{"type":"ScholarlyArticle","title":"Instance-Level Label Propagation with Multi-Instance Learning","authors":[{"name":"Qifan Wang","slug":"qifan-wang","given":"Qifan","family":"Wang","affiliation":"Google Research Mountain View , CA 94043 , US"},{"name":"Gal Chechik","slug":"gal-chechik","given":"Gal","family":"Chechik","affiliation":"Google Research Mountain View , CA 94043 , US"},{"name":"Chen Sun","slug":"chen-sun","given":"Chen","family":"Sun","affiliation":"Google Research Mountain View , CA 94043 , US"},{"name":"Bin Shen","slug":"bin-shen","given":"Bin","family":"Shen","affiliation":"Google Research Mountain View , CA 94043 , US"}],"abstract":"Label propagation is a popular semi-supervised learning technique that transfers information from labeled examples to unlabeled examples through a graph. Most label propagation methods construct a graph based on example-to-example similarity, assuming that the resulting graph connects examples that share similar labels. Unfortunately, example-level similarity is sometimes badly defined. For instance, two images may contain two different objects, but have similar overall appearance due to large similar background. In this case, computing similarities based on whole-image would fail propagating information to the right labels. This paper proposes a novel Instance-Level Label Propagation (ILLP) approach that integrates label propagation with multi-instance learning. Each example is treated as containing multiple instances, as in the case of an image consisting of multiple regions. We first construct a graph based on instance-level similarity and then simultaneously identify the instances carrying the labels and propagate the labels across instances in the graph. Optimization is based on an iterative Expectation Maximization (EM) algorithm.  Experimental results on two benchmark datasets demonstrate the effectiveness of the proposed approach over several state-of-the-art methods.","doi":"10.24963/ijcai.2017/410","url":"https://ijcai.org/proceedings/2017/410","pdf":"https://ijcai.org/proceedings/2017/0410.pdf","markdown":"https://ijcai.org/proceedings/2017/410.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/410","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Machine Learning S-Z","track":"Main track","number":410,"pages":{"first":2943,"last":2949},"keywords":["Machine Learning: Semi-Supervised Learning","Machine Learning: Machine Learning","Machine Learning: Multi-instance/Multi-label/Multi-view learning"],"videos":[],"references":[{"position":1,"raw":"[Amores, 2013] Jaume Amores. Multiple instance classification: Review, taxonomy and comparative study. Artif. Intell., 201:81- 105, 2013.","title":"Jaume Amores. Multiple instance classification: Review, taxonomy and comparative study","authors":"Amores","year":"2013","venue":"Artif. Intell","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Andrews et al., 2002] Stuart Andrews, Ioannis Tsochantaridis, and Thomas Hofmann. Support vector machines for multiple- instance learning. In NIPS, pages 561-568, 2002.","title":"Support vector machines for multipleinstance learning","authors":"Stuart Andrews, Ioannis Tsochantaridis, Thomas Hofmann","year":"2002","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Belkin et al., 2006] Mikhail Belkin, Partha Niyogi, and Vikas Sindhwani. Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. Journal of Machine Learning Research, 7:2399-2434, 2006.","title":"Manifold regularization: A geometric framework for learning from labeled and unlabeled examples","authors":"Mikhail Belkin, Partha Niyogi, Vikas Sindhwani","year":"2006","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chapelle et al., 2010] Olivier Chapelle, Bernhard Schlkopf, and Alexander Zien. Semi-Supervised Learning. The MIT Press, 1st edition, 2010.","title":"Semi-Supervised Learning","authors":"Chapelle","year":"2010","venue":null,"doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Chen et al., 2006] Yixin Chen, Jinbo Bi, and James Ze Wang. Miles: Multiple-instance learning via embedded instance selec- tion. IEEE Trans. Pattern Anal. Mach. Intell., 28(12):1931-1947, 2006.","title":"Miles: Multiple-instance learning via embedded instance selection","authors":"Yixin Chen, Jinbo Bi, James Ze, Wang","year":"2006","venue":"IEEE Trans. Pattern Anal. Mach. Intell","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Dietterich et al., 1997] Thomas G. Dietterich, Richard H. Lathrop, and Tomás Lozano-Pérez. Solving the multiple instance problem with axis-parallel rectangles. Artif. Intell., 89(1-2):31-71, 1997.","title":"Solving the multiple instance problem with axis-parallel rectangles","authors":"G Thomas, Richard H Dietterich, Tomás Lathrop, Lozano-Pérez","year":"1997","venue":"Artif. Intell","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Ding and Riloff, 2016] Haibo Ding and Ellen Riloff. Acquiring knowledge of affective events from blogs using label propaga- tion. In AAAI, pages 2935-2942, 2016.","title":"Acquiring knowledge of affective events from blogs using label propagation","authors":"Haibo Ding, Ellen Riloff","year":"2016","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Fu and Robles-Kelly, 2009] Zhouyu Fu and Antonio Robles- Kelly. An instance selection approach to multiple