{"type":"ScholarlyArticle","title":"Semi-Supervised Optimal Transport for Heterogeneous Domain Adaptation","authors":[{"name":"Yuguang Yan","slug":"yuguang-yan","given":"Yuguang","family":"Yan","affiliation":"School of Software Engineering , South China University of Technology , China"},{"name":"Wen Li","slug":"wen-li","given":"Wen","family":"Li","affiliation":"Computer Vision Laboratory , ETH Zurich , Switzerland"},{"name":"Hanrui Wu","slug":"hanrui-wu","given":"Hanrui","family":"Wu","affiliation":"School of Software Engineering , South China University of Technology , China"},{"name":"Huaqing Min","slug":"huaqing-min","given":"Huaqing","family":"Min","affiliation":"School of Software Engineering , South China University of Technology , China"},{"name":"Mingkui Tan","slug":"mingkui-tan","given":"Mingkui","family":"Tan","affiliation":"School of Software Engineering , South China University of Technology , China"},{"name":"Qingyao Wu","slug":"qingyao-wu","given":"Qingyao","family":"Wu","affiliation":"School of Software Engineering , South China University of Technology , China"}],"abstract":"Heterogeneous domain adaptation (HDA) aims to exploit knowledge from a heterogeneous source domain to improve the learning performance in a target domain. Since the feature spaces of the source and target domains are different, the transferring of knowledge is extremely difficult. In this paper, we propose a novel semi-supervised algorithm for HDA by exploiting the theory of optimal transport (OT), a powerful tool originally designed for aligning two different distributions. To match the samples between heterogeneous domains, we propose to preserve the semantic consistency between heterogeneous domains by incorporating label information into the entropic Gromov-Wasserstein discrepancy, which is a metric in OT for different metric spaces, resulting in a new semi-supervised scheme. Via the new scheme, the target and transported source samples with the same label are enforced to follow similar distributions. Lastly, based on the Kullback-Leibler metric, we develop an efficient algorithm to optimize the resultant problem. Comprehensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of our proposed method.","doi":"10.24963/ijcai.2018/412","url":"https://ijcai.org/proceedings/2018/412","pdf":"https://ijcai.org/proceedings/2018/0412.pdf","markdown":"https://ijcai.org/proceedings/2018/412.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/412","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Machine Learning","track":"Main track","number":412,"pages":{"first":2969,"last":2975},"keywords":["Machine Learning: Transfer, Adaptation, Multi-task Learning","Machine Learning Applications: Applications of Supervised Learning"],"videos":[],"references":[{"position":1,"raw":"[Amini et al., 2009] Massih Amini, Nicolas Usunier, and Cyril Goutte. Learning from multiple partially observed views-an ap- plication to multilingual text categorization. In NIPS, pages 28- 36, 2009.","title":"Learning from multiple partially observed views-an application to multilingual text categorization","authors":"Massih Amini, Nicolas Usunier, Cyril Goutte","year":"2009","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bay et al., 2006] Herbert Bay, Tinne Tuytelaars, and Luc Van Gool. Surf: Speeded up robust features. In ECCV, pages 404-417, 2006.","title":"Surf: Speeded up robust features","authors":"Herbert Bay, Tinne Tuytelaars, Luc Van Gool","year":"2006","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Benamou et al., 2015] Jean-David Benamou, Guillaume Carlier, Marco Cuturi, Luca Nenna, and Gabriel Peyré. Iterative breg- man projections for regularized transportation problems. SIAM Journal on Scientific Computing, 37(2):A1111-A1138, 2015.","title":"Iterative bregman projections for regularized transportation problems","authors":"Jean-David Benamou, Guillaume Carlier, Marco Cuturi, Luca Nenna, Gabriel Peyré","year":"2015","venue":"SIAM Journal on Scientific Computing","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chang and Lin, 2011] Chih-Chung Chang and Chih-Jen Lin. Lib- svm: a library for support vector machines. TIST, 2(3):27, 2011.","title":"Libsvm: a library for support vector machines","authors":"Chang, Lin ; ] Chih-Chung Chang, Chih-Jen Lin","year":"2011","venue":"TIST","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Courty et al., 2017] Nicolas Courty, Rémi Flamary, Devis Tuia, and Alain Rakotomamonjy. Optimal transport for domain adap- tation. TPAMI, 39(9):1853-1865, 2017.","title":"Optimal transport for domain adaptation","authors":"Nicolas Courty, Rémi Flamary, Devis Tuia, Alain Rakotomamonjy","year":"2017","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Cuturi, 2013] Marco Cuturi. Sinkhorn distances: Lightspeed com- putation of optimal transport. In NIPS, pages 2292-2300, 2013.","title":"Sinkhorn distances: Lightspeed computation of optimal transport","authors":"Marco