Unsupervised domain adaptation (UDA) aims to transfer the knowledge from labeled source domain to unlabeled target domain. The main challenge of UDA stems from the domain shift between the source and target domains. Currently, in the discrete classification problems, most existing UDA methods usually adopt the distribution alignment strategy while enforcing unstable instances to pass through the low-density areas. However, the scenario of ordinal regression (OR) is rarely researched in UDA, and the traditional UDA methods cannot preferably handle OR since they do not preserve the order relationships in data labels, like in human age estimation. To address this issue, we proposed a structure-oriented adaptation strategy, namely, structure preserved ordinal unsupervised domain adaptation (SPODA). More specifically, on one hand, the global structure information was modeled and embedded into an auto-encoder framework via a low-rank transferred structure matrix. On the other hand, the local structure information was preserved through a weighted pair-wise strategy in the latent space. Guided by both the local and global structure information, a well-performance latent space was generated, whose geometric structure was adopted to further obtain a more discriminant ordinal regressor. To further enhance its generalization, a counterpart of SPODA with deep architecture was developed. Finally, extensive experiments indicated that in addressing the OR problem, SPODA was more effective and advanced than existing related domain adaptation methods.
Citation: Qing Tian, Canyu Sun. Structure preserved ordinal unsupervised domain adaptation[J]. Electronic Research Archive, 2024, 32(11): 6338-6363. doi: 10.3934/era.2024295
Unsupervised domain adaptation (UDA) aims to transfer the knowledge from labeled source domain to unlabeled target domain. The main challenge of UDA stems from the domain shift between the source and target domains. Currently, in the discrete classification problems, most existing UDA methods usually adopt the distribution alignment strategy while enforcing unstable instances to pass through the low-density areas. However, the scenario of ordinal regression (OR) is rarely researched in UDA, and the traditional UDA methods cannot preferably handle OR since they do not preserve the order relationships in data labels, like in human age estimation. To address this issue, we proposed a structure-oriented adaptation strategy, namely, structure preserved ordinal unsupervised domain adaptation (SPODA). More specifically, on one hand, the global structure information was modeled and embedded into an auto-encoder framework via a low-rank transferred structure matrix. On the other hand, the local structure information was preserved through a weighted pair-wise strategy in the latent space. Guided by both the local and global structure information, a well-performance latent space was generated, whose geometric structure was adopted to further obtain a more discriminant ordinal regressor. To further enhance its generalization, a counterpart of SPODA with deep architecture was developed. Finally, extensive experiments indicated that in addressing the OR problem, SPODA was more effective and advanced than existing related domain adaptation methods.
| [1] | Y. Liu, Z. Zhou, B. Sun, Cot: Unsupervised domain adaptation with clustering and optimal transport, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2023), 19998–20007. https://doi.org/10.1109/CVPR52729.2023.01915 |