instance learning. In CVPR, pages 911-918, 2009.","title":"Zhouyu Fu and Antonio Robles-Kelly. An instance selection approach to multiple instance learning","authors":"Robles-Kelly Fu","year":"2009","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Fujiwara and Irie, 2014] Yasuhiro Fujiwara and Go Irie. Efficient label propagation. In ICML, pages 784-792, 2014.","title":"Efficient label propagation","authors":"Irie Fujiwara, Yasuhiro Fujiwara, Go Irie","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Golub and Van Loan, 2012] Gene H Golub and Charles F Van Loan. Matrix computations, volume 3. JHU Press, 2012.","title":"Matrix computations","authors":"Van Golub, ; Loan, H Gene, Charles F Golub, Van Loan","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Gong et al., 2015] Chen Gong, Dacheng Tao, Keren Fu, and Jie Yang. Fick's law assisted propagation for semisupervised learning. IEEE Trans. Neural Netw. Learning Syst., 26(9):2148- 2162, 2015.","title":"Fick's law assisted propagation for semisupervised learning","authors":"Chen Gong, Dacheng Tao, Keren Fu, Jie Yang","year":"2015","venue":"IEEE Trans. Neural Netw. Learning Syst","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Gong et al., 2016] Chen Gong, Dacheng Tao, Jie Yang, and Wei Liu. Teaching-to-learn and learning-to-teach for multi-label propagation. In AAAI, pages 1610-1616, 2016.","title":"Teaching-to-learn and learning-to-teach for multi-label propagation","authors":"Chen Gong, Dacheng Tao, Jie Yang, Wei Liu","year":"2016","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Hadiji et al., 2015] Fabian Hadiji, Martin Mladenov, Christian Bauckhage, and Kristian Kersting. Computer science on the move: Inferring migration regularities from the web via compressed label propagation. In IJCAI, pages 171-177, 2015.","title":"Computer science on the move: Inferring migration regularities from the web via compressed label propagation","authors":"Fabian Hadiji, Martin Mladenov, Christian Bauckhage, Kristian Kersting","year":"2015","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2015/31"},{"position":14,"raw":"[Jia and Zhang, 2008] Yangqing Jia and Changshui Zhang. Instance-level semisupervised multiple instance learning. In AAAI, pages 640-645, 2008.","title":"Instance-level semisupervised multiple instance learning","authors":"Zhang Jia, Yangqing Jia, Changshui Zhang","year":"2008","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Jin et al., 2007] Rong Jin, Ming Wu, and Rahul Sukthankar. Semi- supervised collaborative text classification. In ECML, pages 600- 607, 2007.","title":"Semisupervised collaborative text classification","authors":"Rong Jin, Ming Wu, Rahul Sukthankar","year":"2007","venue":"ECML","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kang et al., 2006] Feng Kang, Rong Jin, and Rahul Sukthankar. Correlated label propagation with application to multi-label learning. In CVPR, pages 1719-1726, 2006.","title":"Correlated label propagation with application to multi-label learning","authors":"Feng Kang, Rong Jin, Rahul Sukthankar","year":"2006","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Karasuyama and Mamitsuka, 2013] Masayuki Karasuyama and Hiroshi Mamitsuka. Manifold-based similarity adaptation for label propagation. In NIPS, pages 1547-1555, 2013.","title":"Manifold-based similarity adaptation for label propagation","authors":"Mamitsuka Karasuyama, Masayuki Karasuyama, Hiroshi Mamitsuka","year":"2013","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Kim et al., 2009] Soo-Min Kim, Patrick Pantel, Lei Duan, and Scott Gaffney. Improving web page classification by label- propagation over click graphs. In CIKM, pages 1077-1086, 2009.","title":"Improving web page classification by labelpropagation over click graphs","authors":"Kim","year":"2009","venue":"CIKM","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Kim et al., 2015] Kwang In Kim, James Tompkin, Hanspeter Pfister, and Christian Theobalt. Context-guided diffusion for label propagation on graphs. In ICCV, pages 2776-2784, 2015.","title":"Context-guided diffusion for label propagation on graphs","authors":"In Kwang, James Kim, Hanspeter Tompkin, Christian Pfister, Theobalt","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Maron and Lozano-Pérez, 1997] Oded Maron and Tomás Lozano- Pérez. A framework for multiple-instance learning. In NIPS, 1997.","title":"A framework for multiple-instance learning","authors":"Lozano-Pérez ; Maron, Oded Maron, Tomás Lozano-Pérez","year":"1997","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Rahmani and Goldman, 2006] Rouhollah Rahmani and Sally A. Goldman. MISSL: multiple-instance semi-supervised learning. In ICML, pages 705-712, 2006.","title":"MISSL: multiple-instance semi-supervised learning","authors":"Goldman Rahmani, Rouhollah Rahmani, Sally A Goldman","year":"2006","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Shi and Malik, 2000] Jianbo Shi and Jitendra Malik. Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Mach. Intell., 22(8):888-905, 2000.","title":"Normalized cuts and image segmentation","authors":"Jianbo Shi, Jitendra Malik","year":"2000","venue":"IEEE Trans. Pattern Anal. Mach. Intell","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Strelow et al., 2016] Dennis Strelow, Qifan Wang, Luo Si, and Anders Eriksson. General, nested, and constrained wiberg minimization. IEEE Trans. Pattern Anal. Mach. Intell., 38(9):1803-1815, 2016.","title":"General, nested, and constrained wiberg minimization","authors":"Dennis Strelow, Qifan Wang, Luo Si, Anders Eriksson","year":"2016","venue":"IEEE Trans. Pattern Anal. Mach. Intell","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[von Luxburg, 2007] Ulrike von Luxburg. A tutorial on spectral clustering. Statistics and Computing, 17(4):395-416, 2007.","title":"A tutorial on spectral clustering","authors":"Ulrike Von, Luxburg","year":"2007","venue":"Statistics and Computing","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Wang and Tsotsos, 2016] Bo Wang and John K. Tsotsos. Dynamic label propagation for semi-supervised multi-class multi-label classification. Pattern Recognition, 52:75-84, 2016.","title":"Dynamic label propagation for semi-supervised multi-class multi-label classification","authors":"Bo Wang, John K Tsotsos","year":"2016","venue":"Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Wang and Zhang, 2006] Fei Wang and Changshui Zhang. Label propagation through linear neighborhoods. In ICML, pages 985- 992, 2006.","title":"Label propagation through linear neighborhoods","authors":"Wang, Zhang, Fei Wang, Changshui Zhang","year":"2006","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Wang et al., 2012] Qifan Wang, Luo Si, and Dan Zhang. A dis- criminative data-dependent mixture-model approach for multiple instance learning in image classification. In ECCV (4), pages 660-673, 2012.","title":"A discriminative data-dependent mixture-model approach for multiple instance learning in image classification","authors":"Qifan Wang, Luo Si, Dan Zhang","year":"2012","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Wang et al., 2014] Qifan Wang, Lingyun Ruan, and Luo Si. Adaptive knowledge transfer for multiple instance learning in image classification. In AAAI, pages 1334-1340, 2014.","title":"Adaptive knowledge transfer for multiple instance learning in image classification","authors":"Wang","year":"2014","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Weiss et al., 2008] Yair Weiss, Antonio Torralba, and Robert Fergus. Spectral hashing. In NIPS, pages 1753-1760, 2008.","title":"Spectral hashing","authors":"Yair Weiss, Antonio Torralba, Robert Fergus","year":"2008","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zelnik-Manor and Perona, 2004] Lihi Zelnik-Manor and Pietro Perona. Self-tuning spectral clustering. In NIPS, 2004.","title":"Self-tuning spectral clustering","authors":"Lihi Zelnik, -Manor, Pietro Perona","year":"2004","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Zhang and Goldman, 2001] Qi Zhang and Sally A. Goldman. Em- dd: An improved multiple-instance learning technique. In NIPS, pages 1073-1080, 2001.","title":"Emdd: An improved multiple-instance learning technique","authors":"Goldman Zhang, Qi Zhang, Sally A Goldman","year":"2001","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zhang and Zhou, 2009] Yin Zhang and Zhi-Hua Zhou. Non- metric label propagation. In IJCAI, pages 1357-1362, 2009.","title":"Nonmetric label propagation","authors":"Yin Zhang, Zhi-Hua Zhou","year":"2009","venue":"IJCAI","doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Zhang et al., 2013] Dan Zhang, Jingrui He, and Richard D. Lawrence. M2ls: Multi-instance learning from multiple infor- mation sources. In KDD, 2013.","title":"M2ls: Multi-instance learning from multiple information sources","authors":"Dan Zhang, Jingrui He, Richard D Lawrence","year":"2013","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Zhou and Xu, 2007] Zhi-Hua Zhou and Jun-Ming Xu. On the relation between multi-instance learning and semi-supervised learning. In ICML, pages 1167-1174, 2007.","title":"On the relation between multi-instance learning and semi-supervised learning","authors":"Xu ; Zhi-Hua Zhou, Jun-Ming Zhou, Xu","year":"2007","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[Zhou et al., 2003] Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal, Jason Weston, and Bernhard Schölkopf. Learning with local and global consistency. In NIPS, pages 321-328, 2003.","title":"Learning with local and global consistency","authors":"Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal, Jason Weston, Bernhard Schölkopf","year":"2003","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":36,"raw":"[Zhou, 2004] Zhi-Hua Zhou. Multi-Instance learning: A survey. Technical report, 2004.","title":"Multi-Instance learning: A survey","authors":"Zhi-Hua Zhou","year":"2004","venue":null,"doi":null,"url":null,"internal_url":null},{"position":37,"raw":"[Zhu and Ghahramani, 2002] Xiaojin Zhu and Zoubin Ghahra- mani. Learning from labeled and unlabeled data with label propagation. Technical report, 2002.","title":"Learning from labeled and unlabeled data with label propagation","authors":"Xiaojin Zhu, Zoubin Ghahramani","year":"2002","venue":null,"doi":null,"url":null,"internal_url":null},{"position":38,"raw":"[Zhu, 2006] Xiaojin Zhu. Semi-supervised learning literature survey, 2006.","title":"Semi-supervised learning literature survey","authors":"Xiaojin Zhu","year":"2006","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}