Class conditional distribution alignment for domain adaptation

	Class conditional distribution alignment for domain adaptation
	Kai CAO 1; Zhipeng TU 1; Yang MING 1
刊名	控制理论与技术:英文版
	2020
卷号	18.0 期号:1.0 页码:72-80
关键词	Domain adaptation distribution alignment feature cluster
ISSN号	2095-6983
英文摘要	In this paper,we study the problem of domain adaptation,which is a crucial ingredient in transfer learning with two domains,that is,the source domain with labeled data and the target domain with none or few labels.Domain adaptation aims to extract knowledge from the source domain to improve the performance of the learning task in the target domain.A popular approach to handle this problem is via adversarial training,which is explained by the H△H-distance theory.However,traditional adversarial network architectures just align the marginal feature distribution in the feature space.The alignment of class condition distribution is not guaranteed.Therefore,we proposed a novel method based on pseudo labels and the cluster assumption to avoid the incorrect class alignment in the feature space.The experiments demonstrate that our framework improves the accuracy on typical transfer learning tasks.
语种	中文
CSCD记录号	CSCD:6709874
内容类型	期刊论文
源URL	[http://ir.amss.ac.cn/handle/2S8OKBNM/56125]
专题	中国科学院数学与系统科学研究院
作者单位	1.Key Lab of Systems and Control,Academy of Mathematics and Systems Science,Chinese Academy of Sciences 2.中国科学院大学
推荐引用方式 GB/T 7714	Kai CAO,Zhipeng TU,Yang MING. Class conditional distribution alignment for domain adaptation[J]. 控制理论与技术:英文版,2020,18.0(1.0):72-80.
APA	Kai CAO,Zhipeng TU,&Yang MING.(2020).Class conditional distribution alignment for domain adaptation.控制理论与技术:英文版,18.0(1.0),72-80.
MLA	Kai CAO,et al."Class conditional distribution alignment for domain adaptation".控制理论与技术:英文版 18.0.1.0(2020):72-80.