Multi-EPL: Accurate multi-source domain adaptation.

Given multiple source datasets with labels, how can we train a target model with no labeled data? Multi-source domain adaptation (MSDA) aims to train a model using multiple source datasets different from a target dataset in the absence of target data labels. MSDA is a crucial problem applicable to m...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Seongmin Lee, Hyunsik Jeon, U Kang
Formato:	article
Lenguaje:	EN
Publicado:	Public Library of Science (PLoS) 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/f8c839edfd0f4b78b96169a1b2fb80cc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f8c839edfd0f4b78b96169a1b2fb80cc
record_format	dspace
spelling	oai:doaj.org-article:f8c839edfd0f4b78b96169a1b2fb80cc2021-12-02T20:15:09ZMulti-EPL: Accurate multi-source domain adaptation.1932-620310.1371/journal.pone.0255754https://doaj.org/article/f8c839edfd0f4b78b96169a1b2fb80cc2021-01-01T00:00:00Zhttps://doi.org/10.1371/journal.pone.0255754https://doaj.org/toc/1932-6203Given multiple source datasets with labels, how can we train a target model with no labeled data? Multi-source domain adaptation (MSDA) aims to train a model using multiple source datasets different from a target dataset in the absence of target data labels. MSDA is a crucial problem applicable to many practical cases where labels for the target data are unavailable due to privacy issues. Existing MSDA frameworks are limited since they align data without considering labels of the features of each domain. They also do not fully utilize the target data without labels and rely on limited feature extraction with a single extractor. In this paper, we propose Multi-EPL, a novel method for MSDA. Multi-EPL exploits label-wise moment matching to align the conditional distributions of the features for the labels, uses pseudolabels for the unavailable target labels, and introduces an ensemble of multiple feature extractors for accurate domain adaptation. Extensive experiments show that Multi-EPL provides the state-of-the-art performance for MSDA tasks in both image domains and text domains, improving the accuracy by up to 13.20%.Seongmin LeeHyunsik JeonU KangPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 16, Iss 8, p e0255754 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Seongmin Lee Hyunsik Jeon U Kang Multi-EPL: Accurate multi-source domain adaptation.
description	Given multiple source datasets with labels, how can we train a target model with no labeled data? Multi-source domain adaptation (MSDA) aims to train a model using multiple source datasets different from a target dataset in the absence of target data labels. MSDA is a crucial problem applicable to many practical cases where labels for the target data are unavailable due to privacy issues. Existing MSDA frameworks are limited since they align data without considering labels of the features of each domain. They also do not fully utilize the target data without labels and rely on limited feature extraction with a single extractor. In this paper, we propose Multi-EPL, a novel method for MSDA. Multi-EPL exploits label-wise moment matching to align the conditional distributions of the features for the labels, uses pseudolabels for the unavailable target labels, and introduces an ensemble of multiple feature extractors for accurate domain adaptation. Extensive experiments show that Multi-EPL provides the state-of-the-art performance for MSDA tasks in both image domains and text domains, improving the accuracy by up to 13.20%.
format	article
author	Seongmin Lee Hyunsik Jeon U Kang
author_facet	Seongmin Lee Hyunsik Jeon U Kang
author_sort	Seongmin Lee
title	Multi-EPL: Accurate multi-source domain adaptation.
title_short	Multi-EPL: Accurate multi-source domain adaptation.
title_full	Multi-EPL: Accurate multi-source domain adaptation.
title_fullStr	Multi-EPL: Accurate multi-source domain adaptation.
title_full_unstemmed	Multi-EPL: Accurate multi-source domain adaptation.
title_sort	multi-epl: accurate multi-source domain adaptation.
publisher	Public Library of Science (PLoS)
publishDate	2021
url	https://doaj.org/article/f8c839edfd0f4b78b96169a1b2fb80cc
work_keys_str_mv	AT seongminlee multieplaccuratemultisourcedomainadaptation AT hyunsikjeon multieplaccuratemultisourcedomainadaptation AT ukang multieplaccuratemultisourcedomainadaptation
_version_	1718374622035968000

Multi-EPL: Accurate multi-source domain adaptation.

Ejemplares similares