Discriminative Joint Matching for Unsupervised Domain Adaptation
ZHANG Yong1,2, XIA Tianqi1, HUANG Dan1
1. School of Computer and Information Technology, Liaoning Normal University, Dalian 116081 2. School of Information Engineering, Huzhou University, Hu-zhou 313000
Abstract：The transfer effect of domain adaption is poor due to the large differences between domains. It can be improved by reducing the domain difference. However, the discriminability of later classification is ignored. A discriminative joint matching algorithm is proposed to handle this problem. Differentiation treatments are conducted according to different categories between domains. Feature matching and instance reweighting are combined to improve the migration effect. The joint probability distribution is employed to measure the difference of data distribution between domains. The transferability is enhanced by reducing the distance between the same domains. The discriminability is improved by expanding the distance between different domains. Feature matching and instance weighting are combined in the process of feature dimensionality reduction to jointly construct a feature transformation matrix. The experimental results show that the classification result of the proposed algorithm on 18 tasks is better.
[1] ZHUANG F Z, QI Z Y, DUAN K Y, et al. A Comprehensive Survey on Transfer Learning. Proceedings of the IEEE, 2021, 109(1): 43-76. [2] WANG J D, FENG W J, CHEN Y Q, et al. Visual Domain Adaptation with Manifold Embedded Distribution Alignment // Proc of the 26th ACM International Conference on Multimedia. New York, USA: ACM, 2018: 402-410. [3] 江 悠,张道强,张俊艺.基于多图核的迁移学习方法.模式识别与人工智能, 2020, 33(6): 488-495. (JIANG Y, ZHANG D Q, ZHANG J Y. Multi-graph Kernel Based Transfer Learning Method. Pattern Recognition and Artificial Intelligence, 2020, 33(6): 488-495.) [4] SONG P. Transfer Linear Subspace Learning for Cross-Corpus Speech Emotion Recognition. IEEE Transactions on Affective Computing, 2019, 10(2): 265-275. [5] 郑伟龙,石振锋,吕宝粮.用异质迁移学习构建跨被试脑电情感模型.计算机学报, 2020, 43(2): 177-189. (ZHENG W L, SHI Z F, LÜ B L. Building Cross-Subject EEG-Based Affective Models Using Heterogeneous Transfer Learning. Chinese Journal of Computers, 2020, 43(2): 177-189.) [6] PAN S J, YANG Q. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359. [7] PAN S J, TSANG I W, KWOK J T, et al. Domain Adaptation via Transfer Component Analysis. IEEE Transactions on Neural Networks, 2011, 22(2): 199-210. [8] LONG M S, WANG J M, DING G G, et al. Transfer Feature Lear-ning with Joint Distribution Adaptation // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2013: 2200-2207. [9] ZHAGN W, WU D R. Transferability versus Discriminability: Joint Probability Distribution Adaptation(JPDA)[C/OL]. [2021-01-07]. https://arxiv.org/pdf/1912.00320v1.pdf. [10] 李 萍,倪志伟,朱旭辉,等.联合类间及域间分布适配的迁移学习方法.模式识别与人工智能, 2020, 33(1): 1-10. (LI P, NI Z W, ZHU X H, et al. Transfer Learning with Joint Inter-Class and Inter-Domain Distributional Adaptation. Pattern Re-cognition and Artificial Intelligence, 2020, 33(1): 1-10.) [11] GRETTON A, BORGWARDT K M, RASCH M J, et al. A Kernel Two-Sample Test. Journal of Machine Learning Research, 2012, 13: 723-773. [12] WANG J D, CHEN Y Q, HAO S J, et al. Balanced Distribution Adaptation for Transfer Learning // Proc of the IEEE International Conference on Data Mining. Washington, USA: IEEE, 2017: 1129-1134. [13] HUANG J Y, SMOLA A J, GRETTON A, et al. Correcting Sample Selection Bias by Unlabeled Data // SCHÖLKOPF B, PLATT J, HOFMANN T, eds. Advances in Neural Information Processing Systems 19. Cambridge, USA: The MIT Press, 2007: 601-608. [14] SUGIYAMA M, SUZUKI T, NAKAJIMA S, et al. Direct Importance Estimation for Covariate Shift Adaptation. Annals of the Institute of Statistical Mathematics, 2008, 60(4): 699-746. [15] DAI W Y, YANG Q, XUE G R, et al. Boosting for Transfer Lear-ning // Proc of the 24th International Conference on Machine learning. New York, USA: ACM, 2007: 193-200. [16] LONG M S, WANG J M, DING G G, et al. Transfer Joint Ma-tching for Unsupervised Domain Adaptation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2014: 1410-1417. [17] ZHANG W, WU D R. Discriminative Joint Probability Maximum Mean Discrepancy(DJP-MMD) for Domain Adaptation // Proc of the International Joint Conference on Neural Networks. Washington, USA: IEEE, 2020. DOI: 10.1109/IJCNN48605.2020.9207365. [18] GU Q Q, LI Z H, HAN J W. Joint Feature Selection and Subspace Learning // Proc of the 22nd International Joint Conference on Artificial Intelligence. Berlin, Germany: Springer, 2011, II: 1294-1299. [19] CHU W S, DE LA TORRE F, COHN J F. Selective Transfer Machine for Personalized Facial Action Unit Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2013: 3515-3522. [20] SHLENS J. A Tutorial on Principal Component Analysis. International Journal of Remote Sensing[C/OL]. [2021-01-07]. https://www.cs.cmu.edu/~elaw/papers/pca.pdf. [21] NIE F P, HUANG H, CAI X, et al. Efficient and Robust Feature Selection via Joint l2,1-norms Minimization // Proc of the 23rd International Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2010: 1813-1821. [22] BUSCEMA B. MetaNet*: The Theory of Independent Judges. Substance Use and Misuse, 1998, 33(2): 439-461. [23] SAENKO K, KULIS B, FRITZ M, et al. Adapting Visual Category Models to New Domains // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2010: 213-226. [24] WANG J D, CHEN Y Q, YU H, et al. Easy Transfer Learning by Exploiting Intra-Domain Structures // Proc of the IEEE International Conference on Multimedia and Expo(ICME). Washington, USA: IEEE, 2019: 1210-1215. [25] DEMSAR J. Statistical Comparisons of Classifiers over Multiple Data Sets. Journal of Machine Learning Research, 2006, 7: 1-30.
2021, Vol. 34 
