{"type":"ScholarlyArticle","title":"Self-Supervised Deep Low-Rank Assignment Model for Prototype Selection","authors":[{"name":"Xingxing Zhang","slug":"xingxing-zhang","given":"Xingxing","family":"Zhang","affiliation":"Institute of Information Science , Beijing Jiaotong University , Beijing , China"},{"name":"Zhenfeng Zhu","slug":"zhenfeng-zhu","given":"Zhenfeng","family":"Zhu","affiliation":"Institute of Information Science , Beijing Jiaotong University , Beijing , China"},{"name":"Yao Zhao","slug":"yao-zhao","given":"Yao","family":"Zhao","affiliation":"Institute of Information Science , Beijing Jiaotong University , Beijing , China; Beijing Key Laboratory of Advanced Information Science and Network Technology , Beijing , China"},{"name":"Deqiang Kong","slug":"deqiang-kong","given":"Deqiang","family":"Kong","affiliation":"Institute of Information Science , Beijing Jiaotong University , Beijing , China"}],"abstract":"Prototype selection is a promising technique for removing redundancy and irrelevance from large-scale data. Here, we consider it as a task assignment problem, which refers to assigning each element of a source set to one representative, i.e., prototype. However, due to the outliers and uncertain distribution on source, the selected prototypes are generally less representative and interesting. To alleviate this issue, we develop in this paper a Self-supervised Deep Low-rank Assignment model (SDLA). By dynamically integrating a low-rank assignment model with deep representation learning, our model effectively ensures the goodness-of-exemplar and goodness-of-discrimination of selected prototypes. Specifically, on the basis of a denoising autoencoder, dissimilarity metrics on source are continuously self-refined in embedding space with weak supervision from selected prototypes, thus preserving categorical similarity. Conversely, working on this metric space, similar samples tend to select the same prototypes by designing a low-rank assignment model. Experimental results on applications like text clustering and image classification (using prototypes) demonstrate our method is considerably superior to the state-of-the-art methods in prototype selection.","doi":"10.24963/ijcai.2018/436","url":"https://ijcai.org/proceedings/2018/436","pdf":"https://ijcai.org/proceedings/2018/0436.pdf","markdown":"https://ijcai.org/proceedings/2018/436.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/436","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":436,"pages":{"first":3141,"last":3147},"keywords":["Machine Learning: Active Learning","Machine Learning: Clustering","Machine Learning: Semi-Supervised Learning"],"videos":[],"references":[{"position":1,"raw":"[Bengio et al., 2013] Yoshua Bengio, Aaron Courville, and Pascal Vincent. Representation learning: A review and new perspectives. TPAMI, 35(8):1798-1828, 2013.","title":"Representation learning: A review and new perspectives","authors":"Yoshua Bengio, Aaron Courville, Pascal Vincent","year":"2013","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Cai et al., 2005] Deng Cai, Xiaofei He, and Jiawei Han. Document clustering using locality preserving indexing. TPAMI, 17(12):1624-1637, 2005.","title":"Document clustering using locality preserving indexing","authors":"Deng Cai, Xiaofei He, Jiawei Han","year":"2005","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Chan, 1987] Tony F Chan. Rank revealing QR factoriza- tions. Linear algebra and its applications, 88:67-82, 1987.","title":"Rank revealing QR factorizations","authors":"Tony F Chan","year":"1987","venue":"Linear algebra and its applications","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chen and Cai, 2011] Xinlei Chen and Deng Cai. Large s- cale spectral clustering with landmark-based representa- tion. In AAAI, pages 313-318, 2011.","title":"Large scale spectral clustering with landmark-based representation","authors":"Chen, Cai, Xinlei Chen, Deng Cai","year":"2011","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Cong et al., 2011] Yang Cong, Junsong Yuan, and Ji Liu. S- parse reconstruction cost for abnormal event detection. In CVPR, pages 3449-3456, 2011.","title":"Sparse reconstruction cost for abnormal event detection","authors":"Yang Cong, Junsong Yuan, Ji Liu","year":"2011","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Cong et al., 2017] Yang Cong, Ji Liu, Gan Sun, Quanzeng You, Yuncheng Li, and Jiebo Luo. Adaptive greedy dic- tionary selection for web media summarization. TIP, 26(1):185-195, 2017.","title":"Adaptive greedy dictionary selection for web media summarization","authors":"Yang Cong, Ji Liu, Gan Sun, Quanzeng You, Yuncheng Li, Jiebo