenburg; K. R. Muller

A new discriminative kernel from probabilistic models

K. Tsuda, M. Kawanabe, G. Ratsch, S. Sonnenburg, K. R. Muller

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Recently. Jaakkola and Haussler proposed a method for constructing kernel functions from probabilistic models. Their so called "Fisher kernel" has been combined with discriminative classifiers such as SVM and applied successfully in e.g. DNA and protein analysis. Whereas the Fisher kernel (FK) is calculated from the marginal log-likelihood, we propose the TOP kernel derived from Tangent vectors Of Posterior log-odds. Furthermore we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing FK and TOP. In experiments our new discriminative TOP kernel compares favorably to the Fisher kernel.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 14 - Proceedings of the 2001 Conference, NIPS 2001
Publisher	Neural information processing systems foundation
ISBN (Print)	0262042088, 9780262042086
Publication status	Published - 2002
Externally published	Yes
Event	15th Annual Neural Information Processing Systems Conference, NIPS 2001 - Vancouver, BC, Canada Duration: 2001 Dec 3 → 2001 Dec 8

Publication series

Name	Advances in Neural Information Processing Systems
ISSN (Print)	1049-5258

Other

Other	15th Annual Neural Information Processing Systems Conference, NIPS 2001
Country/Territory	Canada
City	Vancouver, BC
Period	01/12/3 → 01/12/8

Bibliographical note

Copyright:
Copyright 2014 Elsevier B.V., All rights reserved.

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems
Signal Processing

Cite this

A new discriminative kernel from probabilistic models. / Tsuda, K.; Kawanabe, M.; Ratsch, G. et al.
Advances in Neural Information Processing Systems 14 - Proceedings of the 2001 Conference, NIPS 2001. Neural information processing systems foundation, 2002. (Advances in Neural Information Processing Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Tsuda, K, Kawanabe, M, Ratsch, G, Sonnenburg, S & Muller, KR 2002, A new discriminative kernel from probabilistic models. in Advances in Neural Information Processing Systems 14 - Proceedings of the 2001 Conference, NIPS 2001. Advances in Neural Information Processing Systems, Neural information processing systems foundation, 15th Annual Neural Information Processing Systems Conference, NIPS 2001, Vancouver, BC, Canada, 01/12/3.

@inproceedings{e98f27e22e4a4e14bf2adb82569553b0,

title = "A new discriminative kernel from probabilistic models",

abstract = "Recently. Jaakkola and Haussler proposed a method for constructing kernel functions from probabilistic models. Their so called {"}Fisher kernel{"} has been combined with discriminative classifiers such as SVM and applied successfully in e.g. DNA and protein analysis. Whereas the Fisher kernel (FK) is calculated from the marginal log-likelihood, we propose the TOP kernel derived from Tangent vectors Of Posterior log-odds. Furthermore we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing FK and TOP. In experiments our new discriminative TOP kernel compares favorably to the Fisher kernel.",

author = "K. Tsuda and M. Kawanabe and G. Ratsch and S. Sonnenburg and Muller, {K. R.}",

year = "2002",

language = "English",

isbn = "0262042088",

series = "Advances in Neural Information Processing Systems",

publisher = "Neural information processing systems foundation",

booktitle = "Advances in Neural Information Processing Systems 14 - Proceedings of the 2001 Conference, NIPS 2001",

}

TY - GEN

T1 - A new discriminative kernel from probabilistic models

AU - Tsuda, K.

AU - Kawanabe, M.

AU - Ratsch, G.

AU - Sonnenburg, S.

AU - Muller, K. R.

PY - 2002

Y1 - 2002

N2 - Recently. Jaakkola and Haussler proposed a method for constructing kernel functions from probabilistic models. Their so called "Fisher kernel" has been combined with discriminative classifiers such as SVM and applied successfully in e.g. DNA and protein analysis. Whereas the Fisher kernel (FK) is calculated from the marginal log-likelihood, we propose the TOP kernel derived from Tangent vectors Of Posterior log-odds. Furthermore we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing FK and TOP. In experiments our new discriminative TOP kernel compares favorably to the Fisher kernel.

AB - Recently. Jaakkola and Haussler proposed a method for constructing kernel functions from probabilistic models. Their so called "Fisher kernel" has been combined with discriminative classifiers such as SVM and applied successfully in e.g. DNA and protein analysis. Whereas the Fisher kernel (FK) is calculated from the marginal log-likelihood, we propose the TOP kernel derived from Tangent vectors Of Posterior log-odds. Furthermore we develop a theoretical framework on feature extractors from probabilistic models and use it for analyzing FK and TOP. In experiments our new discriminative TOP kernel compares favorably to the Fisher kernel.

UR - http://www.scopus.com/inward/record.url?scp=84899032930&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899032930&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84899032930

SN - 0262042088

SN - 9780262042086

T3 - Advances in Neural Information Processing Systems

BT - Advances in Neural Information Processing Systems 14 - Proceedings of the 2001 Conference, NIPS 2001

PB - Neural information processing systems foundation

T2 - 15th Annual Neural Information Processing Systems Conference, NIPS 2001

Y2 - 3 December 2001 through 8 December 2001

ER -

A new discriminative kernel from probabilistic models

Abstract

Publication series

Other

Bibliographical note

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this