计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (7): 345-353.doi: 10.11896/jsjkx.230500144

• 人工智能 • 上一篇    下一篇

基于熵值过滤和类质心优化的无监督域适应

田青1,3,4, 卢章虎2, 杨宏2   

  1. 1 南京信息工程大学软件学院 南京 210044
    2 南京信息工程大学计算机学院、网络空间安全学院 南京 210044
    3 南京信息工程大学数字取证教育部工程研究中心 南京 210044
    4 南京大学计算机软件新技术国家重点实验室 南京210023
  • 收稿日期:2023-05-22 修回日期:2023-10-25 出版日期:2024-07-15 发布日期:2024-07-10
  • 通讯作者: 田青(tianqing@nuist.edu.cn)
  • 基金资助:
    国家自然科学基金(62176128);南京大学计算机软件新技术国家重点实验室开放课题(KFKT2022B06);中央高校基本科研基金(NJ2022028);江苏省“青蓝工程”人才计划项目

Unsupervised Domain Adaptation Based on Entropy Filtering and Class Centroid Optimization

TIAN Qing1,3,4, LU Zhanghu2, YANG Hong2   

  1. 1 School of Software,Nanjing University of Information Science and Technology,Nanjing 210044,China
    2 School of Computer Science,Nanjing University of Information Science and Technology,Nanjing 210044,China
    3 Engineering Research Center of Digital Forensics,Ministry of Education,Nanjing University of Information Science and Technology,Nanjing 210044,China
    4 State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210023,China
  • Received:2023-05-22 Revised:2023-10-25 Online:2024-07-15 Published:2024-07-10
  • About author:TIAN Qing,born in 1984,Ph.D,professor.His main research interests include machine learning and pattern recognition.
  • Supported by:
    National Natural Science Foundation of China(62176128),Open Projects Program of State Key Laboratory for Novel Software Technology of Nanjing University(KFKT2022B06),Fundamental Research Funds for the Central Universities(NJ2022028) and Qing-Lan Project of Jiangsu Province of China.

PDF (PC)

381

摘要： 无监督域适应作为机器学习领域的新兴研究方向之一,其主要利用源域监督信息来辅助无标记目标域的学习。截至目前,已有较多无监督域适应方法被提出,但在关系挖掘方面仍存在一些不足之处。具体来说,现有方法通常对目标域样本采取一致性处理策略,而忽略了目标域样本在关系挖掘中的差异性,因此文中提出了熵值过滤和类质心优化方法。所提方法利用生成对抗网络架构对目标域样本进行标记,利用所获伪标签计算样本熵值,并与所设阈值进行比较,从而进一步划分目标域样本。对于简单样本,分配伪标签;对于困难样本,该方法结合对比学习思想,利用源域和简单样本来学习更加鲁棒的分类器对困难样本分类,并进一步获得源和目标域的类质心。通过优化域间和实例对比对齐,来减小域间和域内的差异。最后,在3个标准数据集上与目前几种先进的领域自适应方法进行了对比实验,实验结果表明所提方法的性能均优于对比方法。

关键词: 迁移学习, 无监督域适应, 对抗学习, 对比学习, 类质心优化

Abstract: As one of the emerging research directions in the field of machine learning,unsupervised domainadaptation mainly uses source domain supervision information to assist the learning of unlabeled target domains.Recently,many unsupervised domain adaptation methods have been proposed,but there are still some deficiencies in relation mining.Specifically,existing methods usually adopt a consistent processing strategy for target domain samples,while ignoring the discrepancy in target domain samples in relation mining.Therefore,this paper proposes a novel method called entropy filtering and class centroid optimization(EFCO).The proposed method utilizes a generative adversarial network architecture to label target domain samples.With the obtained pseudo-labels,the sample entropy value is calculated and compared with a predefined threshold to further categorize target domain samples.Simple samples are assigned pseudo-labels,while difficult samples are classified using the idea of contrastive learning.By combining source domain data and simple samples,a more robust classifier is learned to classify difficult samples,and class centroids of the source and target domains are obtained.Inter-domain and intra-domain discrepancies are minimized by optimizing inter-domain contrastive alignment and instance contrastive alignment.Finally,it is compared with several advanced domain adaptation methods on three standard data sets,and the results indicate that the performance of the proposed method outperforms the comparison methods.

