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.
Kazuki EGASHIRA
The University of Tokyo
Atsuyuki MIYAI
The University of Tokyo
Qing YU
The University of Tokyo
Go IRIE
Tokyo University of Science
Kiyoharu AIZAWA
The University of Tokyo
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Kazuki EGASHIRA, Atsuyuki MIYAI, Qing YU, Go IRIE, Kiyoharu AIZAWA, "Negative Learning to Prevent Undesirable Misclassification" in IEICE TRANSACTIONS on Information,
vol. E107-D, no. 1, pp. 144-147, January 2024, doi: 10.1587/transinf.2023EDL8056.
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.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023EDL8056/_p
Copy
@ARTICLE{e107-d_1_144,
author={Kazuki EGASHIRA, Atsuyuki MIYAI, Qing YU, Go IRIE, Kiyoharu AIZAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Negative Learning to Prevent Undesirable Misclassification},
year={2024},
volume={E107-D},
number={1},
pages={144-147},
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={},
doi={10.1587/transinf.2023EDL8056},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - Negative Learning to Prevent Undesirable Misclassification
T2 - IEICE TRANSACTIONS on Information
SP - 144
EP - 147
AU - Kazuki EGASHIRA
AU - Atsuyuki MIYAI
AU - Qing YU
AU - Go IRIE
AU - Kiyoharu AIZAWA
PY - 2024
DO - 10.1587/transinf.2023EDL8056
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E107-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2024
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.
ER -