Skip to main content
Springer Nature Link
Log in

Heterogeneous domain adaptation by class centroid matching and local discriminative structure preservation

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Heterogeneous domain adaptation (HDA) aims at facilitating the target model training by leveraging knowledge from the heterogeneous source domain. HDA is a challenging problem since the domains are not consistent in not only data distribution but also feature space. Most HDA methods attend to search for a subspace, where the features and the distributions across domains can be aligned. However, these methods barely consider the shared semantic label space of two domains and do not align the decision boundaries of the two domains, which may cause misclassification. To address the above issue, we propose a novel HDA method called Class centroid Matching and local Discriminative structure Preservation (CMDP), which can transfer discriminative semantic source knowledge to the target domain. Specifically, we project cross-domain samples to regress the label matrix to align the discriminative directions of two domains. Then, we introduce the inner product strategy to align the distance and angle of the class centroids across domains, such that the discriminative source knowledge can more sufficiently transfer to the target domain. Besides, to further improve the quality of the class centroids in each domain, we propose a novel cross-domain graph embedding strategy to exploit the structure information of data more thoroughly. A simple and efficient optimization algorithm is designed to solve the CMDP model. Extensive experiments on heterogeneous datasets validate the superiority of our proposal over several advanced methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability

The datasets generated during and/or analyzed during the current study can be accessed in https://gitee.com/caumath/cdmp.git.

Notes

  1. https://www.imageclef.org/2014/adaptation.

References

  1. Alipour N, Tahmoresnezhad J (2022) Heterogeneous domain adaptation with statistical distribution alignment and progressive pseudo label selection. Appl Intell 52(7):8038–8055

    Article  Google Scholar 

  2. Amini M, Usunier N, Goutte C (2009) Learning from multiple partially observed views—an application to multilingual text categorization. In: Proc NIPS, pp 28–36

  3. Chen W-Y, Hsu T-MH, Tsai Y-HH, Wang Y-CF, Chen M-S (2016) Transfer neural trees for heterogeneous domain adaptation. In: Proc ECCV, pp 399–414

  4. Chen W, Hu H (2020) Generative attention adversarial classification network for unsupervised domain adaptation. Pattern Recogn 107:107440

    Article  Google Scholar 

  5. Chen S, Harandi M, Jin X, Yang X (2021) Semi-supervised domain adaptation via asymmetric joint distribution matching. IEEE Trans Neural Netw Learn Syst 32(12):5708–5722

    Article  MathSciNet  Google Scholar 

  6. Duan L, Xu D, Tsang I (2012) Learning with augmented features for heterogeneous domain adaptation. In: Proc ICML, pp 711–718

  7. Fang W-C, Chiang Y-T (2018) A discriminative feature mapping approach to heterogeneous domain adaptation. Pattern Recogn Lett 106:13–19

    Article  Google Scholar 

  8. Gong B, Shi Y, Sha F, Grauman K (2012) Geodesic flow kernel for unsupervised domain adaptation. In: 2012 IEEE conference on computer vision and pattern recognition, pp 2066–2073

  9. Guo H, Pasunuru R, Bansal M (2020) Multi-source domain adaptation for text classification via distancenet-bandits. In: Proc AAAI, pp 7830–7838

  10. Guo Y, Wang X, Li C, Ying S (2023) Domain adaptive semantic segmentation by optimal transport. CoRR arXiv:2303.16435

  11. Hoffman J, Rodner E, Donahue J, Kulis B, Saenko K (2014) Asymmetric and category invariant feature transformations for domain adaptation. Int J Comput Vis 109(1):28–41

    Article  MathSciNet  Google Scholar 

  12. Hsieh Y-T, Tao S-Y, Tsai Y-HH, Yeh Y-R, Wang Y-CF (2016) Recognizing heterogeneous cross-domain data via generalized joint distribution adaptation. In: IEEE international conference on multimedia and expo, pp 1–6

  13. Huang S, Guo J, Yu Z, Wen Y (2023) Effective domain awareness and adaptation approach via mask substructure for multi-domain neural machine translation. Neural Comput Appl 35(19):14047–14060

    Article  Google Scholar 

  14. Jemni SK, Ammar S, Kessentini Y (2022) Domain and writer adaptation of offline Arabic handwriting recognition using deep neural networks. Neural Comput Appl 34(3):2055–2071

    Article  Google Scholar 

  15. Li W, Duan L, Xu D, Tsang IW (2014) Learning with augmented features for supervised and semi-supervised heterogeneous domain adaptation. IEEE Trans Pattern Anal Mach Intell 36(6):1134–1148

    Article  Google Scholar 

  16. Li J, Jing M, Lu K, Zhu L, Shen HT (2019) Locality preserving joint transfer for domain adaptation. IEEE Trans Image Process 28(12):6103–6115