Cuturi","year":"2013","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Donahue et al., 2014] Jeff Donahue, Yangqing Jia, Oriol Vinyals, Judy Hoffman, Ning Zhang, Eric Tzeng, and Trevor Darrell. De- caf: A deep convolutional activation feature for generic visual recognition. In ICML, volume 32, pages 647-655, 2014.","title":"Decaf: A deep convolutional activation feature for generic visual recognition","authors":"Jeff Donahue, Yangqing Jia, Oriol Vinyals, Judy Hoffman, Ning Zhang, Eric Tzeng, Trevor Darrell","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Duan et al., 2012] Lixin Duan, Dong Xu, and Ivor W Tsang. Learning with augmented features for heterogeneous domain adaptation. In ICML, pages 711-718, 2012.","title":"Learning with augmented features for heterogeneous domain adaptation","authors":"Lixin Duan, Dong Xu, Ivor W Tsang","year":"2012","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Griffin et al., 2007] Gregory Griffin, Alex Holub, and Pietro Per- ona. Caltech-256 object category dataset. California Institute of Technology, 2007.","title":"Caltech-256 object category dataset","authors":"Gregory Griffin, Alex Holub, Pietro Perona","year":"2007","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Li et al., 2014] Wen Li, Lixin Duan, Dong Xu, and Ivor W Tsang. Learning with augmented features for supervised and semi-supervised heterogeneous domain adaptation. TPAMI, 36(6):1134-1148, 2014.","title":"Learning with augmented features for supervised and semi-supervised heterogeneous domain adaptation","authors":"Wen Li, Lixin Duan, Dong Xu, Ivor W Tsang","year":"2014","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Liu et al., 2017] Tongliang Liu, Qiang Yang, and Dacheng Tao. Understanding how feature structure transfers in transfer learn- ing. In IJCAI, pages 2365-2371, 2017.","title":"Understanding how feature structure transfers in transfer learning","authors":"Liu","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2017/329"},{"position":12,"raw":"[Long et al., 2014] Mingsheng Long, Jianmin Wang, Guiguang Ding, Jiaguang Sun, and Philip S Yu. Transfer joint matching for unsupervised domain adaptation. In CVPR, pages 1410-1417, 2014.","title":"Transfer joint matching for unsupervised domain adaptation","authors":"Mingsheng Long, Jianmin Wang, Guiguang Ding, Jiaguang Sun, Philip S Yu","year":"2014","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Long et al., 2015] Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. Learning transferable features with deep adap- tation networks. In ICML, pages 97-105, 2015.","title":"Learning transferable features with deep adaptation networks","authors":"Mingsheng Long, Yue Cao, Jianmin Wang, Michael Jordan","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Luo et al., 2017] Yong Luo, Yonggang Wen, Tongliang Liu, and Dacheng Tao. General heterogeneous transfer distance met- ric learning via knowledge fragments transfer. In IJCAI, pages 2450-2456, 2017.","title":"General heterogeneous transfer distance metric learning via knowledge fragments transfer","authors":"Yong Luo, Yonggang Wen, Tongliang Liu, Dacheng Tao","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2017/341"},{"position":15,"raw":"[Moon and Carbonell, 2017] Seungwhan Moon and Jaime Car- bonell. Completely heterogeneous transfer learning with attention-what and what not to transfer. In IJCAI, pages 2508- 2514, 2017.","title":"Seungwhan Moon and Jaime Carbonell. Completely heterogeneous transfer learning with attention-what and what not to transfer","authors":"Carbonell Moon","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Pan and Yang, 2010] Sinno Jialin Pan and Qiang Yang. A survey on transfer learning. TKDE, 22(10):1345-1359, 2010.","title":"A survey on transfer learning","authors":"Jialin Sinno, Qiang Pan, Yang","year":"2010","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Pan et al., 2011] Sinno Jialin Pan, Ivor W Tsang, James T Kwok, and Qiang Yang. Domain adaptation via transfer component anal- ysis. TNN, 22(2):199-210, 2011.","title":"Domain adaptation via transfer component analysis","authors":"Ivor W Sinno Jialin Pan, James T Tsang, Qiang Kwok, Yang","year":"2011","venue":"TNN","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Patel et al., 2015] Vishal M Patel, Raghuraman Gopalan, Ruonan Li, and Rama Chellappa. Visual domain adaptation: A survey of recent advances. IEEE Signal Processing Magazine, 32(3):53- 69, 2015.","title":"Visual domain adaptation: A survey of recent advances","authors":"M Vishal, Raghuraman Patel, Ruonan Gopalan, Rama Li, Chellappa","year":"2015","venue":"IEEE Signal Processing Magazine","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Perrot et al., 2016] Michaël Perrot, Nicolas Courty, Rémi Flamary, and Amaury Habrard. Mapping estimation for discrete optimal transport. In NIPS, pages 4197-4205, 2016.","title":"Mapping estimation for discrete optimal