Abstract
Black-box unsupervised domain adaptation (\(\textrm{B}^2 \textrm{UDA}\)) is a challenging task in unsupervised domain adaptation, where the source model is treated as a black box and only its output is accessible. Previous works have treated the source models as a pseudo-labeling tool and formulated \(\textrm{B}^2 \textrm{UDA}\) as a noisy labeled learning (LNL) problem. However, they have ignored the gap between the “shift noise” caused by the domain shift and the hypothesis noise in LNL. To alleviate the negative impact of shift noise on \(\textrm{B}^2 \textrm{UDA}\), we propose a novel framework called Classifier Decoupling Training (CDT), which introduces two additional classifiers to assist model training with a new label-confidence sampling. First, we introduce a self-training classifier to learn robust feature representation from the low-confidence samples, which is discarded during testing, and the final classifier is only trained with a few high-confidence samples. This step decouples the training of high-confidence and low-confidence samples to mitigate the impact of noise labels on the final classifier while avoiding overfitting to the few confident samples. Second, an adversarial classifier optimizes the feature distribution of low-confidence samples to be biased toward high-confidence samples through adversarial training, which greatly reduces intra-class variation. Third, we further propose a novel ETP-entropy Sampling (E\(^2\)S) to collect class-balanced high-confidence samples, which leverages the early-time training phenomenon into LNL. Extensive experiments on several benchmarks show that the proposed CDT achieves \(88.2\%\), \(71.6\%\), and \(81.3\%\) accuracies on Office-31, Office-Home, and VisDA-17, respectively, which outperforms state-of-the-art methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Asano, Y.M., Rupprecht, C., Vedaldi, A.: Self-labelling via simultaneous clustering and representation learning. arXiv preprint: arXiv:1911.05371 (2019)
Chang, H., Shejwalkar, V., Shokri, R., Houmansadr, A.: Cronus: robust and heterogeneous collaborative learning with black-box knowledge transfer. arXiv preprint: arXiv:1912.11279 (2019)
Chen, X., Wang, S., Long, M., Wang, J.: Transferability vs. discriminability: batch spectral penalization for adversarial domain adaptation. In: ICML (2019)
Chen, Y.C., Lin, Y.Y., Yang, M.H., Huang, J.B.: CrDoCo: pixel-level domain transfer with cross-domain consistency. In: CVPR (2019)
Cheng, D., et al.: Instance-dependent label-noise learning with manifold-regularized transition matrix estimation. In: CVPR (2022)
Cui, K., Huang, J., Luo, Z., Zhang, G., Zhan, F., Lu, S.: GenCo: generative co-training on data-limited image generation. arXiv preprint: arXiv:2110.01254 (2021)
Dou, Q., Ouyang, C., Chen, C., Chen, H., Heng, P.A.: Unsupervised cross-modality domain adaptation of convnets for biomedical image segmentations with adversarial loss. In: IJCAI (2018)
Ganin, Y., et al.: Domain-adversarial training of neural networks. JMLR (2016)
Guo, C., Pleiss, G., Sun, Y., Weinberger, K.Q.: On calibration of modern neural networks. In: ICML (2017)
Han, B., et al.: Co-teaching: robust training of deep neural networks with extremely noisy labels. In: NeurIPS (2018)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)
Hendrycks, D., Gimpel, K.: A baseline for detecting misclassified and out-of-distribution examples in neural networks. arXiv preprint: arXiv:1610.02136 (2016)
Huang, J., Guan, D., Xiao, A., Lu, S.: Cross-view regularization for domain adaptive panoptic segmentation. In: CVPR (2021)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: NeurIPS (2012)
Lee, S., Kim, D., Kim, N., Jeong, S.G.: Drop to adapt: learning discriminative features for unsupervised domain adaptation. In: CVPR (2019)
Liang, J., Hu, D., Wang, Y., He, R., Feng, J.: Source data-absent unsupervised domain adaptation through hypothesis transfer and labeling transfer. PAMI (2020)
Liang, J., Hu, D., Feng, J.: Do we really need to access the source data? Source hypothesis transfer for unsupervised domain adaptation. In: ICML (2020)
Liang, J., Hu, D., Feng, J.: Domain adaptation with auxiliary target domain-oriented classifier. In: CVPR (2021)
Liang, J., Hu, D., Feng, J., He, R.: DINE: domain adaptation from single and multiple black-box predictors. In: CVPR (2022)
Liu, S., Niles-Weed, J., Razavian, N., Fernandez-Granda, C.: Early-learning regularization prevents memorization of noisy labels. In: NeurIPS (2020)
Long, M., Cao, Z., Wang, J., Jordan, M.I.: Conditional adversarial domain adaptation. In: NeurIPS (2018)
Luo, Y.W., Ren, C.X.: Conditional bures metric for domain adaptation. In: CVPR (2021)
Pan, S.J., Yang, Q.: A survey on transfer learning. TKDE (2010)
Peng, Q., Ding, Z., Lyu, L., Sun, L., Chen, C.: Toward better target representation for source-free and black-box domain adaptation. arXiv preprint: arXiv:2208.10531 (2022)
Peng, X., Usman, B., Kaushik, N., Hoffman, J., Wang, D., Saenko, K.: VisDA: the visual domain adaptation challenge. arXiv preprint: arXiv:1710.06924 (2017)
Saenko, K., Kulis, B., Fritz, M., Darrell, T.: Adapting visual category models to new domains. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010. LNCS, vol. 6314, pp. 213–226. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15561-1_16
Venkateswara, H., Eusebio, J., Chakraborty, S., Panchanathan, S.: Deep hashing network for unsupervised domain adaptation. In: CVPR (2017)
Xu, R., Li, G., Yang, J., Lin, L.: Larger norm more transferable: an adaptive feature norm approach for unsupervised domain adaptation. In: ICCV (2019)
Yin, H., Mallya, A., Vahdat, A., Alvarez, J.M., Kautz, J., Molchanov, P.: See through gradients: image batch recovery via gradinversion. In: CVPR (2021)
Yin, H., et al.: Dreaming to distill: Data-free knowledge transfer via deepinversion. In: CVPR (2020)
Zhang, H., Zhang, Y., Jia, K., Zhang, L.: Unsupervised domain adaptation of black-box source models. arXiv preprint: arXiv:2101.02839 (2021)
Zhang, Y., Liu, T., Long, M., Jordan, M.: Bridging theory and algorithm for domain adaptation. In: ICML (2019)
Zhu, J.Y., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. In: ICCV (2017)
Acknowledgments
This work is supported by the National Natural Science Foundation of China (NO. 62102151), Shanghai Sailing Program (21YF1411200), CCF-Tencent Open Research Fund, the Open Research Fund of Key Laboratory of Advanced Theory and Application in Statistics and Data Science, Ministry of Education (KLATASDS2305), the Fundamental Research Funds for the Central Universities.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Chen, X., Shen, Y., Luo, X., Zhang, Y., Li, K., Lin, S. (2024). Classifier Decoupled Training for Black-Box Unsupervised Domain Adaptation. In: Liu, Q., et al. Pattern Recognition and Computer Vision. PRCV 2023. Lecture Notes in Computer Science, vol 14427. Springer, Singapore. https://doi.org/10.1007/978-981-99-8435-0_2
Download citation
DOI: https://doi.org/10.1007/978-981-99-8435-0_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-8434-3
Online ISBN: 978-981-99-8435-0
eBook Packages: Computer ScienceComputer Science (R0)