Constraining pseudo-label in self-training unsupervised domain adaptation with energy-based model

doi:10.1002/int.22930

CORC > 长春光学精密机械与物理研究所 > 中国科学院长春光学精密机械与物理研究所

	Constraining pseudo-label in self-training unsupervised domain adaptation with energy-based model
	L. S. Kong; B. Hu; X. C. Liu; J. Lu; J. You and X. F. Liu
刊名	International Journal of Intelligent Systems
	2022
卷号	37 期号:10 页码:8092-8112
ISSN号	0884-8173
DOI	10.1002/int.22930
英文摘要	Deep learning is usually data starved, and the unsupervised domain adaptation (UDA) is developed to introduce the knowledge in the labeled source domain to the unlabeled target domain. Recently, deep self-training presents a powerful means for UDA, involving an iterative process of predicting the target domain and then taking the confident predictions as hard pseudo-labels for retraining. However, the pseudo-labels are usually unreliable, thus easily leading to deviated solutions with propagated errors. In this paper, we resort to the energy-based model and constrain the training of the unlabeled target sample with an energy function minimization objective. It can be achieved via a simple additional regularization or an energy-based loss. This framework allows us to gain the benefits of the energy-based model, while retaining strong discriminative performance following a plug-and-play fashion. The convergence property and its connection with classification expectation minimization are investigated. We deliver extensive experiments on the most popular and large-scale UDA benchmarks of image classification as well as semantic segmentation to demonstrate its generality and effectiveness.
URL标识	查看原文
语种	英语
内容类型	期刊论文
源URL	[http://ir.ciomp.ac.cn/handle/181722/66484]
专题	中国科学院长春光学精密机械与物理研究所
推荐引用方式 GB/T 7714	L. S. Kong,B. Hu,X. C. Liu,et al. Constraining pseudo-label in self-training unsupervised domain adaptation with energy-based model[J]. International Journal of Intelligent Systems,2022,37(10):8092-8112.
APA	L. S. Kong,B. Hu,X. C. Liu,J. Lu,&J. You and X. F. Liu.(2022).Constraining pseudo-label in self-training unsupervised domain adaptation with energy-based model.International Journal of Intelligent Systems,37(10),8092-8112.
MLA	L. S. Kong,et al."Constraining pseudo-label in self-training unsupervised domain adaptation with energy-based model".International Journal of Intelligent Systems 37.10(2022):8092-8112.