Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation

In this paper, we focus the problem of unsupervised domain adaptation which transfers knowledge from a well-labeled source domain to an unlabeled target domain with distinctive distributions. Based on Gromov-Hausdorff’s theory, we proposed two kinds of feature mappings in the model of joi...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Wenjuan Ren, Shie Zhou, Zhanpeng Yang, Quan Shi, Xian Sun, Luyi Yang
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Domain adaptation metric information matrix maximum mean discrepancy Toeplitz matrix convolutional filter mask Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/16d7b5987e864a538b69a0f7c9ea5d1e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:16d7b5987e864a538b69a0f7c9ea5d1e
record_format	dspace
spelling	oai:doaj.org-article:16d7b5987e864a538b69a0f7c9ea5d1e2021-11-18T00:11:24ZMetric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation2169-353610.1109/ACCESS.2021.3123281https://doaj.org/article/16d7b5987e864a538b69a0f7c9ea5d1e2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9585694/https://doaj.org/toc/2169-3536In this paper, we focus the problem of unsupervised domain adaptation which transfers knowledge from a well-labeled source domain to an unlabeled target domain with distinctive distributions. Based on Gromov-Hausdorff’s theory, we proposed two kinds of feature mappings in the model of joint distribution adaptation by embedding the original feature subspace to a common subspace. It can been seen as a part of feature embedding used for the models based feature alignment. Our experiments show that constructed mappings have the abilities to alleviate the feature discrepancy and mitigate the distribution shift between source domain and target domains.Wenjuan RenShie ZhouZhanpeng YangQuan ShiXian SunLuyi YangIEEEarticleDomain adaptationmetric information matrixmaximum mean discrepancyToeplitz matrixconvolutional filter maskElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 148017-148023 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Domain adaptation metric information matrix maximum mean discrepancy Toeplitz matrix convolutional filter mask Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Domain adaptation metric information matrix maximum mean discrepancy Toeplitz matrix convolutional filter mask Electrical engineering. Electronics. Nuclear engineering TK1-9971 Wenjuan Ren Shie Zhou Zhanpeng Yang Quan Shi Xian Sun Luyi Yang Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
description	In this paper, we focus the problem of unsupervised domain adaptation which transfers knowledge from a well-labeled source domain to an unlabeled target domain with distinctive distributions. Based on Gromov-Hausdorff’s theory, we proposed two kinds of feature mappings in the model of joint distribution adaptation by embedding the original feature subspace to a common subspace. It can been seen as a part of feature embedding used for the models based feature alignment. Our experiments show that constructed mappings have the abilities to alleviate the feature discrepancy and mitigate the distribution shift between source domain and target domains.
format	article
author	Wenjuan Ren Shie Zhou Zhanpeng Yang Quan Shi Xian Sun Luyi Yang
author_facet	Wenjuan Ren Shie Zhou Zhanpeng Yang Quan Shi Xian Sun Luyi Yang
author_sort	Wenjuan Ren
title	Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
title_short	Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
title_full	Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
title_fullStr	Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
title_full_unstemmed	Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation
title_sort	metric information matrix for maximum mean discrepancy for domain adaptation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/16d7b5987e864a538b69a0f7c9ea5d1e
work_keys_str_mv	AT wenjuanren metricinformationmatrixformaximummeandiscrepancyfordomainadaptation AT shiezhou metricinformationmatrixformaximummeandiscrepancyfordomainadaptation AT zhanpengyang metricinformationmatrixformaximummeandiscrepancyfordomainadaptation AT quanshi metricinformationmatrixformaximummeandiscrepancyfordomainadaptation AT xiansun metricinformationmatrixformaximummeandiscrepancyfordomainadaptation AT luyiyang metricinformationmatrixformaximummeandiscrepancyfordomainadaptation
_version_	1718425146027409408

Metric Information Matrix for Maximum Mean Discrepancy for Domain Adaptation

Ejemplares similares