Negative Learning to Prevent Undesirable Misclassification

Kazuki Egashira; Atsuyuki Miyai; Qing Yu; Go Irie; Kiyoharu Aizawa

doi:10.1587/transinf.2023EDL8056

Negative Learning to Prevent Undesirable Misclassification

Kazuki Egashira, Atsuyuki Miyai, Qing Yu, Go Irie, Kiyoharu Aizawa

Department of Information and Computer Technology

Research output: Contribution to journal › Article › peer-review

Abstract

We propose a novel classification problem setting where Undesirable Classes (UCs) are defined for each class. UC is the class you specifically want to avoid misclassifying. To address this setting, we propose a framework to reduce the probabilities for UCs while increasing the probability for a correct class.

Original language	English
Pages (from-to)	144-147
Number of pages	4
Journal	IEICE Transactions on Information and Systems
Volume	E107.D
Issue number	1
DOIs	https://doi.org/10.1587/transinf.2023EDL8056
Publication status	Published - Jan 2024

Keywords

classification
negative learning

Access to Document

10.1587/transinf.2023EDL8056

Cite this

@article{dd0b175e7af047719af1f3127430fd8e,

title = "Negative Learning to Prevent Undesirable Misclassification",

abstract = "We propose a novel classification problem setting where Undesirable Classes (UCs) are defined for each class. UC is the class you specifically want to avoid misclassifying. To address this setting, we propose a framework to reduce the probabilities for UCs while increasing the probability for a correct class.",

keywords = "classification, negative learning",

author = "Kazuki Egashira and Atsuyuki Miyai and Qing Yu and Go Irie and Kiyoharu Aizawa",

note = "Publisher Copyright: Copyright {\textcopyright} 2024 The Institute of Electronics, Information and Communication Engineers.",

year = "2024",

month = jan,

doi = "10.1587/transinf.2023EDL8056",

language = "English",

volume = "E107.D",

pages = "144--147",

journal = "IEICE Transactions on Information and Systems",

issn = "0916-8532",

publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",

number = "1",

}

TY - JOUR

T1 - Negative Learning to Prevent Undesirable Misclassification

AU - Egashira, Kazuki

AU - Miyai, Atsuyuki

AU - Yu, Qing

AU - Irie, Go

AU - Aizawa, Kiyoharu

PY - 2024/1

Y1 - 2024/1

N2 - We propose a novel classification problem setting where Undesirable Classes (UCs) are defined for each class. UC is the class you specifically want to avoid misclassifying. To address this setting, we propose a framework to reduce the probabilities for UCs while increasing the probability for a correct class.

AB - We propose a novel classification problem setting where Undesirable Classes (UCs) are defined for each class. UC is the class you specifically want to avoid misclassifying. To address this setting, we propose a framework to reduce the probabilities for UCs while increasing the probability for a correct class.

KW - classification

KW - negative learning

UR - http://www.scopus.com/inward/record.url?scp=85182590776&partnerID=8YFLogxK

U2 - 10.1587/transinf.2023EDL8056

DO - 10.1587/transinf.2023EDL8056

M3 - Article

AN - SCOPUS:85182590776

SN - 0916-8532

VL - E107.D

SP - 144

EP - 147

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

IS - 1

ER -

Negative Learning to Prevent Undesirable Misclassification

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this