Key words: Transfer learning, Unsupervised domain adaptation, Adversarial learning, Contrastive learning, Class centroid optimization

中图分类号: 

  • TP181
[1]SUN B,SAENKO K.Deep coral:Correlation alignment for deep domain adaptation[C]//Proceedings of European Conference on Computer Vision.Springer,2016:443-450.
[2]TIAN Q,SUN H,MA C,et al.Heterogeneous Domain Adaptation With Structure and Classification Space Alignment[J].IEEE Transactions on Cybernetics,2022,52(10):10328-10338.
[3]LONG M,CAO Z,WANG J,et al.Conditional adversarial do-main adaptation[J].Advances in Neural Information Processing Systems,2018,31:1647-1657.
[4]PEI Z,CAO Z,LONG M,et al.Multi-Adversarial domain adaptation[C]//Proceedings of AAAI Conference on Artificial Intelligence.AAAI Press,2018:3934-3941.
[5]GANIN Y,LEMPITSKY V.Unsupervised domain adaptationby backpropagation[C]//Proceedings of International Confe-rence on Machine Learning.PMLR,2015:1180-1189.
[6]TIAN Q,ZHU Y,SUN H,CHEN S,et al.Unsupervised Domain Adaptation Through Dynamically Aligning Both the Feature and Label Spaces[J].IEEE Transactions on Circuits and Systems for Video Technology,2022,32(12):8562-8573.
[7]LONG M,ZHU H,WANG J,et al.Unsupervised domain adaptation with residual transfer networks[J].Advances in Neural Information Processing Systems,2016,29:136-144.
[8]TIAN Q,PENG S,MA T.Source-free Unsupervised DomainAdaptation with Trusted Pseudo Samples[J].ACM Transactions on Intelligent Systems and Technology,2023,14(2):2157-6904.
[9]OORD A,LI Y,VINYALS O.Representation learning with contrastive predictive coding[J].arXiv:1807.03748,2018.
[10]HE K,FAN H,WU Y,et al.Momentum contrast for unsupervised visual representation learning[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2020:9726-9735.
[11]KANG G,JIANG L,YANG Y,et al.Contrastive adaptation network for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2019:4893-4902.
[12]CHEN T,KORNBLITH S,NOROUZI M,et al.A simpleframework for contrastive learning of visual representations[C]//Proceedings of International Conference on Machine Learning.PMLR,2020:1597-1607.
[13]GRILL J B,STRUB F,ALTCHE F,et al.Bootstrap your own latent-a new approach to self-supervised learning[J].Advances in Neural Information Processing Systems,2020,33:21271-21284.
[14]SINGH A,CHAKRABORTY O,VARSHNEY A,et al.Semi-supervised action recognition with temporal contrastive learning[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2021:10389-10399.
[15]XIE S,ZHENG Z,CHEN L,et al.Learning semantic representations for unsupervised domain adaptation[C]//Proceedings of International Conference on Machine Learning.PMLR,2018:5423-5432.
[16]PINHEIRO P O.Unsupervised domain adaptation with similarity learning[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2018:8004-8013.
[17]PAN Y,YAO T,LI Y,et al.Transferrable prototypical net-works for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2019:2239-2247.
[18]YAN H,LI Z,WANG Q,et al.Weighted and class-specific ma-ximum mean discrepancy for unsupervised domain adaptation[J].IEEE Transactions on Multimedia,2019,22(9):2420-2433.
[19]HU L,KAN M,SHAN S,et al.Unsupervised domain adaptation with hierarchical gradient synchronization[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2020:4043-4052.
[20]LU Y,ZHU Q,ZHANG B,et al.Weighted correlation embedding learning for domain adaptation[J].IEEE Transactions on Image Processing,2022,31:5303-5316.
[21]CHEN C,XIE W,HUANG W,et al.Progressive feature alignment for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2019:627-636.
[22]WANG X,LI L,YE W,et al.Transferable attention for domain adaptation[C]//Proceedings of AAAI Conference on Artificial Intelligence.AAAI Press,2019:5345-5352.
[26]ZHANG W,OUYANG W,LI W,et al.Collaborative and adversarial network for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2018:3801-3809.
[24]ZHU Y,ZHUANG F,WANG J,et al.Deep subdomain adapta-tion network for image classification[J].IEEE Transactions on Neural Networks and Learning Systems,2020,32(4):1713-1722.
[25]DENG Z,LUO Y,ZHU J.Cluster alignment with a teacher for unsupervised domain adaptation[C]//Proceedings of IEEE International Conference on Computer Vision.IEEE,2019:9944-9953.
[26]WANG Q,BRECKON T.Unsupervised domain adaptation via structured prediction based selective pseudo-labeling[C]//Proceedings of AAAI Conference on Artificial Intelligence.AAAI Press,2020:6243-6250.
[27]WANG M,DENG W.Cycle label-consistent networks for unsupervised domain adaptation[J].Neurocomputing,2021,422:186-199.
[28]MORERIO P,VOLPI R,RAGONESI R,et al.Generative pseudo-label refinement for unsupervised domain adaptation[C]//Proceedings of IEEE Winter Conference on Applications of Computer Vision.IEEE,2020:3130-3139.