| [2] |
M. Wang, S. Wang, X. Yang, J. Yuan, W. Zhang, Equity in unsupervised domain adaptation by nuclear norm maximization, IEEE Trans. Circuits Syst. Video Technol., 34 (2024), 5533–5545. https://doi.org/10.1109/TCSVT.2023.3346444 doi: 10.1109/TCSVT.2023.3346444
|
| [3] |
M. Wang, Y. Liu, J. Yuan, S. Wang, Z. Wang, W. Wang, Inter-class and inter-domain semantic augmentation for domain generalization, IEEE Trans. Image Process., 33 (2024), 1338–1347. https://doi.org/10.1109/TIP.2024.3354420 doi: 10.1109/TIP.2024.3354420
|
| [4] | T. Mikolov, M. Karafiát, L. Burget, J.Cernockỳ, S. Khudanpur, Recurrent neural network based language model, in 11th Annual Conference of the International Speech Communication Association, (2010), 1045–1048. https://doi.org/10.21437/INTERSPEECH.2010-343 |
| [5] | I. Sutskever, J. Martens, G. Hinton, Generating text with recurrent neural networks, in Proceedings of the 28th international conference on machine learning, (2011), 1017–1024. |
| [6] | T. Mikolov, S. Kombrink, L. Burget, J. Černockỳ, S. Khudanpur, Extensions of recurrent neural network language model, in 2011 IEEE International Conference on Acoustics, Speech and Signal Processing, (2019), 5528–5531. https://doi.org/10.1109/ICASSP.2011.5947611 |
| [7] | Y. Wang, S. Tang, Y. Lei, W. Song, S. Wang, M. Zhang, Disenhan: Disentangled heterogeneous graph attention network for recommendation, in Proceedings of the 29th ACM International Conference on Information Knowledge Management, (2020), 1605–1614. https://doi.org/10.1145/3340531.3411996 |
| [8] | Y. Wang, Y. Li, S. Li, W. Song, J. Fan, S. Gao, et al., Deep graph mutual learning for cross-domain recommendation, in International Conference on Database Systems for Advanced Applications, (2022), 298–305. https://doi.org/10.1007/978-3-031-00126-0_22 |
| [9] |
Y. Wang, X. Luo, C. Chen, X. Hua, M. Zhang, W. Ju, Disensemi: Semi-supervised graph classification via disentangled representation learning, IEEE Trans. Neural Networks Learn. Syst., (2024). https://doi.org/10.1109/TNNLS.2024.3431871 doi: 10.1109/TNNLS.2024.3431871
|
| [10] | I. Shlizerman, S. Suwajanakorn, S. Seitz. Illumination-aware age progression, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2014), 3334–3341. https://doi.org/10.1109/CVPR.2014.426 |
| [11] |
H. Liu, J. Lu, J. Feng, J. Zhou, Ordinal deep learning for facial age estimation, IEEE Trans. Circuits Syst. Video Technol., 29 (2019). https://doi.org/10.1109/TCSVT.2017.2782709 doi: 10.1109/TCSVT.2017.2782709
|
| [12] | J. Wang, W. Feng, Y. Chen, H. Yu, M. Huang, P. Yu, Visual domain adaptation with manifold embedded distribution alignment, in Proceedings of the 26th ACM international Conference on Multimedia, (2018), 402–410. https://doi.org/10.1145/3240508.3240512 |
| [13] |
C. Ren, Y. Zhai, Y. Luo, H. Yan, Towards unsupervised domain adaptation via domain-transformer, Int. J. Comput. Vis., 132 (2024), 6163–6183. https://link.springer.com/article/10.1007/s11263-024-02174-9 doi: 10.1007/s11263-024-02174-9
|
| [14] | S. David, J. Blitzer, K. Crammer, F. Pereira, Analysis of representations for domain adaptation, Adv. Neural Inf. Process. Syst., 19 (2007), 137–144. |
| [15] |
S. David, J. Blitzer, K. Crammer, A. Kulesza, F. Pereira, J. W. Vaughan, A theory of learning from different domains, Mach. learn., 79 (2010), 151–175. https://doi.org/10.1007/s10994-009-5152-4 doi: 10.1007/s10994-009-5152-4
|
| [16] | V. Vapnik, Estimation of Dependences Based on Empirical Data, Springer Science & Business Media, 2006. https://doi.org/10.1007/0-387-34239-7 |
