基于域不变判别特征学习的深子域自适应方法

肖铠祥; 吴晓鸰; 冯永晋; Hoon Heo

doi:10.12052/gdutxb.240096

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于域不变判别特征学习的深子域自适应方法

计算机科学与技术 | 更新时间：2026-05-27

- 基于域不变判别特征学习的深子域自适应方法
- Deep Subdomain Adaptation Method Based on Domain Invariant Discriminant Feature Learning
- 广东工业大学学报 2025年42卷第4期页码：59-70
- 作者机构：
  
  1. 广东工业大学计算机学院,广东,广州,510006
  2. 三星电机有限公司,京畿道,水原市,16674
- 作者简介：
  
  吴晓鸰(1979–) ，女，副教授，博士，主要研究方向为信息安全、物联网，E-mail：xl.wu@gdut.edu.cn
- 基金信息：
  
  广东省重点领域研发计划项目(2019B010139002);广东省国际科技合作领域项目(2019A050513010)
- DOI：10.12052/gdutxb.240096
  中图分类号： TP183
- 纸质出版：2025
- 稿件说明：
移动端阅览
肖铠祥, 吴晓鸰, 冯永晋, 等. 基于域不变判别特征学习的深子域自适应方法[J]. 广东工业大学学报, 2025, 42(4): 59–70.

Xiao Kaixiang, Wu Xiaoling, Feng Yongjin, et al. Deep Subdomain Adaptation Method Based on Domain Invariant Discriminant Feature Learning[J]. 2025, 42(4): 59-70.
肖铠祥, 吴晓鸰, 冯永晋, 等. 基于域不变判别特征学习的深子域自适应方法[J]. 广东工业大学学报, 2025, 42(4): 59–70. DOI： 10.12052/gdutxb.240096

Xiao Kaixiang, Wu Xiaoling, Feng Yongjin, et al. Deep Subdomain Adaptation Method Based on Domain Invariant Discriminant Feature Learning[J]. 2025, 42(4): 59-70. DOI： 10.12052/gdutxb.240096.

摘要

无监督域自适应旨在将源领域已标记数据的知识迁移到无标记的目标领域。基于域对齐的无监督域自适应方法通过最小化跨域特征分布差异来学习域不变特征，忽略了目标域判别特征的学习，造成学习到的域不变特征在目标域中判别性差，不同类特征易混淆，导致模型性能下降。为了解决域对齐过程中域不变特征在目标域判别性差的问题，本文提出了一种基于域不变判别特征学习的深子域自适应方法(deep Subdomain Adaptation method based on domain invariant Discriminant Feature learning

DFSA)，结合深子域自适应方法与最小化类混淆约束来增加域不变特征的类间差异，并添加一致性正则化来减少域不变特征的类内差异，通过以上组合挖掘域不变判别特征，以更好地保留域不变特征在目标域的判别性，提高域自适应能力。在Office-31和Office-Home基准数据集上验证了DFSA的有效性，结果表明DFSA有效提高了域自适应性能，比现有的深子域自适应方法在2个基准数据集上的平均准确率分别提升了0.79个百分点和3.51个百分点。

Abstract

Unsupervised domain adaptation aims to transfer knowledge from labeled data in the source domain to the unlabeled target domain. The unsupervised domain adaptation method based on domain alignment learns domain invariant features by minimizing the difference in cross domain feature distribution

ignoring the learning of discriminative features in the target domain

resulting in poor discriminability of the learned domain invariant features in the target domain

confusion of different types of features

and performance decrease of the model. To address the problem of poor discriminative power of domain invariant features in the target domain during domain alignment

this paper proposes a deep Subdomain Adaptive method based on domain invariant Discriminative Feature learning (DFSA). The deep subdomain adaptive method is combined with the constraints of minimizing class confusion to increase the inter-class differences of domain invariant features

and consistency regularization is added to reduce the intra-class differences of domain invariant features. By doing so

domain invariant discriminative features are mined to better preserve the discriminative power of domain invariant features in the target domain and improve domain adaptability. The effectiveness of DFSA was validated on the Office-31 and Office-Home benchmark datasets

and the results showed that DFSA effectively improves domain adaptation performance

with an average accuracy improvement of approximately 0.79 percentage points and 3.51 percentage points

respectively

when compared with existing deep subdomain adaptation methods on the two benchmark datasets.

关键词

Keywords

references

ZHANG Y C, LIU T L, LONG M S, et al . Bridging theory and algorithm for domain adaptation[C ] //36th International Conference on Machine Learning (ICML). Long Beach: IEEE, 2019: 7404-7413.

SI L, DONG H, QIANG W, et al . Regularized hypothesis-induced wasserstein divergence for unsupervised domain adaptation[J ] . Knowledge-Based Systems, 2024, 283: 111162.

LI Y, ZHU Y, YING S. Attention-guided optimal transport for unsupervised domain adaptation with class structure prior[J]. Neural Processing Letters, 2023, 55(9): 12547-12567.

