Domain-attention Conditional Wasserstein Distance for Multi-source Domain Adaptation

被引:22
作者
Wu, Hanrui [1 ]
Yan, Yuguang [2 ]
Ng, Michael K. [2 ]
Wu, Qingyao [1 ]
机构
[1] South China Univ Technol, Sch Software Engn, Guangzhou 510000, Peoples R China
[2] Univ Hong Kong, Dept Math, Hong Kong, Peoples R China
基金
中国国家自然科学基金;
关键词
Domain adaptation; multiple sources; optimal transport; attention;
D O I
10.1145/3391229
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Multi-source domain adaptation has received considerable attention due to its effectiveness of leveraging the knowledge from multiple related sources with different distributions to enhance the learning performance. One of the fundamental challenges in multi-source domain adaptation is how to determine the amount of knowledge transferred from each source domain to the target domain. To address this issue, we propose a new algorithm, called Domain-attention Conditional Wasserstein Distance (DCWD), to learn transferred weights for evaluating the relatedness across the source and target domains. In DCWD, we design a new conditional Wasserstein distance objective function by taking the label information into consideration to measure the distance between a given source domain and the target domain. We also develop an attention scheme to compute the transferred weights of different source domains based on their conditional Wasserstein distances to the target domain. After that, the transferred weights can be used to reweight the source data to determine their importance in knowledge transfer. We conduct comprehensive experiments on several real-world data sets, and the results demonstrate the effectiveness and efficiency of the proposed method.
引用
收藏
页数:19
相关论文
共 54 条
[1]  
[Anonymous], 2007, P ACM MM
[2]  
[Anonymous], 2018, ADV NEUR IN
[3]  
[Anonymous], 2010, Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on, DOI DOI 10.1109/CVPR.2010.5539857
[4]  
[Anonymous], 2016, 30 AAAI C ART INT
[5]  
Aytar Y, 2011, IEEE I CONF COMP VIS, P2252, DOI 10.1109/ICCV.2011.6126504
[6]  
Bahdanau D, 2014, 3 INT C LEARN REPR
[7]  
Benamou JD, 2000, NUMER MATH, V84, P375, DOI 10.1007/s002119900117
[8]   Sliced and Radon Wasserstein Barycenters of Measures [J].
Bonneel, Nicolas ;
Rabin, Julien ;
Peyre, Gabriel ;
Pfister, Hanspeter .
JOURNAL OF MATHEMATICAL IMAGING AND VISION, 2015, 51 (01) :22-45
[9]   LIBSVM: A Library for Support Vector Machines [J].
Chang, Chih-Chung ;
Lin, Chih-Jen .
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY, 2011, 2 (03)
[10]   Re-weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation [J].
Chen, Qingchao ;
Liu, Yang ;
Wang, Zhaowen ;
Wassell, Ian ;
Chetty, Kevin .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :7976-7985