| [17] |
S. Pan, I. Tsang, J. Kwok, Q. Yang, Domain adaptation via transfer component analysis, IEEE Trans. Neural Networks Learn. Syst., 22 (2010), 199–210. https://doi.org/10.1109/TNN.2010.2091281 doi: 10.1109/TNN.2010.2091281
|
| [18] | R. Combes, H. Zhao, Y. Wang, G. Gordon, Domain adaptation with conditional distribution matching and generalized label shift, Adv. Neural Inf. Process. Syst., 33 (2020). |
| [19] | J. Wang, Y. Chen, S. Hao, W. Feng, Z. Shen, Balanced distribution adaptation for transfer learning, in 2017 IEEE International Conference on Data Mining, (2017), 1129–1134. https://doi.org/10.1109/ICDM.2017.150 |
| [20] | M. Long, J. Wang, G. Ding, J. Sun, P. Yu, Transfer feature learning with joint distribution adaptation, in Proceedings of the IEEE International Conference on Computer Vision, (2013), 2200–2207. https://doi.org/10.1109/ICCV.2013.274 |
| [21] |
S. Li, S. Song, G. Huang, Prediction reweighting for domain adaptation, IEEE Trans. Neural Networks Learn. Syst., 28 (2016), 1682–1695. https://doi.org/10.1109/TNNLS.2016.2538282 doi: 10.1109/TNNLS.2016.2538282
|
| [22] | S. Chen, F. Zhou, Q. Liao, Visual domain adaptation using weighted subspace alignment, in 2016 Visual Communications and Image Processing (VCIP), (2016), 1–4. https://doi.org/10.1109/VCIP.2016.7805516 |
| [23] |
L. Zhang, S. Wang, G. Huang, W. Zuo, J. Yang, D. Zhang, Manifold criterion guided transfer learning via intermediate domain generation, IEEE Trans. Neural Networks Learn. Syst., 30 (2019), 3759–3773. https://doi.org/10.1109/TNNLS.2019.2899037 doi: 10.1109/TNNLS.2019.2899037
|
| [24] | Y. Grandvalet, Y. Bengio, Semi-supervised learning by entropy minimization, Neural Inf. Process. Syst., (2004), 529–536. |
| [25] | S. Ahmed, D. Raychaudhuri, S. Paul, S. Oymak, A. RoyChowdhury, Unsupervised multi-source domain adaptation without access to source data, in Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition, (2021), 10103–10112. https://doi.org/10.1109/CVPR46437.2021.00997 |
| [26] |
Q. Tian, Y. Zhu, H. Sun, S. Chen, H. Yin, Unsupervised domain adaptation through dynamically aligning both the feature and label spaces, IEEE Trans. Circuits Syst. Video Technol., 32 (2022), 8562–8573. https://doi.org/10.1109/TCSVT.2022.3192135 doi: 10.1109/TCSVT.2022.3192135
|
| [27] | S. Roy, M. Trapp, A. Pilzer, J. Kannala, N. Sebe, E. Ricci, et al., Uncertainty-guided source-free domain adaptation, in European Conference on Computer Vision, (2022), 537–555. https://doi.org/10.1007/978-3-031-19806-9_31 |
| [28] | H. Mao, L. Du, Y. Zheng, Q. Fu, Z. Li, X. Chen, et al., Source free graph unsupervised domain adaptation, in Proceedings of the 17th ACM International Conference on Web Search and Data Mining, (2024), 520–528. https://doi.org/10.1145/3616855.3635802 |
| [29] | X. Wu, L. Cheng, S. Zhang, Open set domain adaptation with entropy minimization, in Pattern Recognition and Computer Vision: Third Chinese Conference, (2020), 29–41. |
| [30] | J. Kundu, N. Venkat, A. Revanur, R. Babu, Towards inheritable models for open-set domain adaptation, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2020), 12376–12385. https://doi.org/10.1109/CVPR42600.2020.01239 |
| [31] | K. Saito, K. Saenko, Ovanet: One-vs-all network for universal domain adaptation, in Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition, (2021), 9000–9009. https://doi.org/10.1109/ICCV48922.2021.00887 |
