{"type":"ScholarlyArticle","title":"Multi-Label Co-Training","authors":[{"name":"Yuying Xing","slug":"yuying-xing","given":"Yuying","family":"Xing","affiliation":"College of Computer and Information Science , Southwest University , Chongqing 400715 , China"},{"name":"Guoxian Yu","slug":"guoxian-yu","given":"Guoxian","family":"Yu","affiliation":"College of Computer and Information Science , Southwest University , Chongqing 400715 , China"},{"name":"Carlotta Domeniconi","slug":"carlotta-domeniconi","given":"Carlotta","family":"Domeniconi","affiliation":"Department of Computer Science , George Mason University , Fairfax 22030 , USA"},{"name":"Jun Wang","slug":"jun-wang","given":"Jun","family":"Wang","affiliation":"College of Computer and Information Science , Southwest University , Chongqing 400715 , China"},{"name":"Zili Zhang","slug":"zili-zhang","given":"Zili","family":"Zhang","affiliation":"College of Computer and Information Science , Southwest University , Chongqing 400715 , China; School of Information Technology , Deakin University , Geelong , VIC 3220 , Australia"}],"abstract":"Multi-label learning aims at assigning a set of appropriate labels to multi-label samples.  Although it has been successfully applied in various domains in recent years, most multi-label learning methods require sufficient labeled training samples, because of the large number of possible label sets.  Co-training, as an important branch of semi-supervised learning, can leverage unlabeled samples, along with scarce labeled ones, and can potentially help with the large labeled data requirement. However, it is a difficult challenge to combine multi-label learning with co-training. Two distinct issues are associated with the challenge: (i) how to solve the widely-witnessed class-imbalance problem in multi-label learning; and (ii) how to select samples with confidence, and  communicate their predicted labels among  classifiers for model refinement. To address these issues, we introduce an approach called Multi-Label Co-Training (MLCT). MLCT leverages information concerning the co-occurrence  of pairwise labels to address the class-imbalance challenge; it introduces a predictive reliability measure to select samples, and applies label-wise filtering to confidently communicate labels of selected samples among co-training classifiers.  MLCT performs favorably against related competitive multi-label learning methods on benchmark datasets and it is also robust to the input parameters.","doi":"10.24963/ijcai.2018/400","url":"https://ijcai.org/proceedings/2018/400","pdf":"https://ijcai.org/proceedings/2018/0400.pdf","markdown":"https://ijcai.org/proceedings/2018/400.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/400","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":400,"pages":{"first":2882,"last":2888},"keywords":["Machine Learning: Classification","Machine Learning: Semi-Supervised Learning","Machine Learning: Multi-instance;Multi-label;Multi-view learning"],"videos":[],"references":[{"position":1,"raw":"[Blum and Mitchell, 1998] Avrim Blum and Tom Mitchell. Combining labeled and unlabeled data with co-training. In COLT, pages 92-100, 1998.","title":"Combining labeled and unlabeled data with co-training","authors":"Avrim Blum, Tom Mitchell","year":"1998","venue":"COLT","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bucak et al., 2011] Serhat Selcuk Bucak, Rong Jin, and Anil K Jain. Multi-label learning with incomplete class assignments. In CVPR, pages 2801-2808, 2011.","title":"Multi-label learning with incomplete class assignments","authors":"Selcuk Serhat, Rong Bucak, Anil K Jin, Jain","year":"2011","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Chen et al., 2011] Minmin Chen, Yixin Chen, and Kilian Q Weinberger. Automatic feature decomposition for single view co-training. In ICML, pages 953-960, 2011.","title":"Automatic feature decomposition for single view co-training","authors":"Minmin Chen, Yixin Chen, Kilian Q Weinberger","year":"2011","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Demšar, 2006] Janez Demšar. Statistical comparisons of classifiers over multiple data sets. Journal of Machine Learning Research, 7(1):1-30, 2006.","title":"Janez Demšar. Statistical comparisons of classifiers over multiple data sets","authors":"Demšar","year":"2006","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Du et al., 2011] Jun Du, Charles X Ling, and Zhihua Zhou. When does cotraining work in real data? TKDE, 23(5):788- 799, 2011.","title":"When does cotraining work in real