LONG M, ZHU H, WANG J, et al . Deep transfer learning with joint adaptation networks[C ] //34th International Conference on Machine Learning(ICML). Sydney: IEEE, 2017: 2208-2217.

XIE B, YUAN L, LI S, et al . Active learning for domain adaptation: an energy-based approach[J ] . Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(8): 8708-8716.

ZHU Y, ZHUANG F, WANG D. Aligning domain-specific distribution and classifier for cross-domain classification from multiple sources[C]//Proceedings of the AAAI Conference on Artificial Intelligence(AAAI). Honolulu: AAAI, 2019: 5989-5996.

WANG Q, MENG F, BRECKON T P. Data augmentation with norm-AE and selective pseudo-labelling for unsupervised domain adaptation[J]. Neural Networks, 2023, 161: 614-625.

GANIN Y, USTINOVA E, AJAKAN H, et al . Domain-adversarial training of neural networks[J ] . Journal of Machine Learning Research, 2016, 17(59): 1-35.

TZENG E, HOFFMAN J, SAENKO K, et al . Adversarial discriminative domain adaptation[C ] //30th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE, 2017: 2962-2971.

GAO Z, ZHANG S, HUANG K, et al . Gradient distribution alignment certificates better adversarial domain adaptation[C ] //2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal: IEEE, 2021: 8917-8926.

GHIFARY M, BALDUZZI D, KLEIJN W B, et al . Scatter component analysis: a unified framework for domain adaptation and domain generalization[J ] . Journal of IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39: 1414-1430.

SHUANG L, SHIJI S, GAO H, et al . Domain invariant and class discriminative feature learning for visual domain adaptation[J ] . IEEE Trans Image Process, 2018, 27: 4260-4273.

DAN J, JIN T, CHI H, et al . Trust-aware conditional adversarial domain adaptation with feature norm alignment[J ] . Neural Networks, 2023, 168: 518-530.

ZHU Y, ZHUANG F, WANG J, et al . Deep subdomain adaptation network for image classification[J ] . IEEE Transactions on Neural Networks and Learning Systems, 2020, 32(4): 1713-1722.

WANG P, YANG Y, XIA Y, et al . Information maximizing adaptation network with label distribution priors for unsupervised do main adaptation[J ] . IEEE Transactions on Multimedia, 2023, 25: 6026-6039.

ZHAO F, LIU W, WEN C. A new method of image classification based on domain adaptation[J]. Sensors, 2022, 22(4): 1315.

DU S L, HUANG B H, LI R P, et al . Domain adaptation based on weighted classification loss and nuclear-norm[J ] . Application Research of Computers, 2023, 40(6): 1734-1738.

CHEN X Y, SUN H. Centroid alignment for unsupervised domain adaptation[J]. Journal of Chinese Computer Systems, 2022, 43(4): 822-827.

GAO J, MA X, XU C. Learning transferable conceptual prototypes for interpretable unsupervised domain adaptation[J]. IEEE Transactions on Image Processing, 2024, 33: 5284-5297.

LUO Y W, REN C X, DAI D Q, et al . Unsupervised domain adaptation via discriminative manifold propagation[J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(3): 1653-1669.

JIN Y, WANG X, LONG M, et al . Minimum class confusion for versatile domain adaptation[C ] //Computer Vision-ECCV 2020: 16th European Conference(ECCV). Glasgow: ECCV, 2020: 464- 480.

SOHN K, BERTHELOT D, CARLINI N, et al . Fixmatch: simplifying semi-supervised learning with consistency and confidence[J ] . Advances in Neural Information Processing Systems, 2020, 33: 596-608.

WANG X, ZHUO J, ZHANG M, et al . Revisiting unsupervised domain adaptation models: a smoothness perspective[C ] //Proceedings of the Asian Conference on Computer Vision(ACCV). Macao: ACCV. 2022: 1504-1521.

CUBUK E D, ZOPH B, SHLENS J, et al . Randaugment: practical automated data augmentation with a reduced search space[C ] //IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 2020: 3008-3017.

SAENKO K, KULIS B, FRITZ M, et al . Adapting visual category models to new domains[C ] //11th European Conference on Computer Vision(ECCV). Heraklion: ECCV, 2010: 213-226.

VENKATESWARA H, EUSEBIO J, CHAKRABORTY S, et al . Deep hashing network for unsupervised domain adaptation[C ] //30th IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE, 2017: 5385-5394.

HE K, ZHANG X, REN S , et al . Deep residual learning for image recognition[C ] //Proceedings of the IEEE Conference on Computer vision and Pattern Recognition(CVPR). Las Vegas: IEEE, 2016: 770-778.

LONG M S, CAO Z J, WANG J M, et al . Conditional adversarial domain adaptation[C ] //32nd Conference on Neural Information Processing Systems (NIPS). Montreal: NIPS, 2018: 1640-1650.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向电力场景图像的动态自适应增量目标检测DA-IOD算法

基于主动学习和视觉状态空间模型的热点检测器

基于深度学习的自适应采样及记忆增强压缩感知方法

融合沟道倾斜角的3D NAND存储单元建模及编程策略优化

双模态迭代交叉注意力融合集成框架