| [32] | Y. Lu, M. Shen, A. Ma, X. Xie, J. Lai, Mlnet: Mutual learning network with neighborhood invariance for universal domain adaptation, in Proceedings of the AAAI Conference on Artificial Intelligence, 38 (2024), 3900–3908. https://doi.org/10.1609/aaai.v38i4.28182 |
| [33] | F. Qiao, L. Zhao, X. Peng, Learning to learn single domain generalization, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2020), 12556–12565. https://doi.org/10.1109/CVPR42600.2020.01257 |
| [34] | K. Ricanek, T Tesafaye, Morph: A longitudinal image database of normal adult age-progression, in 7th International Conference on Automatic Face and Gesture Recognition (FGR06), (2006), 341–345. https://doi.org/10.1109/FGR.2006.78 |
| [35] | X. Liu, S. Li, Y. Ge, P. Ye, J. You, J. Lu, Recursively conditional gaussian for ordinal unsupervised domain adaptation, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2021), 764–773. https://doi.org/10.1109/ICCV48922.2021.00080 |
| [36] |
X. Liu, S. Li, Y. Ge, P. Ye, J. You, J. Lu, Ordinal unsupervised domain adaptation with recursively conditional gaussian imposed variational disentanglement, IEEE Trans. Pattern Anal. Mach. Intell., (2022), 1–14. https://doi.org/10.1109/TPAMI.2022.3183115 doi: 10.1109/TPAMI.2022.3183115
|
| [37] |
Q. Tian, W. Zhang, M. Cao, L. Wang, S. Chen, H. Yin, Moment-guided discriminative manifold correlation learning on ordinal data, ACM Trans. Intell. Syst. Technol. (TIST), 11 (2020), 1–18. https://doi.org/10.1145/3402445 doi: 10.1145/3402445
|
| [38] | Z. Kang, Y. Lu, Y. Su, C. Li, Z. Xu, Similarity learning via kernel preserving embedding, in Proceedings of the AAAI Conference on Artificial Intelligence, 33 (2019), 4057–4064. https://doi.org/10.1609/aaai.v33i01.33014057 |
| [39] |
C. Geng, S. Chen, Metric learning-guided least squares classifier learning, IEEE Trans. Neural Networks Learn. Syst., 29 (2018), 6409–6414. https://doi.org/10.1109/TNNLS.2018.2830802 doi: 10.1109/TNNLS.2018.2830802
|
| [40] | Y. Ganin, V. Lempitsky, Unsupervised domain adaptation by backpropagation, in International Conference on Machine Learning, (2015), 1180–1189. http://proceedings.mlr.press/v37/ganin15.html |
| [41] |
Y. Yao, Y. Zhang, X. Li, Y. Ye, Discriminative distribution alignment: A unified framework for heterogeneous domain adaptation, Pattern Recognit., 101 (2020), 107165. https://doi.org/10.1016/j.patcog.2019.107165 doi: 10.1016/j.patcog.2019.107165
|
| [42] | W. Zellinger, T. Grubinger, E. Lughofer, T. Natschläger, S. Platz, Central moment discrepancy (cmd) for domain-invariant representation learning, preprint, arXiv: 1702.08811. |
| [43] | B. Sun, J. Feng, K. Saenko, Return of frustratingly easy domain adaptation, in Proceedings of the AAAI Conference on Artificial Intelligence, 30 (2016), 2058–2065. https://doi.org/10.1609/aaai.v30i1.10306 |
| [44] | M Long, Y Cao, J Wang, M Jordan, Learning transferable features with deep adaptation networks, in International Conference on Machine Learning, 37 (2015), 97–105. |
| [45] | C. Chen, Z. Chen, B. Jiang, X. Jin, Joint domain alignment and discriminative feature learning for unsupervised deep domain adaptation, in Proceedings of the AAAI Conference on Artificial Intelligence, 33 (2019), 3296–3303. https://doi.org/10.1609/aaai.v33i01.33013296 |
| [46] | I. Goodfellow, J. Abadie, M. Mirza, B. Xu, D. Farley, S. Ozair, et al., Generative adversarial nets, Adv. Neural Inf. Process. Syst., (2014), 2672–2680. |
| [47] | E. Tzeng, J. Hoffman, K. Saenko, T. Darrell, Adversarial discriminative domain adaptation, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2017), 7167–7176. https://doi.org/10.1109/CVPR.2017.316 |
