Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization

At present, Symmetric Positive Definite (SPD) matrix data is the most common non-Euclidean data in machine learning. Because SPD data don’t form a linear space, most machine learning algorithms can not be carried out directly on SPD data. The first purpose of this paper is to propose a ne...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Qian Li, Zhengming Ma, Shuyu Liu, Yanli Pei
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Domain adaptation learning Riemannian manifold bi-subspace learning symmetric positive definite source linear discriminant analysis regularization Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/4dfa2953c2e244aa9f43b97d25fac455
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4dfa2953c2e244aa9f43b97d25fac455
record_format	dspace
spelling	oai:doaj.org-article:4dfa2953c2e244aa9f43b97d25fac4552021-11-09T00:03:05ZDomain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization2169-353610.1109/ACCESS.2021.3123470https://doaj.org/article/4dfa2953c2e244aa9f43b97d25fac4552021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9590537/https://doaj.org/toc/2169-3536At present, Symmetric Positive Definite (SPD) matrix data is the most common non-Euclidean data in machine learning. Because SPD data don’t form a linear space, most machine learning algorithms can not be carried out directly on SPD data. The first purpose of this paper is to propose a new framework of SPD data machine learning, in which SPD data are transformed into the tangent spaces of Riemannian manifold, rather than a Reproducing Kernel Hilbert Space (RKHS) as usual. Domain adaption learning is a kind of machine learning. The second purpose of this paper is to apply the proposed framework to domain adaption learning (DAL), in which the architecture of bi-subspace learning is adopted. Compared with the commonly-used one subspace learning architecture, the proposed architecture provides a broader optimization space to meet the domain adaption criterion. At last, in order to further improve the classification accuracy, Linear Discriminant Analysis (LDA) regularization of source domain data is added. The experimental results on five real-world datasets demonstrate the out-performance of the proposed algorithm over other five related state-of-the-art algorithms.Qian LiZhengming MaShuyu LiuYanli PeiIEEEarticleDomain adaptation learningRiemannian manifoldbi-subspace learningsymmetric positive definitesource linear discriminant analysis regularizationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 146984-147002 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Domain adaptation learning Riemannian manifold bi-subspace learning symmetric positive definite source linear discriminant analysis regularization Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Domain adaptation learning Riemannian manifold bi-subspace learning symmetric positive definite source linear discriminant analysis regularization Electrical engineering. Electronics. Nuclear engineering TK1-9971 Qian Li Zhengming Ma Shuyu Liu Yanli Pei Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
description	At present, Symmetric Positive Definite (SPD) matrix data is the most common non-Euclidean data in machine learning. Because SPD data don’t form a linear space, most machine learning algorithms can not be carried out directly on SPD data. The first purpose of this paper is to propose a new framework of SPD data machine learning, in which SPD data are transformed into the tangent spaces of Riemannian manifold, rather than a Reproducing Kernel Hilbert Space (RKHS) as usual. Domain adaption learning is a kind of machine learning. The second purpose of this paper is to apply the proposed framework to domain adaption learning (DAL), in which the architecture of bi-subspace learning is adopted. Compared with the commonly-used one subspace learning architecture, the proposed architecture provides a broader optimization space to meet the domain adaption criterion. At last, in order to further improve the classification accuracy, Linear Discriminant Analysis (LDA) regularization of source domain data is added. The experimental results on five real-world datasets demonstrate the out-performance of the proposed algorithm over other five related state-of-the-art algorithms.
format	article
author	Qian Li Zhengming Ma Shuyu Liu Yanli Pei
author_facet	Qian Li Zhengming Ma Shuyu Liu Yanli Pei
author_sort	Qian Li
title	Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
title_short	Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
title_full	Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
title_fullStr	Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
title_full_unstemmed	Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization
title_sort	domain adaption based on symmetric matrices space bi-subspace learning and source linear discriminant analysis regularization
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/4dfa2953c2e244aa9f43b97d25fac455
work_keys_str_mv	AT qianli domainadaptionbasedonsymmetricmatricesspacebisubspacelearningandsourcelineardiscriminantanalysisregularization AT zhengmingma domainadaptionbasedonsymmetricmatricesspacebisubspacelearningandsourcelineardiscriminantanalysisregularization AT shuyuliu domainadaptionbasedonsymmetricmatricesspacebisubspacelearningandsourcelineardiscriminantanalysisregularization AT yanlipei domainadaptionbasedonsymmetricmatricesspacebisubspacelearningandsourcelineardiscriminantanalysisregularization
_version_	1718441383201603584

Domain Adaption Based on Symmetric Matrices Space Bi-Subspace Learning and Source Linear Discriminant Analysis Regularization

Ejemplares similares