Domain-invariant Graph for Adaptive Semi-supervised Domain Adaptation

doi:10.1145/3487194

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 多媒体计算与图形学团队

	Domain-invariant Graph for Adaptive Semi-supervised Domain Adaptation
	Li, Jinfeng 1; Liu, Weifeng 1; Zhou, Yicong 5; Yu, Jun 4; Tao, Dapeng 3; Xu, Changsheng 2
刊名	ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
	2022-08-01
卷号	18 期号:3 页码:18
关键词	Domain adaptation domain-invariant graph the Nystrom method few labeled source samples
ISSN号	1551-6857
DOI	10.1145/3487194
通讯作者	Li, Jinfeng(lijinfeng_stu@163.com)
英文摘要	Domain adaptation aims to generalize a model from a source domain to tackle tasks in a related but different target domain. Traditional domain adaptation algorithms assume that enough labeled data, which are treated as the prior knowledge are available in the source domain. However, these algorithms will be infeasible when only a few labeled data exist in the source domain, thus the performance decreases significantly. To address this challenge, we propose a Domain-invariant Graph Learning (DGL) approach for domain adaptation with only a few labeled source samples. Firstly, DGL introduces the Nystrom method to construct a plastic graph that shares similar geometric property with the target domain. Then, DGL flexibly employs the Nystrom approximation error to measure the divergence between the plastic graph and source graph to formalize the distribution mismatch from the geometric perspective. Through minimizing the approximation error, DGL learns a domain-invariant geometric graph to bridge the source and target domains. Finally, we integrate the learned domain-invariant graph with the semi-supervised learning and further propose an adaptive semi-supervised model to handle the cross-domain problems. The results of extensive experiments on popular datasets verify the superiority of DGL, especially when only a few labeled source samples are available.
资助项目	National Natural Science Foundation of China[61671480] ; National Natural Science Foundation of China[61836002] ; National Natural Science Foundation of China[62020106007] ; Major Scientific and Technological Projects of CNPC[ZD2019-183-008] ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)[202000009]
WOS关键词	FRAMEWORK ; FEATURES ; KERNEL ; REGULARIZATION ; MATRIX
WOS研究方向	Computer Science
语种	英语
出版者	ASSOC COMPUTING MACHINERY
WOS记录号	WOS:000772650600006
资助机构	National Natural Science Foundation of China ; Major Scientific and Technological Projects of CNPC ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/48201]
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	Li, Jinfeng
作者单位	1.Xidian Univ, China Univ Petr East China, State Key Lab Integrated Serv Networks, 66 Changjiang West Rd, Qingdao 266580, Peoples R China 2.Chinese Acad Sci, Inst Automat, 95 Zhongguancun East Rd, Beijing 100190, Peoples R China 3.Yunnan Univ, Kunming 650091, Yunnan, Peoples R China 4.Hangzhou Dianzi Univ, 1158 2 Dajie, Hangzhou 310018, Peoples R China 5.Univ Macau, Macau, Peoples R China
推荐引用方式 GB/T 7714	Li, Jinfeng,Liu, Weifeng,Zhou, Yicong,et al. Domain-invariant Graph for Adaptive Semi-supervised Domain Adaptation[J]. ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,2022,18(3):18.
APA	Li, Jinfeng,Liu, Weifeng,Zhou, Yicong,Yu, Jun,Tao, Dapeng,&Xu, Changsheng.(2022).Domain-invariant Graph for Adaptive Semi-supervised Domain Adaptation.ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,18(3),18.
MLA	Li, Jinfeng,et al."Domain-invariant Graph for Adaptive Semi-supervised Domain Adaptation".ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS 18.3(2022):18.