Generalization on Unseen Domains via Model-Agnostic Learning for Intelligent Fault Diagnosis

doi:10.1109/TIM.2022.3152316

CORC > 自动化研究所 > 中国科学院自动化研究所 > 综合信息系统研究中心 > 工业智能技术与系统

	Generalization on Unseen Domains via Model-Agnostic Learning for Intelligent Fault Diagnosis
	Wang, Huanjie 1,2; Bai, Xiwei 1; Wang, Sihan 3; Tan, Jie 1,2; Liu, Chengbao 1
刊名	IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
	2022
卷号	71 页码:11
关键词	Fault diagnosis Data models Task analysis Representation learning Adaptation models Training data Training Convolutional neural network (CNN) data-driven fault diagnosis domain generalization (DG) model-agnostic learning rolling bearing
ISSN号	0018-9456
DOI	10.1109/TIM.2022.3152316
通讯作者	Tan, Jie(jie.tan@ia.ac.cn)
英文摘要	Machine learning-based diagnosis methods have achieved remarkable success under the assumption that the training and test data are identically distributed. However, a critical requirement of these methods is the generalization capability to unseen domains when deployed to actual diagnosis scenarios. We introduce the challenging problem of domain generalization, i.e., learning from multiple source domains to produce a model that can directly generalize to unseen domains without target information. We adopt a model-agnostic learning produce that maximizes the dot product of gradients between the source domains. Such a gradient alignment objective encourages finding a common optimization path for all source domains, which helps to focus on invariant representations. Furthermore, we propose two feature regularizations that explicitly regularize the feature space. Global feature regularization aligns class relationships between different domains to preserve the domain-invariant knowledge. Local feature regularization encourages the model to learn domain-agnostic class-specific representations with intraclass compactness and interclass separability. The effectiveness of the proposed method is demonstrated with generalization experiments on two benchmarks.
资助项目	National Key Research and Development Program of China[2018YFB1703401] ; National Nature Science Foundation of China[62003344] ; National Nature Science Foundation of China[U1801263]
WOS关键词	NETWORK ; KERNEL
WOS研究方向	Engineering ; Instruments & Instrumentation
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000766618900020
资助机构	National Key Research and Development Program of China ; National Nature Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/48147]
专题	综合信息系统研究中心_工业智能技术与系统
通讯作者	Tan, Jie
作者单位	1.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Northeastern Univ, Sch Informat Sci & Engn, Shenyang 110819, Peoples R China
推荐引用方式 GB/T 7714	Wang, Huanjie,Bai, Xiwei,Wang, Sihan,et al. Generalization on Unseen Domains via Model-Agnostic Learning for Intelligent Fault Diagnosis[J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2022,71:11.
APA	Wang, Huanjie,Bai, Xiwei,Wang, Sihan,Tan, Jie,&Liu, Chengbao.(2022).Generalization on Unseen Domains via Model-Agnostic Learning for Intelligent Fault Diagnosis.IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,71,11.
MLA	Wang, Huanjie,et al."Generalization on Unseen Domains via Model-Agnostic Learning for Intelligent Fault Diagnosis".IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 71(2022):11.