Multi-label classification is an appealing and challenging supervised learning problem, where multiple labels, rather than a single label, are associated with an unseen test instance. To remove possible noises in labels and features of high-dimensionality, multi-label dimension reduction has attracted more and more attentions in recent years. The existing methods usually suffer from several problems, such as ignoring label outliers and label correlations. In addition, most of them emphasize on conducting dimension reduction in an unsupervised or supervised way, therefore, unable to utilize the label information or a large amount of unlabeled data to improve the performance. In order to cope with these problems, we propose a novel method termed Robust sEmi-supervised multi-lAbel DimEnsion Reduction, shortly READER. From the viewpoint of empirical risk minimization, READER selects most discriminative features for all the labels in a semi-supervised way. Specifically, the ℓ2,1-norm induced loss function and regularization term make READER robust to the outliers in the data points. READER finds a feature subspace so as to keep originally neighbor instances close and embeds labels into a low-dimensional latent space nonlinearly. To optimize the objective function, an efficient algorithm is developed with convergence property. Extensive empirical studies on real-world datasets demonstrate the superior performance of the proposed method.
Lu SUN
Hokkaido University
Mineichi KUDO
Hokkaido University
Keigo KIMURA
Hokkaido University
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
Lu SUN, Mineichi KUDO, Keigo KIMURA, "READER: Robust Semi-Supervised Multi-Label Dimension Reduction" in IEICE TRANSACTIONS on Information,
vol. E100-D, no. 10, pp. 2597-2604, October 2017, doi: 10.1587/transinf.2017EDP7184.
Abstract: Multi-label classification is an appealing and challenging supervised learning problem, where multiple labels, rather than a single label, are associated with an unseen test instance. To remove possible noises in labels and features of high-dimensionality, multi-label dimension reduction has attracted more and more attentions in recent years. The existing methods usually suffer from several problems, such as ignoring label outliers and label correlations. In addition, most of them emphasize on conducting dimension reduction in an unsupervised or supervised way, therefore, unable to utilize the label information or a large amount of unlabeled data to improve the performance. In order to cope with these problems, we propose a novel method termed Robust sEmi-supervised multi-lAbel DimEnsion Reduction, shortly READER. From the viewpoint of empirical risk minimization, READER selects most discriminative features for all the labels in a semi-supervised way. Specifically, the ℓ2,1-norm induced loss function and regularization term make READER robust to the outliers in the data points. READER finds a feature subspace so as to keep originally neighbor instances close and embeds labels into a low-dimensional latent space nonlinearly. To optimize the objective function, an efficient algorithm is developed with convergence property. Extensive empirical studies on real-world datasets demonstrate the superior performance of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017EDP7184/_p
Copy
@ARTICLE{e100-d_10_2597,
author={Lu SUN, Mineichi KUDO, Keigo KIMURA, },
journal={IEICE TRANSACTIONS on Information},
title={READER: Robust Semi-Supervised Multi-Label Dimension Reduction},
year={2017},
volume={E100-D},
number={10},
pages={2597-2604},
abstract={Multi-label classification is an appealing and challenging supervised learning problem, where multiple labels, rather than a single label, are associated with an unseen test instance. To remove possible noises in labels and features of high-dimensionality, multi-label dimension reduction has attracted more and more attentions in recent years. The existing methods usually suffer from several problems, such as ignoring label outliers and label correlations. In addition, most of them emphasize on conducting dimension reduction in an unsupervised or supervised way, therefore, unable to utilize the label information or a large amount of unlabeled data to improve the performance. In order to cope with these problems, we propose a novel method termed Robust sEmi-supervised multi-lAbel DimEnsion Reduction, shortly READER. From the viewpoint of empirical risk minimization, READER selects most discriminative features for all the labels in a semi-supervised way. Specifically, the ℓ2,1-norm induced loss function and regularization term make READER robust to the outliers in the data points. READER finds a feature subspace so as to keep originally neighbor instances close and embeds labels into a low-dimensional latent space nonlinearly. To optimize the objective function, an efficient algorithm is developed with convergence property. Extensive empirical studies on real-world datasets demonstrate the superior performance of the proposed method.},
keywords={},
doi={10.1587/transinf.2017EDP7184},
ISSN={1745-1361},
month={October},}
Copy
TY - JOUR
TI - READER: Robust Semi-Supervised Multi-Label Dimension Reduction
T2 - IEICE TRANSACTIONS on Information
SP - 2597
EP - 2604
AU - Lu SUN
AU - Mineichi KUDO
AU - Keigo KIMURA
PY - 2017
DO - 10.1587/transinf.2017EDP7184
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2017
AB - Multi-label classification is an appealing and challenging supervised learning problem, where multiple labels, rather than a single label, are associated with an unseen test instance. To remove possible noises in labels and features of high-dimensionality, multi-label dimension reduction has attracted more and more attentions in recent years. The existing methods usually suffer from several problems, such as ignoring label outliers and label correlations. In addition, most of them emphasize on conducting dimension reduction in an unsupervised or supervised way, therefore, unable to utilize the label information or a large amount of unlabeled data to improve the performance. In order to cope with these problems, we propose a novel method termed Robust sEmi-supervised multi-lAbel DimEnsion Reduction, shortly READER. From the viewpoint of empirical risk minimization, READER selects most discriminative features for all the labels in a semi-supervised way. Specifically, the ℓ2,1-norm induced loss function and regularization term make READER robust to the outliers in the data points. READER finds a feature subspace so as to keep originally neighbor instances close and embeds labels into a low-dimensional latent space nonlinearly. To optimize the objective function, an efficient algorithm is developed with convergence property. Extensive empirical studies on real-world datasets demonstrate the superior performance of the proposed method.
ER -