transport","authors":"Michaël Perrot, Nicolas Courty, Rémi Flamary, Amaury Habrard","year":"2016","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Peyré and Cuturi, 2017] Gabriel Peyré and Marco Cuturi. Compu- tational optimal transport, 2017.","title":"Computational optimal transport","authors":"Gabriel Peyré, Marco Cuturi","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Peyré et al., 2016] Gabriel Peyré, Marco Cuturi, and Justin Solomon. Gromov-wasserstein averaging of kernel and distance matrices. In ICML, pages 2664-2672, 2016.","title":"Gromov-wasserstein averaging of kernel and distance matrices","authors":"Gabriel Peyré, Marco Cuturi, Justin Solomon","year":"2016","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Redko et al., 2017] Ievgen Redko, Amaury Habrard, and Marc Sebban. Theoretical analysis of domain adaptation with optimal transport. In ECML-PKDD, pages 737-753, 2017.","title":"Ievgen Redko, Amaury Habrard, and Marc Sebban. Theoretical analysis of domain adaptation with optimal transport","authors":"Redko","year":"2017","venue":"ECML-PKDD","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Saenko et al., 2010] Kate Saenko, Brian Kulis, Mario Fritz, and Trevor Darrell. Adapting visual category models to new domains. In ECCV, pages 213-226, 2010.","title":"Adapting visual category models to new domains","authors":"Saenko","year":"2010","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Shao et al., 2015] Ling Shao, Fan Zhu, and Xuelong Li. Trans- fer learning for visual categorization: A survey. TNNLS, 26(5):1019-1034, 2015.","title":"Transfer learning for visual categorization: A survey","authors":"Ling Shao, Fan Zhu, Xuelong Li","year":"2015","venue":"TNNLS","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Tan et al., 2015] Ben Tan, Yangqiu Song, Erheng Zhong, and Qiang Yang. Transitive transfer learning. In KDD, pages 1155- 1164, 2015.","title":"Transitive transfer learning","authors":"Ben Tan, Yangqiu Song, Erheng Zhong, Qiang Yang","year":"2015","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Tsai et al., 2016] Yao-Hung Hubert Tsai, Yi-Ren Yeh, and Yu- Chiang Frank Wang. Learning cross-domain landmarks for het- erogeneous domain adaptation. In CVPR, pages 5081-5090, 2016.","title":"Learning cross-domain landmarks for heterogeneous domain adaptation","authors":"Yao-Hung Hubert Tsai, Yi-Ren Yeh, Yu-Chiang Frank, Wang","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Wu et al., 2014] Qingyao Wu, Michael K. Ng, and Yunming Ye. Cotransfer learning using coupled markov chains with restart. IEEE Intelligent Systems, 29(4):26 -33, 2014.","title":"Cotransfer learning using coupled markov chains with restart","authors":"Qingyao Wu, Michael K Ng, Yunming Ye","year":"2014","venue":"IEEE Intelligent Systems","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Yan et al., 2017a] Hongliang Yan, Yukang Ding, Peihua Li, Qi- long Wang, Yong Xu, and Wangmeng Zuo. Mind the class weight bias: Weighted maximum mean discrepancy for unsupervised do- main adaptation. In CVPR, pages 2272-2281, 2017.","title":"Mind the class weight bias: Weighted maximum mean discrepancy for unsupervised domain adaptation","authors":"Yan","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Yan et al., 2017b] Yuguang Yan, Wen Li, Michael Ng, Mingkui Tan, Hanrui Wu, Huaqing Min, and Qingyao Wu. Learning dis- criminative correlation subspace for heterogeneous domain adap- tation. In IJCAI, pages 3252-3258, 2017.","title":"Learning discriminative correlation subspace for heterogeneous domain adaptation","authors":"Yan","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2017/454"},{"position":30,"raw":"[Yan et al., 2017c] Yuguang Yan, Qingyao Wu, Mingkui Tan, Michael K Ng, Huaqing Min, and Ivor W Tsang. Online het- erogeneous transfer by hedge ensemble of offline and online de- cisions. TNNLS, pages 1-12, 2017.","title":"Online heterogeneous transfer by hedge ensemble of offline and online decisions","authors":"Yan","year":"2017","venue":"TNNLS","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhou et al., 2014] Joey Tianyi Zhou, Sinno Jialin Pan, Ivor W Tsang, and Yan Yan. Hybrid heterogeneous transfer learning through deep learning. In AAAI, pages 2213-2220, 2014.","title":"Hybrid heterogeneous transfer learning through deep learning","authors":"Joey Tianyi Zhou, Sinno Jialin Pan, Ivor W Tsang, Yan Yan","year":"2014","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zhou et al., 2016] Joey Tianyi Zhou, Sinno Jialin Pan, Ivor W Tsang, and Shen-Shyang Ho. Transfer learning for cross- language text categorization through active correspondences con- struction. In AAAI, pages 2400-2406, 2016.","title":"Transfer learning for crosslanguage text categorization through active correspondences construction","authors":"Joey Tianyi Zhou, Sinno Jialin Pan, Ivor W Tsang, Shen-Shyang Ho","year":"2016","venue":"AAAI","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}