Minimum error classification with geometric margin control

Hideyuki Watanabe; Shigeru Katagiri; Kouta Yamada; Erik McDermott; Atsushi Nakamura; Shinji Watanabe; Miho Ohsaki

doi:10.1109/ICASSP.2010.5495645

Minimum error classification with geometric margin control

Hideyuki Watanabe^*, Shigeru Katagiri, Kouta Yamada, Erik McDermott, Atsushi Nakamura, Shinji Watanabe, Miho Ohsaki

^*Corresponding author for this work

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

8 Citations (Scopus)

Abstract

Minimum Classification Error (MCE) training, which can be used to achieve minimum error classification of various types of patterns, has attracted a great deal of attention. However, to increase classification robustness, a conventional MCE framework has no practical optimization procedures like geometric margin maximization in Support Vector Machine (SVM). To realize high robustness in a wide range of classification tasks, we derive the geometric margin for a general class of discriminant functions and develop a new MCE training method that increases the geometric margin value. We also experimentally demonstrate the effectiveness of our new method using prototype-based classifiers.

Original language	English
Title of host publication	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2170-2173
Number of pages	4
ISBN (Print)	9781424442966
DOIs	https://doi.org/10.1109/ICASSP.2010.5495645
Publication status	Published - 2010
Externally published	Yes
Event	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States Duration: 2010 Mar 14 → 2010 Mar 19

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Conference

Conference	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
Country/Territory	United States
City	Dallas, TX
Period	10/3/14 → 10/3/19

Keywords

Discriminative training
Geometric margin
MCE
Margin
Minimum classification error

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2010.5495645

Cite this

Watanabe, H., Katagiri, S., Yamada, K., McDermott, E., Nakamura, A., Watanabe, S., & Ohsaki, M. (2010). Minimum error classification with geometric margin control. In 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings (pp. 2170-2173). [5495645] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2010.5495645

Minimum error classification with geometric margin control. / Watanabe, Hideyuki; Katagiri, Shigeru; Yamada, Kouta et al.

2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2010. p. 2170-2173 5495645 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

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

Watanabe, H, Katagiri, S, Yamada, K, McDermott, E, Nakamura, A, Watanabe, S & Ohsaki, M 2010, Minimum error classification with geometric margin control. in 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings., 5495645, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 2170-2173, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 10/3/14. https://doi.org/10.1109/ICASSP.2010.5495645

Watanabe H, Katagiri S, Yamada K, McDermott E, Nakamura A, Watanabe S et al. Minimum error classification with geometric margin control. In 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2010. p. 2170-2173. 5495645. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2010.5495645

Watanabe, Hideyuki ; Katagiri, Shigeru ; Yamada, Kouta et al. / Minimum error classification with geometric margin control. 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2010. pp. 2170-2173 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{dbe7367b4d184dcbae3b31b158b1f4a9,

title = "Minimum error classification with geometric margin control",

abstract = "Minimum Classification Error (MCE) training, which can be used to achieve minimum error classification of various types of patterns, has attracted a great deal of attention. However, to increase classification robustness, a conventional MCE framework has no practical optimization procedures like geometric margin maximization in Support Vector Machine (SVM). To realize high robustness in a wide range of classification tasks, we derive the geometric margin for a general class of discriminant functions and develop a new MCE training method that increases the geometric margin value. We also experimentally demonstrate the effectiveness of our new method using prototype-based classifiers.",

keywords = "Discriminative training, Geometric margin, MCE, Margin, Minimum classification error",

author = "Hideyuki Watanabe and Shigeru Katagiri and Kouta Yamada and Erik McDermott and Atsushi Nakamura and Shinji Watanabe and Miho Ohsaki",

year = "2010",

doi = "10.1109/ICASSP.2010.5495645",

language = "English",

isbn = "9781424442966",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "2170--2173",

booktitle = "2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings",

note = "2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 ; Conference date: 14-03-2010 Through 19-03-2010",

}

TY - GEN

T1 - Minimum error classification with geometric margin control

AU - Watanabe, Hideyuki

AU - Katagiri, Shigeru

AU - Yamada, Kouta

AU - McDermott, Erik

AU - Nakamura, Atsushi

AU - Watanabe, Shinji

AU - Ohsaki, Miho

PY - 2010

Y1 - 2010

N2 - Minimum Classification Error (MCE) training, which can be used to achieve minimum error classification of various types of patterns, has attracted a great deal of attention. However, to increase classification robustness, a conventional MCE framework has no practical optimization procedures like geometric margin maximization in Support Vector Machine (SVM). To realize high robustness in a wide range of classification tasks, we derive the geometric margin for a general class of discriminant functions and develop a new MCE training method that increases the geometric margin value. We also experimentally demonstrate the effectiveness of our new method using prototype-based classifiers.

AB - Minimum Classification Error (MCE) training, which can be used to achieve minimum error classification of various types of patterns, has attracted a great deal of attention. However, to increase classification robustness, a conventional MCE framework has no practical optimization procedures like geometric margin maximization in Support Vector Machine (SVM). To realize high robustness in a wide range of classification tasks, we derive the geometric margin for a general class of discriminant functions and develop a new MCE training method that increases the geometric margin value. We also experimentally demonstrate the effectiveness of our new method using prototype-based classifiers.

KW - Discriminative training

KW - Geometric margin

KW - MCE

KW - Margin

KW - Minimum classification error

UR - http://www.scopus.com/inward/record.url?scp=78049357326&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049357326&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5495645

DO - 10.1109/ICASSP.2010.5495645

M3 - Conference contribution

AN - SCOPUS:78049357326

SN - 9781424442966

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 2170

EP - 2173

BT - 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010

Y2 - 14 March 2010 through 19 March 2010

ER -

Minimum error classification with geometric margin control

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this