Cross-domain activity recognition via substructural optimal transport

doi:10.1016/j.neucom.2021.04.124

	Cross-domain activity recognition via substructural optimal transport
	Lu, Wang 2,3; Chen, Yiqiang 2,3; Wang, Jindong 1; Qin, Xin 2,3
刊名	NEUROCOMPUTING
	2021-09-24
卷号	454 页码:65-75
关键词	Ubiquitous computing Transfer learning Domain adaptation Optimal transport Clustering
ISSN号	0925-2312
DOI	10.1016/j.neucom.2021.04.124
英文摘要	It is expensive and time-consuming to collect sufficient labeled data for human activity recognition (HAR). Domain adaptation is a promising approach for cross-domain activity recognition. Existing methods mainly focus on adapting cross-domain representations via domain-level, class-level, or sample-level distribution matching. However, they might fail to capture the fine-grained locality information in activity data. The domain-and class-level matching are too coarse that may result in under-adaptation, while sample-level matching may be affected by the noise seriously and eventually cause over-adaptation. In this paper, we propose substructure-level matching for domain adaptation (SSDA) to better utilize the locality information of activity data for accurate and efficient knowledge transfer. Based on SSDA, we propose an optimal transport-based implementation, Substructural Optimal Transport (SOT), for cross domain HAR. We obtain the substructures of activities via clustering methods and seeks the coupling of the weighted substructures between different domains. We conduct comprehensive experiments on four public activity recognition datasets (i.e. UCI-DSADS, UCI-HAR, USC-HAD, PAMAP2), which demonstrates that SOT significantly outperforms other state-of-the-art methods w.r.t classification accuracy (9%+ improvement). In addition, SOT is 5x faster than traditional OT-based DA methods with the same hyper-parameters. (c) 2021 Elsevier B.V. All rights reserved.
资助项目	KeyArea Research and Development Program of Guangdong Province[2019B010109001] ; National Natural Science Foundation of China[61972383]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000672469900007
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/17526]
专题	中国科学院计算技术研究所
通讯作者	Chen, Yiqiang
作者单位	1.Microsoft Res Asia, Beijing, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Lu, Wang,Chen, Yiqiang,Wang, Jindong,et al. Cross-domain activity recognition via substructural optimal transport[J]. NEUROCOMPUTING,2021,454:65-75.
APA	Lu, Wang,Chen, Yiqiang,Wang, Jindong,&Qin, Xin.(2021).Cross-domain activity recognition via substructural optimal transport.NEUROCOMPUTING,454,65-75.
MLA	Lu, Wang,et al."Cross-domain activity recognition via substructural optimal transport".NEUROCOMPUTING 454(2021):65-75.