| [48] | H. Tang, K. Jia, Discriminative adversarial domain adaptation, in Proceedings of the AAAI Conference on Artificial Intelligence, 34 (2020), 5940–5947. https://doi.org/10.1609/aaai.v34i04.6054 |
| [49] | K. Saito, K. Watanabe, Y. Ushiku, T. Harada, Maximum classifier discrepancy for unsupervised domain adaptation, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2018), 3723–3732. https://doi.org/10.1109/CVPR.2018.00392 |
| [50] | L. Zhou, M. Ye, X. Zhu, S. Li, Y. Liu, Class discriminative adversarial learning for unsupervised domain adaptation, in Proceedings of the 30th ACM International Conference on Multimedia, (2022), https://doi.org/10.1145/3503161.3548143 |
| [51] | K. Saito, D. Kim, S. Sclaroff, K. Saenko, Universal domain adaptation through self supervision, Adv. Neural Inf. Process. Syst., 33 (2020), 16282–16292. |
| [52] | J. Wang, Y. Chen, H. Yu, M. Huang, Q. Yang, Easy transfer learning by exploiting intra-domain structures, in 2019 IEEE International Conference on Multimedia and Expo (ICME), (2019), 1210–1215. https://doi.org/10.1109/ICME.2019.00211 |
| [53] | Q. Wang, T. Breckon, Unsupervised domain adaptation via structured prediction based selective pseudo-labeling, in Proceedings of the AAAI Conference on Artificial Intelligence, 34 (2020), 6243–6250. https://doi.org/10.1609/aaai.v34i04.6091 |
| [54] |
L. Zhang, J. Fu, S. Wang, D. Zhang, Z. Dong, C. Chen, Guide subspace learning for unsupervised domain adaptation, IEEE Trans. Neural Networks Learn. Syst., 31 (2020), 3374–3388. https://doi.org/10.1109/TNNLS.2019.2944455 doi: 10.1109/TNNLS.2019.2944455
|
| [55] | K. He, H. Fan, Y. Wu, S. Xie, R. Girshick, Momentum contrast for unsupervised visual representation learning, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2020), 9729–9738. https://doi.org/10.1109/CVPR42600.2020.00975 |
| [56] | T. Chen, S. Kornblith, M. Norouzi, G. Hinton, A simple framework for contrastive learning of visual representations, in International Conference on Machine Learning, (2020), 1597–1607. http://proceedings.mlr.press/v119/chen20j.html |
| [57] |
R. Wang, Z. Wu, Z. Weng, J. Chen, G. Qi, Y. Jiang, Cross-domain contrastive learning for unsupervised domain adaptation, IEEE Trans. Multim., 25 (2023), 1665–1673. https://doi.org/10.1109/TMM.2022.3146744 doi: 10.1109/TMM.2022.3146744
|
| [58] | W. Ma, J. Zhang, S. Li, C. Liu, Y. Wang, W. Li, Making the best of both worlds: A domain-oriented transformer for unsupervised domain adaptation, in Proceedings of the 30th ACM International Conference on Multimedia, (2022), 5620–5629. https://doi.org/10.1145/3503161.3548229 |
| [59] | Y. Zhang, Z. Wang, J. Li, J. Zhuang, Z. Lin, Towards effective instance discrimination contrastive loss for unsupervised domain adaptation, in Proceedings of the IEEE/CVF International Conference on Computer Vision, (2023), 11388–11399. https://doi.org/10.1109/ICCV51070.2023.01046 |
| [60] | H. Liu, M. Shao, Y. Fu, Structure-preserved multi-source domain adaptation, in 2016 IEEE 16th International Conference on Data Mining (ICDM), (2016), 1059–1064. https://doi.org/10.1109/ICDM.2016.0136 |
| [61] |
H. Liu, M. Shao, Z. Ding, Y. Fu, Structure-preserved unsupervised domain adaptation, IEEE Trans. Knowl. Data Eng., 31 (2018), 799–812. https://doi.org/10.1109/TKDE.2018.2843342 doi: 10.1109/TKDE.2018.2843342
|
| [62] |
