Transfer Learning with Joint Inter-class and Inter-domain Distributional Adaptation
LI Ping1,2,3, NI Zhiwei1,3, ZHU Xuhui1,3 , SONG Juan1,3
1. School of Management, Hefei University of Technology, Hefei 230009; 2. College of Information Engineering, Fuyang Normal University, Fuyang 236041; 3. Key Laboratory of Process Optimization and Intelligent Decision-Making, Ministry of Education, Hefei University of Technology, Hefei 230009
Abstract：Inter-domain information is lost during the process of inter-domain distributional adaptation. Therefore, it is difficult to train an effective classifier in the source domain, and the performance of generalization and tagging in the target domain are affected. Aiming at this problem, an approach, joint inter-class and inter-domain distributional adaptation for transfer learning, is proposed to address this challenge. The proposed method is formulated by learning a projection matrix to map new representations of respective domains into a common subspace. And the distance-measure method of the maximum mean discrepancy is adopted to compute the distance of inter-class and inter-domain distributions. During the optimization procedure, the inter-domain distributional difference is reduced explicitly, and the inter-class distributional difference is enlarged greatly. The capability of knowledge transfer between different domains is improved. Experiments on transfer learning dataset verify the effectiveness of the proposed approach.
[1] DAI W Y, YANG Q, XUE G R, et al. Boosting for Transfer Lear-ning // Proc of the 24th International Conference on Machine Lear-ning. New York, USA: ACM, 2007: 193-200. [2] 张 倩,李 明,王雪松, 等.一种面向多源领域的实例迁移学习.自动化学报, 2014, 40(6): 1176-1183. (ZHANG Q, LI M, WANG X S, et al. Instance-Based Transfer Learning for Multi-source Domains. Acta Automatica Sinica, 2014, 40(6): 1176-1183.) [3] JIANG S Y, XU Y H, SONG H J, et al. Multi-instance Transfer Metric Learning by Weighted Distribution and Consistent Maximum Likelihood Estimation. Neurocomputing, 2018, 321: 49-60. [4] 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. [5] 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. [6] 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. [7] LONG M S, CAO Y, WANG J M, et al. Learning Transferable Features with Deep Adaptation Networks // Proc of the 32nd International Conference on Machine Learning. Washington, USA: IEEE, 2015: 97-105. [8] PAN S J, SHEN D, YANG Q, et al. Transferring Localization Models across Space // Proc of the 23rd National Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2008, III: 1383-1388. [9] ZHAO Z T, CHEN Y Q, LIU J F, et al. Cross-People Mobile-Phone Based Activity Recognition // Proc of the 22nd International Joint Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2011, III: 2545-2250. [10] WEI Y, ZHU Y, LEUNG C W, et al. Instilling Social to Physical: Co-regularized Heterogeneous Transfer Learning // Proc of the 30th AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2016: 1338-1344. [11] MIHALKOVA L, HUYNH T, MOONEY R J. Mapping and Revising Markov Logic Networks for Transfer Learning // Proc of the 22nd National Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2007, I: 608-614. [12] MIHALKOVA L, MOONEY R J. Transfer Learning by Mapping with Minimal Target Data // Proc of the AAAI Workshop on Transfer Learning for Complex Tasks. Palo Alto, USA: AAAI Press, 2008: 31-36. [13] DAVIS J, DOMINGOS P. Deep Transfer via Second-Order Markov Logic // Proc of the 26th International Conference on Machine Learning. New York, USA: ACM, 2009: 217-224. [14] ZHANG J, LI W Q, OGUNBONA P. Joint Geometrical and Statistical Alignment for Visual Domain Adaptation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 5150-5158. [15] CAO Z J, LONG M S, WANG J M, et al. Partial Transfer Lear-ning with Selective Adversarial Networks // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 2724-2732. [16] CAO Z J, MA L J, LONG M S, et al. Partial Adversarial Domain Adaptation // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 139-155. [17] HAUSSER P, FRERIX T, MORDVINTSEV A, et al. Associative Domain Adaptation // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2017: 2784-2792. [18] HONG W X, WANG Z Z, YANG M, et al. Conditional Generative Adversarial Network for Structured Domain Adaptation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 1335-1344. [19] LONG M S, CAO Z J, WANG J M, et al. Conditional Adversarial Domain Adaptation // Proc of the 32nd Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2018: 1647-1657. [20] MOTIIAN S, PICCIRILLI M, ADJEROH D A, et al. Unified Deep Supervised Domain Adaptation and Generalization // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2017: 5716-5726. [21] PEI Z Y, CAO Z J, LONG M S, et al. Multi-adversarial Domain Adaption // Proc of the 32nd AAAI Conference on Artificial Inte-lligence. Palo Alto, USA: AAAI Press, 2018: 3934-3941. [22] ZHANG J, DING Z W, LI W Q, et al. Importance Weighted Adversarial Nets for Partial Domain Adaptation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 8156-8164. [23] LONG M S, WANG J M, JORDAN M I. Deep Transfer Learning with Joint Adaptation Networks // Proc of the 34th International Conference on Machine Learning. Washington, USA: IEEE, 2017: 2208-2217. [24] 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. [25] COURTY N, FLAMARY R, HABRARD A, et al. Joint Distribution Optimal Transportation for Domain Adaptation[C/OL]. [2019-01-25]. https://arxiv.org/pdf/1705.08848.pdf. [26] SUN B C, FENG J S, SAENKO K. Return of Frustratingly Easy Domain Adaptation[C/OL]. [2019-01-25]. https://arxiv.org/pdf/1511.05547.pdf. [27] BAKTASHMOTLAGH M, HARANDI M T, LOVELL B C, et al. Unsupervised Domain Adaptation by Domain Invariant Projection // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2013: 769-776. [28] HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 770-778. [29] LONG M S, ZHU H, WANG J M, et al. Unsupervised Domain Adaptation with Residual Transfer Networks // LEE D D, SUGIYAMA M, LUXBURG U V, et al., eds. Advances in Neural Information Processing Systems 29. Cambridge, USA: The MIT Press, 2016: 136-144.
2020, Vol. 33 