[29]LITRICO M,DEL BUE A,MORERIO P.Guiding Pseudo-La-bels With Uncertainty Estimation for Source-Free Unsupervised Domain Adaptation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2023:7640-7650.
[30]CHU T,LIU Y,DENG J,et al.Denoised Maximum Classifier Discrepancy for Source-Free Unsupervised Domain Adaptation[C]//Proceedings of the AAAI Conference on Artificial Intelligence.AAAI Press,2022:472-480.
[31]TIAN Y,KRISHNAN D,ISOLA P.Contrastive multiview co-ding[C]//Proceedings of European Conference on Computer Vision.Springer,2020:776-794.
[32]KHOSLA P,TETERWAK P,WANG C,et al.Supervised contrastive learning[J].Advances in Neural Information Processing Systems,2020,33:18661-18673.
[33]GE Y X,ZHU F,CHEN D,et al.Self-paced contrastive learning with hybrid memory for domain adaptive object re-id[J].Advances in Neural Information Processing Systems,2020,33:11309-11321.
[34]THOTA M,LEONTIDIS G.Contrastive domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.2021:2209-2218.
[35]LIU W,FERSTL D,SCHULTER S,et al.Domain adaptationfor semantic segmentation via patch-wise contrastive learning[J].arXiv:2104.11056,2021.
[36]GRETTON A,BORGWARDT K M,RASCH M J,et al.A kernel two-sample test[J].Journal of Machine Learning Research,2012,13(1):723-773.
[37]HUANG J,GUAN D,XIAO A,et al.Category contrast for un-supervised domain adaptation in visual tasks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2022:1203-1214.
[38]PAN S,YANG Q.A survey on transfer learning[J].IEEETransactions on Knowledge and Data Engineering,2010,22(10):1345-1359.
[39]HOFFMAN J,TZENG E,PARK T,et al.Cycada:Cycle-con-sistent adversarial domain adaptation[C]//Proceedings of International Conference on Machine Learning.PMLR,2018:1989-1998.
[40]LECUN Y,BOTTOU L,BENGIO Y,et al.Gradient-basedlearning applied to document recognition[J].IEEE,1998,86(11):2278-2324.
[41]HULL J.A database for handwritten text recognition research[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1994,16(5):550-554.
[42]NETZER Y,WANG T,COATES A,et al.Reading digits in na-tural images with unsupervised feature learning[J].NIPS Workshop on Deep Learning & Unsupervised Feature Learning.2011.
[43]SAENLO K,KULIS B,FRITZ M,et al.Adapting visual category models to new domains[C]//Proceedings of European Conference on Computer Vision.Springer,2010:213-226.
[44]VENKATESWARA H,EUSBIO J,CHAKRABORTY S,et al.Deep hashing network for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2017:5018-5027.
[45]GANIN Y,USTINOVA E,AJAKAN H,et al.Domain-adversarial training of neural networks[J].TheJournal of Machine Learning Research,2016,17(1):2096-2030.
[46]TZENG E,HOFFMAN J,SAENKO K,et al.Adversarial dis-criminative domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2017:7167-7176.
[47]LONG M,CAO Y,WANG J,et al.Learning transferable fea-tures with deep adaptation networks[C]//Proceedings of International Conference on Machine Learning.PMLR,2015:97-105.
[48]SAITO K,WATANABE K,USHIKU Y,et al.Maximum classifier discrepancy for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2018:3723-3732.
[49]XIAO N,ZHANG L.Dynamic weighted learning for unsuper-vised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2021:15242-15251.
[50]LIU H,LONG M,WANG J,et al.Transferable adversarialtraining:A general approach to adapting deep classifiers[C]//Proceedings of International Conference on Machine Learning.PMLR,2019:4013-4022.
[51]LONG M,ZHU H,WANG J,et al.Deep transfer learning with joint adaptation networks[C]//Proceedings of International Conference on Machine Learning.PMLR,2017:2208-2217.
[52]LI S,SONG S,WU C.Layer-wise domain correction for unsupervised domain adaptation[J].Frontiers of Information Technology Electronic Engineering,2018,19(1):91-103.
[53]LIU M,TUZEL O.Coupled generative adversarial networks[J].Advances in Neural Information Processing Systems,2016,29:469-477.
[54]XU R,LI G,YANG J,et al.Larger norm more transferable:An adaptive feature norm approach for unsupervised domain adaptation[C]//Proceedings of IEEE International Conference on Computer Vision.IEEE,2019:1426-1435.
[55]LI M,ZHAI Y,LUO Y,et al.Enhanced transport distance for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2020:13936-13944.
[56]DU Z,LI J,SU H,et al.Cross-domain gradient discrepancy mi-nimization for unsupervised domain adaptation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2021:3937-3946.
[57]WANG H,TIAN J,LI S,et al.Structure-conditioned adversarial learning for unsupervised domain adaptation[J].Neurocompu-ting,2022,497:216-226.
[58]HE K,ZHANG X,REN S,et al.Deep residual learning forimage recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer So-ciety,2016:770-778.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发