Unified Cross-domain Classification via Geometric and Statistical Adaptations

doi:10.1016/j.patcog.2020.107658

	Unified Cross-domain Classification via Geometric and Statistical Adaptations
	Liu, Weifeng 1; Li, Jinfeng 2; Liu, Baodi 1; Guan, Weili 3; Zhou, Yicong 4; Xu, Changsheng5
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
	2021-02-01
卷号	110 页码:9
通讯作者	Liu, Weifeng(liuwf@upc.edu.cn)
摘要	Domain adaptation aims to learn an adaptive classifier for target data using the labelled source data from a different distribution. Most proposed works construct cross-domain classifier by exploring one-sided property of the input data, i.e., either geometric or statistical property. Therefore they may ignore the complementarity between the two properties. Moreover, many previous methods implement knowledge transfer with two separated steps: divergence minimization and classifier construction, which degrades the adaptation robustness. In order to address such problems, we propose a unified cross-domain classi-fication method via geometric and statistical adaptations (UCGS). UCGS models the divergence minimization and classifier construction in a unified way based on structural risk minimization principle and coupled adaptations theory. Specifically, UCGS constructs an adaptive model by simultaneously minimizing the structural risk on labelled source data, using Maximum Mean Discrepancy (MMD) criterion to implement statistical adaptation, and flexibly employing the Nystrom method to explore the geometric connections between domains. A domain-invariant graph is successfully constructed to link the two domains geometrically. The standard supervised methods can be used to instantiate UCGS to handle inter-domain classification problems. Comprehensive experiments show the superiority of UCGS on several real-world datasets. (c) 2020 Elsevier Ltd. All rights reserved.
关键词	Domain adaptation Statistical adaptation Maximum mean discrepancy (MMD) Geometric adaptation Nystrom method
DOI	10.1016/j.patcog.2020.107658
关键词[WOS]	REGULARIZATION ; FRAMEWORK ; KERNEL
收录类别	SCI
语种	英语
资助项目	Major Scientific and Technological Projects of CNPC[ZD2019-183-008] ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)[2020 0 0009] ; Fundamental Research Funds for the Central Universities[20CX05004A]
项目资助者	Major Scientific and Technological Projects of CNPC ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR) ; Fundamental Research Funds for the Central Universities
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000585303400011
出版者	ELSEVIER SCI LTD
引用统计	被引频次：23[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/41650
专题	多模态人工智能系统全国重点实验室_多媒体计算
通讯作者	Liu, Weifeng
作者单位	1.China Univ Petr East China, Coll Control Sci & Engn, Beijing, Peoples R China 2.China Univ Petr East China, Coll Oceanog & Space Informat, Beijing, Peoples R China 3.Monash Univ, Fac Informat Technol, Clayton Campus, Clayton, Vic, Australia 4.Univ Macau, Fac Sci & Technol, Macau, Peoples R China 5.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Liu, Weifeng,Li, Jinfeng,Liu, Baodi,et al. Unified Cross-domain Classification via Geometric and Statistical Adaptations[J]. PATTERN RECOGNITION,2021,110:9.
APA	Liu, Weifeng,Li, Jinfeng,Liu, Baodi,Guan, Weili,Zhou, Yicong,&Xu, Changsheng.(2021).Unified Cross-domain Classification via Geometric and Statistical Adaptations.PATTERN RECOGNITION,110,9.
MLA	Liu, Weifeng,et al."Unified Cross-domain Classification via Geometric and Statistical Adaptations".PATTERN RECOGNITION 110(2021):9.