Attention-Based Multi-Source Domain Adaptation

Yukun Zuo

doi:10.1109/TIP.2021.3065254

2021-01-01

PRE

徐常胜 (Changsheng Xu) *

DOI：10.1109/TIP.2021.3065254

原文链接

原文求助

摘要

En 中文

Multi-source domain adaptation (MSDA) aims to transfer knowledge from multi-source domains to one target domain. Inspired by single-source domain adaptation, existing methods solve MSDA by aligning the data distributions between the target domain and each source domain. However, aligning the target domain with the dissimilar source domain would harm the representation learning. To address the above issue, an intuitive motivation of MSDA is using the attention mechanism to enhance the positive effects of the similar domains, and suppress the negative effects of the dissimilar domains. Therefore, we propose Attention-Based Multi-Source Domain Adaptation (ABMSDA) by considering the domain correlations to alleviate the effects caused by dissimilar domains. To obtain the domain correlations between source and target domains, ABMSDA firstly trains a domain recognition model to calculate the probability that the target images belong to each source domain. Based on the domain correlations, Weighted Moment Distance (WMD) is proposed to pay more attention on the source domains with higher similarities. Furthermore, Attentive Classification Loss (ACL) is developed to constrain that the feature extractor can generate the alignment and discriminative visual representations. The evaluations on two benchmarks demonstrate the effectiveness of the proposed model, e.g., an average of 6.1% improvement on the challenging DomainNet dataset.

Keyword:

Correlation

Adaptation models

Feature extraction

Target recognition

Data models

Transfer learning

Visualization

Multi-source domain adaptation

attention-based multi-source domain adaptation

weighted moment distance

AI总结

对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Transactions on Image Processing

IF:

13.7

论文数:

1.0W

被引数:

8.4W

机构

university of science & technology of china, cas

学者数:

3.2W

论文数: 2.7W

被引数: 74

chinese academy of sciences

学者数:

56.7W

论文数: 45.0W

被引数: 704

引用论文

The mitochondrial complex I inhibitor rotenone triggers a cerebral tauopathy

Journal of Neurochemistry

IF0

2005-10-10

PREAI

Günter U. Höglinger; Annie Lannuzel; Myriam Escobar Khondiker; Patrick P. Michel; Charles Duyckaerts; Jean Féger; Pierre Champy; Annick Prigent; Fadia Medja; Anne Lombes; Wolfgang H. Oertel; Merle Ruberg; Etienne C. Hirsch

Transient Analysis of a Cracked Rotor Passing Through Critical Speed

Journal of Sound and Vibration

IF0

1994-06-01

PREAI

A.S. Sekhar; B.S. Prabhu

Photoactivated Self-Sanitizing Chlorophyllin-Containing Coatings to Prevent Microbial Contamination in Packaged Food

Coatings

IF0

2018-09-19