M. Meng, Q. Chen, J. Wu, Structure preservation adversarial network for visual domain adaptation, Inf. Sci., 579 (2021), 266–280. https://doi.org/10.1016/j.ins.2021.07.085 doi: 10.1016/j.ins.2021.07.085
|
| [63] |
Q. Tian, H. Sun, C. Ma, M. Cao, Y. Chu, S. Chen, Heterogeneous domain adaptation with structure and classiffcation space alignment, IEEE Trans. Cybern., 52 (2022), 10328–10338. https://doi.org/10.1109/TCYB.2021.3070545 doi: 10.1109/TCYB.2021.3070545
|
| [64] | J. Jiang, Y. Ji, X. Wang, Y. Liu, J. Wang, M. Long, Regressive domain adaptation for unsupervised keypoint detection, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2021), 6780–6789. https://doi.org/10.1109/CVPR46437.2021.00671 |
| [65] |
C. Seah, I. Tsang, Y. Ong, Transfer ordinal label learning, IEEE Trans. Neural Networks Learn. Syst., 24 (2013), 1863–1876. https://doi.org/10.1109/TNNLS.2013.2268541 doi: 10.1109/TNNLS.2013.2268541
|
| [66] | X. Chen, S. Wang, J. Wang, M. Long, Representation subspace distance for domain adaptation regression, in International Conference on Machine Learning, (2021), 1749–1759. http://proceedings.mlr.press/v139/chen21u.html |
| [67] | W. Wu, J. He, S. Wang, K. Guan, E. Ainsworth, Distribution-informed neural networks for domain adaptation regression, Adv. Neural Inf. Process. Syst., 35 (2022), 10040–10054. |
| [68] | I. Nejjar, Q. Wang, O. Fink, Dare-gram: Unsupervised domain adaptation regression by aligning inverse gram matrices, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2023), 11744–11754. https://doi.org/10.1109/CVPR52729.2023.01130 |
| [69] | H. Wang, H. He, D. Katabi, Continuously indexed domain adaptation, preprint, arXiv: 2007.01807 |
| [70] | X. Zhong, L. Xu, Y. Li, Z. Liu, E. Chen, A nonconvex relaxation approach for rank minimization problems, in Proceedings of the AAAI Conference on Artificial Intelligence, 29 (2015), 266–280. https://doi.org/10.1609/aaai.v29i1.9482 |
| [71] |
Q. Tian, M. Cao, S. Chen, H. Yin, Structure-exploiting discriminative ordinal multioutput regression, IEEE Trans. Neural Networks Learn. Syst., 32 (2020), 266–280. https://doi.org/10.1109/TNNLS.2020.2978508 doi: 10.1109/TNNLS.2020.2978508
|
| [72] | P. Zadeh, R. Hosseini, S. Sra. Geometric mean metric learning, in International Conference on Machine Learning, (2016), 2464–2471. http://proceedings.mlr.press/v48/zadeh16.html |
| [73] | X. He, P. Niyogi, Locality preserving projections, Adv. Neural Inf. Process. Syst., (2003), 153–160. |
| [74] | Z. Niu, M. Zhou, L. Wang, X. Gao, G. Hua, Ordinal regression with multiple output cnn for age estimation, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2016), 4920–4928. https://doi.org/10.1109/CVPR.2016.532 |
| [75] | S. Moschoglou, A. Papaioannou, C. Sagonas, J. Deng, I. Kotsia, S. Zafeiriou Agedb: the first manually collected, in-the-wild age database, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, (2017), 51–59. https://doi.org/10.1109/CVPRW.2017.250 |
| [76] | B. Sun, K. Saenko, Deep coral: Correlation alignment for deep domain adaptation, in European Conference on Computer Vision, (2016), 443–450. https://doi.org/10.1007/978-3-319-49409-8_35 |
| [77] | C. Yu, J. Wang, Y. Chen, M Huang, Transfer learning with dynamic adversarial adaptation network, in 2019 IEEE International Conference on Data Mining (ICDM), (2019), 778–786. https://doi.org/10.1109/ICDM.2019.00088 |
Figures(7) / Tables(5)
Qing Tian, Canyu Sun. Structure preserved ordinal unsupervised domain adaptation[J]. Electronic Research Archive, 2024, 32(11): 6338-6363. doi: 10.3934/era.2024295
DownLoad: