Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator

Active learning aims to select the most valuable unlabelled samples for annotation. In this paper, we propose a redundancy removal adversarial active learning (RRAAL) method based on norm online uncertainty indicator, which selects samples based on their distribution, uncertainty, and redundancy. RR...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jifeng Guo, Zhiqi Pang, Wenbo Sun, Shi Li, Yu Chen
Formato:	article
Lenguaje:	EN
Publicado:	Hindawi Limited 2021
Materias:	Computer applications to medicine. Medical informatics R858-859.7 Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
Acceso en línea:	https://doaj.org/article/e68b43bc908a4b6e9c50c770dd863947
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e68b43bc908a4b6e9c50c770dd863947
record_format	dspace
spelling	oai:doaj.org-article:e68b43bc908a4b6e9c50c770dd8639472021-11-08T02:35:55ZRedundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator1687-527310.1155/2021/4752568https://doaj.org/article/e68b43bc908a4b6e9c50c770dd8639472021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/4752568https://doaj.org/toc/1687-5273Active learning aims to select the most valuable unlabelled samples for annotation. In this paper, we propose a redundancy removal adversarial active learning (RRAAL) method based on norm online uncertainty indicator, which selects samples based on their distribution, uncertainty, and redundancy. RRAAL includes a representation generator, state discriminator, and redundancy removal module (RRM). The purpose of the representation generator is to learn the feature representation of a sample, and the state discriminator predicts the state of the feature vector after concatenation. We added a sample discriminator to the representation generator to improve the representation learning ability of the generator and designed a norm online uncertainty indicator (Norm-OUI) to provide a more accurate uncertainty score for the state discriminator. In addition, we designed an RRM based on a greedy algorithm to reduce the number of redundant samples in the labelled pool. The experimental results on four datasets show that the state discriminator, Norm-OUI, and RRM can improve the performance of RRAAL, and RRAAL outperforms the previous state-of-the-art active learning methods.Jifeng GuoZhiqi PangWenbo SunShi LiYu ChenHindawi LimitedarticleComputer applications to medicine. Medical informaticsR858-859.7Neurosciences. Biological psychiatry. NeuropsychiatryRC321-571ENComputational Intelligence and Neuroscience, Vol 2021 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Computer applications to medicine. Medical informatics R858-859.7 Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
spellingShingle	Computer applications to medicine. Medical informatics R858-859.7 Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571 Jifeng Guo Zhiqi Pang Wenbo Sun Shi Li Yu Chen Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
description	Active learning aims to select the most valuable unlabelled samples for annotation. In this paper, we propose a redundancy removal adversarial active learning (RRAAL) method based on norm online uncertainty indicator, which selects samples based on their distribution, uncertainty, and redundancy. RRAAL includes a representation generator, state discriminator, and redundancy removal module (RRM). The purpose of the representation generator is to learn the feature representation of a sample, and the state discriminator predicts the state of the feature vector after concatenation. We added a sample discriminator to the representation generator to improve the representation learning ability of the generator and designed a norm online uncertainty indicator (Norm-OUI) to provide a more accurate uncertainty score for the state discriminator. In addition, we designed an RRM based on a greedy algorithm to reduce the number of redundant samples in the labelled pool. The experimental results on four datasets show that the state discriminator, Norm-OUI, and RRM can improve the performance of RRAAL, and RRAAL outperforms the previous state-of-the-art active learning methods.
format	article
author	Jifeng Guo Zhiqi Pang Wenbo Sun Shi Li Yu Chen
author_facet	Jifeng Guo Zhiqi Pang Wenbo Sun Shi Li Yu Chen
author_sort	Jifeng Guo
title	Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
title_short	Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
title_full	Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
title_fullStr	Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
title_full_unstemmed	Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator
title_sort	redundancy removal adversarial active learning based on norm online uncertainty indicator
publisher	Hindawi Limited
publishDate	2021
url	https://doaj.org/article/e68b43bc908a4b6e9c50c770dd863947
work_keys_str_mv	AT jifengguo redundancyremovaladversarialactivelearningbasedonnormonlineuncertaintyindicator AT zhiqipang redundancyremovaladversarialactivelearningbasedonnormonlineuncertaintyindicator AT wenbosun redundancyremovaladversarialactivelearningbasedonnormonlineuncertaintyindicator AT shili redundancyremovaladversarialactivelearningbasedonnormonlineuncertaintyindicator AT yuchen redundancyremovaladversarialactivelearningbasedonnormonlineuncertaintyindicator
_version_	1718443241204875264

Redundancy Removal Adversarial Active Learning Based on Norm Online Uncertainty Indicator

Ejemplares similares