1. Research Center of Precision Sensing and Control, Institute of Automation, Chinese Academy of Sciences, Beijing 100190 2. School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049
Abstract：Aiming at minimization of the joint distribution difference between source domain and target domain in domain adaptation, a two-stage domain adaptation learning method is proposed. In the first stage, the discriminative information of sample labels and the data structure are considered, and a shared projection transformation is learned to minimize the difference of marginal distribution in the shared-projected space. In the second stage, an adaptive classifier with structural risk is learned by the labeled source data and unlabeled target data. The classifier minimizes the difference of conditional distribution of source domain and target domain as well as maintains the manifold consistency underlying the marginal distributions. Experiments on three benchmark datasets show that the method achieves better results on average classification accuracy and the Kappa coefficient.
[1] 庄福振,罗 平,何 清,等.迁移学习研究进展.软件学报, 2015, 26(1): 26-39. (ZHUANG F Z, LUO P, HE Q, et al. Survey on Transfer Learning Research. Journal of Software, 2015, 26(1): 26-39.) [2] GOPALAN R, LI R N, CHELLAPPA R. Domain Adaptation for Object Recognition: An Unsupervised Approach // Proc of the IEEE Conference on Computer Vision. Washington, USA: IEEE, 2011: 999-1006. [3] GONG B Q, SHI Y, SHA F, et al. Geodesic Flow Kernel for Unsupervised Domain Adaptation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2012: 2066-2073. [4] FERNANDO B, HABRAD A, SEBBAN M, et al. Unsupervised Visual Domain Adaptation Using Subspace Alignment // Proc of the IEEE Conference on Computer Vision. Washington, USA: IEEE, 2013: 2960-2967. [5] PAN S J, TSANG I W, KWOK J T, et al. Domain Adaptation via Transfer Component Analysis. IEEE Transaction on Neural Networks, 2011, 22(2): 199-210. [6] LONG M S, WANG J M, DING G G, et al. Transfer Feature Lear-ning with Joint Distribution Adaptation // Proc of the IEEE Confe-rence on Computer Vision. Washington, USA: IEEE, 2013: 2200-2207. [7] VENKATESWARA H, CHAKRABORTY S, MCDANIEL T, et al. Model Selection with Nonlinear Embedding for Unsupervised Domain Adaptation[C/OL]. [2019-05-15]. https://arxiv.org/pdf/1706.07527.pdf. [8] ZHANG J, LI W Q, OGUNBONA P. Joint Geometrical and Statistical Alignment for Visual Domain Adaptation // Proc of the IEEE Conference on Computer Vision. Washington, USA: IEEE, 2017: 5150-5158. [9] LI S, SONG S J, HUANG G, et al. Domain Invariant and Class Discriminative Feature Learning for Visual Domain Adaptation. IEEE Transaction on Image Processing, 2018, 27(9): 4260-4273. [10] YANG J, YAN R, HAUPTMANN A G. Cross-Domain Video Concept Detection Using Adaptive SVMs // Proc of the 15th ACM International Conference on Multimedia. New York, USA: ACM, 2007: 188-197. [11] YAO T, PAN Y W, NGO C W, et al. Semi-supervised Domain Adaptation with Subspace Learning for Visual Recognition // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2015: 2142-2150. [12] ZHANG L, ZHANG D. Robust Visual Knowledge Transfer via Extreme Learning Machine-Based Domain Adaptation. IEEE Transaction on Image Processing, 2016, 25(10): 4959-4973. [13] DUAN L X, TSANG I W, XU D, et al. Domain Transfer SVM for Video Concept Detection // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2009: 1375-1381. [14] LONG M S, WANG J M, DING G G, et al. Adaptation Regulari-zation: A General Framework for Transfer Learning. IEEE Transa-ction on Knowledge and Data Engineering, 2014, 26(5): 1076-1089. [15] 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. [16] PAN S J, YANG Q. A Survey on Transfer Learning. IEEE Trans-actions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359. [17] PAN S J, KWOK J T, YANG Q. Transfer Learning via Dimensionality Reduction // Proc of the 23rd National Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2008, II: 677-682. [18] DONAHUE J, JIA Y Q, VINYALS O, et al. DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition // Proc of the 31st International Conference on Machine Learning. New York, USA: ACM, 2014: 647-655. [19] 龙明盛.迁移学习问题与方法研究.博士学位论文.北京:清华大学, 2014. (LONG M S. Transfer Learning: Problems and Methods. Ph.D. Dissertation. Beijing, China: Tsinghua University, 2014.) [20] FANG C, XU Y, ROCKMORE D N. Unbiased Metric Learning: On the Utilization of Multiple Datasets and Web Images for Softe-ning Bias // Proc of the IEEE International Conference on Compu-ter Vision. Washington, USA: IEEE, 2013: 1657-1664. [21] FUKUNAGA K, NARENDRA P M. A Branch and Bound Algorithm for Computing k-Nearest Neighbors. IEEE Transactions on Computers, 1975, 24(7): 750-753. [22] VAPNIK V N. An Overview of Statistical Learning Theory. IEEE Transaction on Neural Networks, 1999, 10(5): 988-999. [23] SUN B C, FENG J S, SAENKO K. Return of Frustratingly Easy Domain Adaptation // Proc of the 30th Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2016: 2058-2065. [24] LIANG J, HE R, SUN Z N, et al. Aggregating Randomized Clustering-Promoting Invariant Projections for Domain Adaptation. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2019, 41(5): 1027-1042. [25] VOGADO L H S, VERAS R M S, ARAUJO F H D, et al. Leuke-mia Diagnosis in Blood Slides Using Transfer Learning in CNNs and SVM for Classification. Engineering Applications of Artificial Inte-lligence, 2018, 72: 415-422.
2019, Vol. 32 