Luo","year":"2017","venue":"TIP","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Dornaika and Aldine, 2015] Fadi Dornaika and I Kamal Al- dine. Decremental sparse modeling representative s- election for prototype selection. Pattern Recognit., 48(11):3714-3727, 2015.","title":"Decremental sparse modeling representative selection for prototype selection","authors":"Fadi Dornaika, I Kamal, Aldine","year":"2015","venue":"Pattern Recognit","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Duda et al., 2012] Richard O Duda, Peter E Hart, and David G Stork. Pattern classification. John Wiley & Sons, 2012.","title":"Pattern classification","authors":"Peter E Richard O Duda, David G Hart, Stork","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Elhamifar and Kaluza, 2017] Ehsan Elhamifar and MCDP Kaluza. Online summarization via submodular and con- vex optimization. In CVPR, pages 1783-1791, 2017.","title":"Ehsan Elhamifar and MCDP Kaluza. Online summarization via submodular and convex optimization","authors":"Kaluza Elhamifar","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Elhamifar et al., 2012] Ehsan Elhamifar, Guillermo Sapiro, and Rene Vidal. See all by looking at a few: Sparse mod- eling for finding representative objects. In CVPR, pages 1600-1607, 2012.","title":"See all by looking at a few: Sparse modeling for finding representative objects","authors":"Ehsan Elhamifar, Guillermo Sapiro, Rene Vidal","year":"2012","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Elhamifar et al., 2016] Ehsan Elhamifar, Guillermo Sapiro, and S Shankar Sastry. Dissimilarity-based sparse subset selection. TPAMI, 38(11):2182-2197, 2016.","title":"Dissimilarity-based sparse subset selection","authors":"Ehsan Elhamifar, Guillermo Sapiro, S Shankar Sastry","year":"2016","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Fan et al., 2017] Mingyu Fan, Xiaojun Chang, and Dacheng Tao. Structure regularized unsupervised discriminant fea- ture analysis. In AAAI, pages 1870-1876, 2017.","title":"Structure regularized unsupervised discriminant feature analysis","authors":"Mingyu Fan, Xiaojun Chang, Dacheng Tao","year":"2017","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Frey and Dueck, 2007] Brendan J Frey and Delbert Dueck. Clustering by passing messages between data points. Sci- ence, 315(5814):972-976, 2007.","title":"Clustering by passing messages between data points","authors":"J Brendan, Delbert Frey, Dueck","year":"2007","venue":"Science","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Garcia et al., 2012] Salvador Garcia, Joaquin Derrac, Jose Cano, and Francisco Herrera. Prototype selection for n- earest neighbor classification: Taxonomy and empirical s- tudy. TPAMI, 34(3):417-435, 2012.","title":"Salvador Garcia, Joaquin Derrac, Jose Cano, and Francisco Herrera. Prototype selection for nearest neighbor classification: Taxonomy and empirical study","authors":"Garcia","year":"2012","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Hull, 1994] Jonathan J. Hull. A database for handwritten text recognition research. TPAMI, 16(5):550-554, 1994.","title":"A database for handwritten text recognition research","authors":"Jonathan J Hull","year":"1994","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kaufman and Rousseeuw, 1987] Leonard Kaufman and Pe- ter Rousseeuw. Clustering by means of medoids. In Y. Dodge (Ed.) Statistical Data Anal. Based on the ℓ 1 -norm and Related Methods, pages 405-416, 1987.","title":"Clustering by means of medoids","authors":"Rousseeuw ; Kaufman, Peter Kaufman, Rousseeuw","year":"1987","venue":"Statistical Data Anal. Based on the ℓ 1 -norm and Related Methods","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Krause et al., 2008] Andreas Krause, H Brendan McMa- han, Carlos Guestrin, and Anupam Gupta. Robust sub- modular observation selection. J. Mach. Learn. Research, 9:2761-2801, 2008.","title":"Robust submodular observation selection","authors":"Krause","year":"2008","venue":"J. Mach. Learn. Research","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Lazebnik et al., 2006] S. Lazebnik, C. Schmid, and J. Ponce. Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In CVPR, pages 2169-2178, 2006.","title":"Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories","authors":"S Lazebnik, C Schmid, J Ponce","year":"2006","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Lin et al., 2018] Liang Lin, Keze Wang, Deyu Meng, Wangmeng Zuo, and Lei Zhang. Active self-paced learn- ing for cost-effective and progressive face identification. TPAMI, 40(1):7-19, 2018.","title":"Active self-paced learning for cost-effective and progressive face identification","authors":"Liang Lin, Keze Wang, Deyu Meng, Wangmeng Zuo, Lei Zhang","year":"2018","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Meng et al., 2016] Jingjing Meng, Hongxing Wang, Jun- song Yuan, and Yap-Peng Tan. From keyframes to key ob- jects: Video summarization by representative object pro- posal selection. In CVPR, pages 1039-1048, 2016.","title":"From keyframes to key objects: Video summarization by representative object proposal selection","authors":"Jingjing Meng, Hongxing Wang, Junsong Yuan, Yap-Peng Tan","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Nellore and Ward, 2015] Abhinav Nellore and Rachel Ward. Recovery guarantees for exemplar-based cluster- ing. Inform. and Computat., 245:165-180, 2015.","title":"Recovery guarantees for exemplar-based clustering","authors":"Abhinav Nellore, Rachel Ward","year":"2015","venue":"Inform. and Computat","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ng et al., 2002] Andrew Y Ng, Michael I Jordan, and Yair Weiss. On spectral clustering: Analysis and an algorithm. In NIPS, pages 849-856, 2002.","title":"On spectral clustering: Analysis and an algorithm","authors":"Y Andrew, Michael I Ng, Yair Jordan, Weiss","year":"2002","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Okura et al., 2017] Shumpei Okura, Yukihiro Tagami, Shin- go Ono, and Akira Tajima. Embedding-based news rec- ommendation for millions of users. In KDD, pages 1933- 1942, 2017.","title":"Embedding-based news recommendation for millions of users","authors":"Okura","year":"2017","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Simonyan and Zisserman, 2014] Karen Simonyan and An- drew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556, 2014.","title":"Very deep convolutional networks for large-scale image recognition","authors":"Karen Simonyan, Andrew Zisserman","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Song et al., 2017] Hyun Oh Song, Stefanie Jegelka, Vivek Rathod, and Kevin Murphy. Deep metric learning via fa- cility location. In CVPR, pages 5382-5390, 2017.","title":"Deep metric learning via facility location","authors":"Hyun Oh Song, Stefanie Jegelka, Vivek Rathod, Kevin Murphy","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Vincent et al., 2008] Pascal Vincent, Hugo Larochelle, Yoshua Bengio, and Pierre-Antoine Manzagol. Extract- ing and composing robust features with denoising autoen- coders. In ICML, pages 1096-1103, 2008.","title":"Extracting and composing robust features with denoising autoencoders","authors":"Pascal Vincent, Hugo Larochelle, Yoshua Bengio, Pierre-Antoine Manzagol","year":"2008","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Wang and Zhang, 2013] Shusen Wang and Zhihua Zhang. Improving CUR matrix decomposition and the nyström approximation via adaptive sampling. J. Mach. Learn. Res., 14(9):2729-2769, 2013.","title":"Improving CUR matrix decomposition and the nyström approximation via adaptive sampling","authors":"Wang, Zhang ; ] Shusen Wang, Zhihua Zhang","year":"2013","venue":"J. Mach. Learn. Res","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Wang et al., 2017] Hongxing Wang, Yoshinobu Kawahara, Chaoqun Weng, and Junsong Yuan. Representative selec- tion with structured sparsity. Pattern Recognit., 63:268- 278, 2017.","title":"Representative selection with structured sparsity","authors":"Hongxing Wang, Yoshinobu Kawahara, Chaoqun Weng, Junsong Yuan","year":"2017","venue":"Pattern Recognit","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Yin et al., 2016] Ming Yin, Junbin Gao, and Zhouchen Lin. Laplacian regularized low-rank representation and its ap- plications. TPAMI, 38(3):504-517, 2016.","title":"Laplacian regularized low-rank representation and its applications","authors":"Ming Yin, Junbin Gao, Zhouchen Lin","year":"2016","venue":"TPAMI","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhang et al., 2018] Xingxing Zhang, Zhenfeng Zhu, Yao Zhao, and Dongxia Chang. Learning a gener- al assignment model for video analytics. TCSVT, DOI:10.1109/TCSVT.2017.2713480, in press, 2018.","title":"Learning a general assignment model for video analytics","authors":"Xingxing Zhang, Zhenfeng Zhu, Yao Zhao, Dongxia Chang","year":"2018","venue":"TCSVT","doi":"10.1109/tcsvt.2017.2713480","url":null,"internal_url":null},{"position":31,"raw":"[Zhuang et al., 2012] Liansheng Zhuang, Haoyuan Gao, Zhouchen Lin, Yi Ma, Xin Zhang, and Nenghai Yu. Non- negative low rank and sparse graph for semi-supervised learning. In CVPR, pages 2328-2335, 2012.","title":"Nonnegative low rank and sparse graph for semi-supervised learning","authors":"Liansheng Zhuang, Haoyuan Gao, Zhouchen Lin, Yi Ma, Xin Zhang, Nenghai Yu","year":"2012","venue":"CVPR","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}