    Article  MathSciNet  Google Scholar 

  17. Li J, Lu K, Huang Z, Zhu L, Shen HT (2019) Heterogeneous domain adaptation through progressive alignment. IEEE Trans Neural Netw Learn Syst 30(5):1381–1391

    Article  MathSciNet  Google Scholar 

  18. Li J, Li Z, Lü S (2021) Unsupervised double weighted domain adaptation. Neural Comput Appl 33(8):3545–3566

    Article  Google Scholar 

  19. Li Y, Cheng L, Peng Y, Wen Z, Ying S (2019) Manifold alignment and distribution adaptation for unsupervised domain adaptation. In: IEEE international conference on multimedia and expo, ICME 2019, Shanghai, China, July 8–12, 2019, pp 688–693

  20. Lin Q, Wang Z, Chen Y, Zhong P (2022) Supervised multi-view classification via the sparse learning joint the weighted elastic loss. Signal Process 191:108362

    Article  Google Scholar 

  21. Li H, Pan SJ, Wang S, Kot AC (2020) Heterogeneous domain adaptation via nonlinear matrix factorization. IEEE Trans Neural Netw Learn Syst 31(3):984–996

    Article  MathSciNet  Google Scholar 

  22. Luo Y, Wen Y, Liu T, Tao D (2019) Transferring knowledge fragments for learning distance metric from a heterogeneous domain. IEEE Trans Pattern Anal Mach Intell 41(4):1013–1026

    Article  Google Scholar 

  23. Ma N, Wang H, Zhang Z, Zhou S, Chen H, Bu J (2023) Source-free semi-supervised domain adaptation via progressive Mixup. Knowl Based Syst 262:110208

    Article  Google Scholar 

  24. Shao L, Zhu F, Li X (2015) Transfer learning for visual categorization: a survey. IEEE Trans Neural Netw Learn Syst 26(5):1019–1034

    Article  MathSciNet  Google Scholar 

  25. Suykens JAK, Vandewalle J (1999) Least squares support vector machine classifiers. Neural Process Lett 9(3):293–300

    Article  Google Scholar 

  26. Tang H, Jia K (2021) Vicinal and categorical domain adaptation. Pattern Recogn 115:107907

    Article  Google Scholar 

  27. Tian Q, Zhou J, Chu Y (2022) Joint bi-adversarial learning for unsupervised domain adaptation. Knowl Based Syst 248:108903

    Article  Google Scholar 

  28. Tian Q, Ma C, Cao M, Chen S, Yin H (2022) A convex discriminant semantic correlation analysis for cross-view recognition. IEEE Trans Cybern 52(2):849–861

    Article  Google Scholar 

  29. Tian Q, Sun H, Ma C, Cao M, Chu Y, Chen S (2022) Heterogeneous domain adaptation with structure and classification space alignment. IEEE Trans Cybern 52(10):10328–10338

    Article  Google Scholar 

  30. Tianyi J, Ivor Z, Tsang W, Jialin S, Tan PM (2014) Heterogeneous domain adaptation for multiple classes. In: Proc AISTATS, pp 1095–1103

  31. Tsai Y-HH, Yeh Y-R, Wang Y-CF (2016) Heterogeneous domain adaptation with label and structure consistency. In: Proc ICASSP, pp 2842–2846

  32. Tsai Y-HH, Yeh Y-R, Wang Y-CF (2016) Learning cross-domain landmarks for heterogeneous domain adaptation. In: Proc CVPR, pp 5081–5090

  33. van der Maaten L (2014) Accelerating t-SNE using tree-based algorithms. J Mach Learn Res 15(1):3221–3245

    MathSciNet  Google Scholar 

  34. Wang Q, Breckon TP (2022) Cross-domain structure preserving projection for heterogeneous domain adaptation. Pattern Recogn 123:108362

    Article  Google Scholar 

  35. Wang Z, Men M, Zhong P (2022) Re-weighting regression and sparsity regularization for multi-view classification. Appl Intell 52(7):7442–7458

    Article  Google Scholar 

  36. Wang Y, Zhang Y, Xu L, Qi S, Yao Y, Qian W, Greenwald SE, Qi L (2023) TSP-UDANet: two-stage progressive unsupervised domain adaptation network for automated cross-modality cardiac segmentation. Neural Comput Appl 35(30):22189–22207

    Article  Google Scholar 

  37. Wang C, Mahadevan S (2011) Heterogeneous domain adaptation using manifold alignment. In: Proc IJCAI, pp 1541–1546

  38. Wen J, Yuan J, Zheng Q, Liu R, Gong Z, Zheng N (2022) Hierarchical domain adaptation with local feature patterns. Pattern Recogn 124:108445

