中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
  5. 多媒体计算与图形学团队
Unified Cross-domain Classification via Geometric and Statistical Adaptations

文献类型：期刊论文

作者Liu, Weifeng4; Li, Jinfeng3; Liu, Baodi4; Guan, Weili2; Zhou, Yicong1; Xu, Changsheng5
刊名PATTERN RECOGNITION
出版日期2021-02-01
卷号110页码:9
ISSN号0031-3203
关键词Domain adaptation Statistical adaptation Maximum mean discrepancy (MMD) Geometric adaptation Nystrom method
DOI10.1016/j.patcog.2020.107658
通讯作者Liu, Weifeng(liuwf@upc.edu.cn)
英文摘要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.
WOS关键词REGULARIZATION ; FRAMEWORK ; KERNEL
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]
WOS研究方向Computer Science ; Engineering
语种英语
出版者ELSEVIER SCI LTD
WOS记录号WOS:000585303400011
资助机构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
源URL[http://ir.ia.ac.cn/handle/173211/41649]  
专题自动化研究所_模式识别国家重点实验室_多媒体计算与图形学团队
通讯作者Liu, Weifeng
作者单位1.Univ Macau, Fac Sci & Technol, Macau, Peoples R China
2.Monash Univ, Fac Informat Technol, Clayton Campus, Clayton, Vic, Australia
3.China Univ Petr East China, Coll Oceanog & Space Informat, Beijing, Peoples R China
4.China Univ Petr East China, Coll Control Sci & Engn, Beijing, 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.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。