data?","authors":"Du","year":"2011","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Gibaja and Ventura, 2015] Eva Gibaja and Sebastián Ventu- ra. A tutorial on multilabel learning. ACM Computing Surveys, 47(3):52, 2015.","title":"Eva Gibaja and Sebastián Ventura. A tutorial on multilabel learning","authors":"Ventura Gibaja","year":"2015","venue":"ACM Computing Surveys","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gibaja et al., 2016] Eva L Gibaja, Jose M Moyano, and Se- bastián Ventura. An ensemble-based approach for multi- view multi-label classification. Progress in Artificial Intel- ligence, 5(4):251-259, 2016.","title":"An ensemble-based approach for multiview multi-label classification","authors":"Jose M Eva L Gibaja, Sebastián Moyano, Ventura","year":"2016","venue":"Progress in Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Gönen, 2014] Mehmet Gönen. Coupled dimensionality reduction and classification for supervised and semi- supervised multilabel learning. Pattern Recognition Letters, 38:132-141, 2014.","title":"Coupled dimensionality reduction and classification for supervised and semisupervised multilabel learning","authors":"Mehmet Gönen","year":"2014","venue":"Pattern Recognition Letters","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Guillaumin et al., 2010] Matthieu Guillaumin, Jakob Ver- beek, and Cordelia Schmid. Multimodal semi-supervised learning for image classification. In CVPR, pages 902-909, 2010.","title":"Multimodal semi-supervised learning for image classification","authors":"Matthieu Guillaumin, Jakob Verbeek, Cordelia Schmid","year":"2010","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Guo and Schuurmans, 2012] Yuhong Guo and Dale Schu- urmans. Semi-supervised multi-label classification: a si- multaneous large-margin, subspace learning approach. In ECML/PKDD, pages 355-370, 2012.","title":"Semi-supervised multi-label classification: a simultaneous large-margin, subspace learning approach","authors":"Schuurmans ; Guo, Dale Guo, Schuurmans","year":"2012","venue":"ECML/PKDD","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Huang and Zhou, 2012] Shengjun Huang and Zhihua Zhou. Multi-label learning by exploiting label correlations locally. In AAAI, pages 949-955, 2012.","title":"Multi-label learning by exploiting label correlations locally","authors":"Shengjun Huang, Zhihua Zhou","year":"2012","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kong et al., 2013] Xiangnan Kong, Michael K Ng, and Zhi- hua Zhou. Transductive multilabel learning via label set propagation. TKDE, 25(3):704-719, 2013.","title":"Transductive multilabel learning via label set propagation","authors":"Kong","year":"2013","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Levati et al., 2017] Jurica Levati, Michelangelo Ceci, Dragi Kocev, and Sao Deroski. Self-training for multi-target regression with tree ensembles. Knowledge-Based Systems, 123(C):41-60, 2017.","title":"Self-training for multi-target regression with tree ensembles","authors":"Jurica Levati, Michelangelo Ceci, Dragi Kocev, Sao Deroski","year":"2017","venue":"Knowledge-Based Systems","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Qian and Davidson, 2010] Buyue Qian and Ian Davidson. Semi-supervised dimension reduction for multi-label clas- sification. In AAAI, pages 569-574, 2010.","title":"Buyue Qian and Ian Davidson. Semi-supervised dimension reduction for multi-label classification","authors":"Davidson Qian","year":"2010","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Quevedo et al., 2012] José Ramón Quevedo, Oscar Luaces, and Antonio Bahamonde. Multilabel classifiers with a probabilistic thresholding strategy. Pattern Recognition, 45(2):876-883, 2012.","title":"José Ramón Quevedo, Oscar Luaces, and Antonio Bahamonde. Multilabel classifiers with a probabilistic thresholding strategy","authors":"Quevedo","year":"2012","venue":"Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Sun and Lee, 2017] Kaiwei Sun and Chong Ho Lee. Ad- dressing class-imbalance in multi-label learning via two- stage multi-label hypernetwork. Neurocomputing, 266:375- 389, 2017.","title":"Kaiwei Sun and Chong Ho Lee. Addressing class-imbalance in multi-label learning via twostage multi-label hypernetwork","authors":"Lee Sun","year":"2017","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Tan et al., 2017] Qiaoyu Tan, Yanming Yu, Guoxian Yu, and Jun Wang. Semi-supervised multi-label classification using incomplete label information. Neurocomputing, 260:192- 202, 2017.","title":"Semi-supervised multi-label classification using incomplete