    Article  Google Scholar 

  39. Wu H, Zhu H, Yan Y, Wu J, Zhang Y, Ng MK (2021) Heterogeneous domain adaptation by information capturing and distribution matching. IEEE Trans Image Process 30:6364–6376

    Article  MathSciNet  Google Scholar 

  40. Yan S, Xu D, Zhang B, Zhang H-J, Yang Q, Lin S (2007) Graph embedding and extensions: a general framework for dimensionality reduction. IEEE Trans Pattern Anal Mach Intell 29(1):40–51

    Article  Google Scholar 

  41. Yang B, Yuen PC (2021) Learning adaptive geometry for unsupervised domain adaptation. Pattern Recogn 110:107638

    Article  Google Scholar 

  42. Yang L, Zhong P (2020) Discriminative and informative joint distribution adaptation for unsupervised domain adaptation. Knowl-Based Syst 207:106394

    Article  Google Scholar 

  43. Yao Y, Zhang Y, Li X, Ye Y (2020) Discriminative distribution alignment: a unified framework for heterogeneous domain adaptation. Pattern Recogn 101:107165

    Article  Google Scholar 

  44. Ye Y, Fu S, Chen J (2023) Learning cross-domain representations by vision transformer for unsupervised domain adaptation. Neural Comput Appl 35(15):10847–10860

    Article  Google Scholar 

  45. Ye H, Sheng X, Zhan D, He P (2018) Distance metric facilitated transportation between heterogeneous domains. In: Lang J (ed) Proceedings of the 27th international joint conference on artificial intelligence, pp 3012–3018

  46. Zhang Y, Hu W, Sun B, He J, Yu L (2023) Cocongan: cooperative contrastive learning for few-shot cross-domain heterogeneous face translation. Neural Comput Appl 35(20):15019–15032

    Article  Google Scholar 

  47. Zhao R, Mao K (2014) Supervised adaptive-transfer plsa for cross-domain text classification. In: ICDM Workshops, pp 259–266

  48. Zhou L, Ye M, Xiao S (2022) Domain adaptation based on source category prototypes. Neural Comput Appl 34(23):21191–21203

    Article  Google Scholar 

  49. Zhou K, Liu Z, Qiao Y, Xiang T, Loy CC (2023) Domain generalization: a survey. IEEE Trans Pattern Anal Mach Intell 45(4):4396–4415

    Google Scholar 

  50. Zhou J, Tian Q, Lu Z (2023) Progressive decoupled target-into-source multi-target domain adaptation. Inf Sci 634:140–156

    Article  Google Scholar 

  51. Zhuang F, Qi Z, Duan K, Xi D, Zhu Y, Zhu H, Xiong H, He Q (2021) A comprehensive survey on transfer learning. Proc IEEE 109(1):43–76

    Article  Google Scholar 

  52. Zhuang F, Qi Z, Duan K, Xi D, Zhu Y, Zhu H, Xiong H, He Q (2021) A comprehensive survey on transfer learning. Proc IEEE 109(1):43–76

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the Chinese Universities Scientific Fund (No. 2022TC109), the Double First-class Project of China Agricultural University (2022), and the Double First-class International Cooperation Project of China Agricultural University (No. 10020799).

Author information

Authors and Affiliations

  1. College of Science, China Agricultural University, Beijing, 100083, China

    Yuqing Chen & Ping Zhong

  2. College of Information and Electrical Engineering, China Agricultural University, Beijing, 100083, China

    Heng Zhou & Zhi Wang

  3. National Innovation Center for Digital Fishery, China Agricultural University, Beijing, 100083, China

    Heng Zhou, Zhi Wang & Ping Zhong

  4. Key Laboratory of Smart Farming Technologies for Aquatic Animal and Livestock, Ministry of Agriculture and Rural Affairs, China Agricultural University, Beijing, 100083, China

    Heng Zhou, Zhi Wang & Ping Zhong

  5. Beijing Engineering and Technology Research Center for Internet of Things in Agriculture, China Agricultural University, Beijing, 100083, China

    Heng Zhou, Zhi Wang & Ping Zhong

Authors
  1. Yuqing Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Heng Zhou
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Zhi Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ping Zhong
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Yuqing Chen contributed to Methodology, Formal analysis, Visualization, and Writing–original draft. Heng Zhou contributed to Writing—review & editing. Zhi Wang contributed to Software. Ping Zhong contributed to Supervision, Conceptualization, and Project administration.

Corresponding author

Correspondence to Ping Zhong.

Ethics declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Ethical approval

No human subjects were involved in any of the authors’ research for this article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chen, Y., Zhou, H., Wang, Z. et al. Heterogeneous domain adaptation by class centroid matching and local discriminative structure preservation. Neural Comput & Applic 36, 12865–12881 (2024). https://doi.org/10.1007/s00521-024-09786-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-024-09786-9

Keywords