label information","authors":"Qiaoyu Tan, Yanming Yu, Guoxian Yu, Jun Wang","year":"2017","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Wu and Zhang, 2013] Le Wu and Minling Zhang. Multi- label classification with unlabeled data: An inductive ap- proach. In ACML, pages 197-212, 2013.","title":"Multilabel classification with unlabeled data: An inductive approach","authors":"Zhang ; ] Le Wu, Minling Wu, Zhang","year":"2013","venue":"ACML","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Wu et al., 2015] Baoyuan Wu, Siwei Lyu, and Bernard Ghanem. Ml-mg: multi-label learning with missing la- bels using a mixed graph. In ICCV, pages 4157-4165, 2015.","title":"Ml-mg: multi-label learning with missing labels using a mixed graph","authors":"Wu","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Yu et al., 2012] Guoxian Yu, Carlotta Domeniconi, Huzefa Rangwala, Guoji Zhang, and Zhiwen Yu. Transductive multi-label ensemble classification for protein function pre- diction. In KDD, pages 1077-1085, 2012.","title":"Transductive multi-label ensemble classification for protein function prediction","authors":"Yu","year":"2012","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Yu et al., 2014] Hsiang-Fu Yu, Prateek Jain, Purushottam Kar, and Inderjit Dhillon. Large-scale multi-label learning with missing labels. In ICML, pages 593-601, 2014.","title":"Large-scale multi-label learning with missing labels","authors":"Hsiang-Fu Yu, Prateek Jain, Purushottam Kar, Inderjit Dhillon","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Zhan and Zhang, 2017] Wang Zhan and Minling Zhang. Inductive semi-supervised multi-label learning with co- training. In KDD, pages 1305-1314, 2017.","title":"Inductive semi-supervised multi-label learning with cotraining","authors":"Wang Zhan, Minling Zhang","year":"2017","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhang and Yeung, 2009] Yu Zhang and Dit Yan Yeung. Semi-supervised multi-task regression. In ECML/PKDD, pages 617-631, 2009.","title":"Semi-supervised multi-task regression","authors":"Yu Zhang, Dit Yan, Yeung","year":"2009","venue":"ECML/PKDD","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhang and Zhou, 2006] Minling Zhang and Zhihua Zhou. Multilabel neural networks with applications to functional genomics and text categorization. TKDE, 18(10):1338- 1351, 2006.","title":"Multilabel neural networks with applications to functional genomics and text categorization","authors":"Zhou ; ] Minling Zhang, Zhihua Zhang, Zhou","year":"2006","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zhang and Zhou, 2007] Minling Zhang and Zhihua Zhou. Ml-knn: A lazy learning approach to multi-label learning. Pattern Recognition, 40(7):2038-2048, 2007.","title":"Ml-knn: A lazy learning approach to multi-label learning","authors":"Zhou ; ] Minling Zhang, Zhihua Zhang, Zhou","year":"2007","venue":"Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang and Zhou, 2014] Minling Zhang and Zhihua Zhou. A review on multi-label learning algorithms. TKDE, 26(8):1819-1837, 2014.","title":"A review on multi-label learning algorithms","authors":"Zhou Zhang, Minling Zhang, Zhihua Zhou","year":"2014","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang et al., 2015a] Minling Zhang, Yukun Li, and Xuying Liu. Towards class-imbalance aware multi-label learning. In IJCAI, pages 4041-4047, 2015.","title":"Towards class-imbalance aware multi-label learning","authors":"Zhang","year":"2015","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2015/567"},{"position":28,"raw":"[Zhang et al., 2015b] Xiang Zhang, Naiyang Guan, Zhigang Luo, and Xuejun Yang. Constrained projective non- negative matrix factorization for semi-supervised multi- label learning. In ICMLA, pages 588-593, 2015.","title":"Constrained projective nonnegative matrix factorization for semi-supervised multilabel learning","authors":"Zhang","year":"2015","venue":"ICMLA","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhou and Li, 2005] Zhihua Zhou and Ming Li. Tri-training: Exploiting unlabeled data using three classifiers. TKDE, 17(11):1529-1541, 2005.","title":"Tri-training: Exploiting unlabeled data using three classifiers","authors":"Zhihua Zhou, Ming Li","year":"2005","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhu, 2008] Xiaojin Zhu. Semi-supervised learning literature survey. Technical report, University of Wisconsin-Madison, Tech. Rep.: 1530, 2008.","title":"Semi-supervised learning literature survey","authors":"Xiaojin Zhu","year":"2008","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}