{"type":"ScholarlyArticle","title":"Extracting Privileged Information from Untagged Corpora for Classifier Learning","authors":[{"name":"Yazhou Yao","slug":"yazhou-yao","given":"Yazhou","family":"Yao","affiliation":"† University of Technology Sydney , NSW , Australia"},{"name":"Jian Zhang","slug":"jian-zhang","given":"Jian","family":"Zhang","affiliation":"† University of Technology Sydney , NSW , Australia"},{"name":"Fumin Shen","slug":"fumin-shen","given":"Fumin","family":"Shen","affiliation":"§ University of Electronic Science and Technology of China , Chengdu , China"},{"name":"Wankou Yang","slug":"wankou-yang","given":"Wankou","family":"Yang","affiliation":"Southeast University , Nanjing , China; Nanjing University of Science and Technology , Nanjing , China"},{"name":"Xian-Sheng Hua","slug":"xian-sheng-hua","given":"Xian-Sheng","family":"Hua","affiliation":"Southeast University , Nanjing , China"},{"name":"Zhenmin Tang","slug":"zhenmin-tang","given":"Zhenmin","family":"Tang","affiliation":"‡ DAMO Academy , Alibaba Group , Hangzhou , China"}],"abstract":"The performance of data-driven learning approaches is often unsatisfactory when the training data is inadequate either in quantity or quality. Manually labeled privileged information (PI), \\eg attributes, tags or properties, is usually incorporated to improve classifier learning. However, the process of manually labeling is time-consuming and labor-intensive. To address this issue, we propose to enhance classifier learning by extracting PI from untagged corpora, which can effectively eliminate the dependency on manually labeled data. In detail, we treat each selected PI as a subcategory and learn one classifier for per subcategory independently. The classifiers for all subcategories are then integrated together to form a more powerful category classifier. Particularly, we propose a new instance-level multi-instance learning (MIL) model to simultaneously select a subset of training images from each subcategory and learn the optimal classifiers based on the selected images. Extensive experiments demonstrate the superiority of our approach.","doi":"10.24963/ijcai.2018/151","url":"https://ijcai.org/proceedings/2018/151","pdf":"https://ijcai.org/proceedings/2018/0151.pdf","markdown":"https://ijcai.org/proceedings/2018/151.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/151","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":"Computer Vision","track":"Main track","number":151,"pages":{"first":1085,"last":1091},"keywords":["Humans and AI: Cognitive Systems","Computer Vision: Language and Vision"],"videos":[],"references":[{"position":1,"raw":"[Cilibrasi and Vitanyi, 2007] R Cilibrasi and P Vitanyi. The google similarity distance. IEEE Transactions on Knowl- edge and Data Engineering, 19(3): 370-383, 2007.","title":"The google similarity distance","authors":"R Cilibrasi, Vitanyi","year":"2007","venue":"IEEE Transactions on Knowledge and Data Engineering","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Liu et al., 2015] X Liu, L Wang, GB Huang, J Zhang, J Yin Multiple kernel extreme learning machine. Neurocomput- ing, 149: 253-264, 2015.","title":"Multiple kernel extreme learning machine","authors":"L Liu, G B Wang, J Huang, J Zhang, Yin","year":"2015","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Everingham et al., 2010] M Everingham, L Gool, C Williams, and A Zisserman. The pascal visual ob- ject classes (voc) challenge. International Journal of Computer Vision, 88(2): 303-338, 2010.","title":"The pascal visual object classes (voc) challenge","authors":"Everingham, C Gool, Williams, Zisserman","year":"2010","venue":"International Journal of Computer Vision","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Hoai and Zisserman, 2013] M Hoai and A Zisserman. Dis- criminative sub-categorization. IEEE International Con- ference on Computer Vision and Pattern Recognition, 1666-1673, 2013.","title":"Discriminative sub-categorization","authors":"M Hoai, Zisserman","year":"2013","venue":"IEEE International Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Kelley, 1960] James E Kelley, Jr. The cutting-plane method for solving convex programs. Journal of the Society for Industrial and Mathematics, 8(4):703-712, 1960.","title":"The cutting-plane method for solving convex programs","authors":"Jr James E Kelley","year":"1960","venue":"Journal of the Society for Industrial and Mathematics","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Li et al., 2011] W Li, L Duan, D Xu, and I Tsang. Text- based image retrieval using progressive multi-instance learning. IEEE International Conference on Computer Vi- sion, 2049-2055, 2011.","title":"Textbased image retrieval using progressive multi-instance learning","authors":"L Li, D Duan, Xu, Tsang","year":"2011","venue":"IEEE International Conference on Computer Vision","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Shen et al., 2017] F Shen, Y Mu, Y Yang, W Liu, H Shen. Classification by Retrieval: Binarizing Data and Classi- fiers. ACM SIGIR Conference on Research and Develop- ment in Information Retrieval, 595-604, 2017.","title":"Classification by Retrieval: Binarizing Data and Classifiers","authors":"Y Shen, Y Mu, W Yang, H Liu, Shen","year":"2017","venue":"ACM SIGIR Conference on Research and Development in Information Retrieval","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Sindhwani et al., 2005] V Sindhwani, P Niyogi, and M Belkin, \"A co-regularization approach to semi-supervised learning with multiple views,\" International Conference on Machine Learning, 74-79, 2005.","title":"A co-regularization approach to semi-supervised learning with multiple views","authors":"Sindhwani, M Niyogi, Belkin","year":"2005","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Lin et al., 2012] Y Lin, E Aiden, J Orwant, and S Petrov. Syntactic annotations for the google books ngram corpus. ACL System Demonstrations, 169-174, 2012.","title":"Syntactic annotations for the google books ngram corpus","authors":"E Lin, J Aiden, Orwant, Petrov","year":"2012","venue":"ACL System Demonstrations","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Ristin et al., 2015] M Ristin, J Gall, M Guillaumin, and L Gool. From categories to subcategories: large-scale im- age classification with partial class label refinement. IEEE Conference on Computer Vision and Pattern Recognition, 231-239, 2015.","title":"From categories to subcategories: large-scale image classification with partial class label refinement","authors":"Ristin, M Gall, L Guillaumin, Gool","year":"2015","venue":"IEEE Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Liu et al., 2013] X Liu, L Wang, J Yin, E Zhu, J Zhang. An efficient approach to integrating radius information into multiple kernel learning. IEEE transactions on cybernet- ics, 43(2): 557-569, 2013.","title":"An efficient approach to integrating radius information into multiple kernel learning","authors":"Liu","year":"2013","venue":"IEEE transactions on cybernetics","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Krizhevsky and Hinton, 2009] A Krizhevsky and G Hinton. Learning multiple layers of features from tiny images. 2009.","title":"Learning multiple layers of features from tiny images","authors":"A Krizhevsky, G Hinton","year":"2009","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Wang et al., 2013] X Wang, and Z Tu. Max-margin multiple-instance dictionary learning. International Con- ference on Machine Learning, 846-854, 2013.","title":"Max-margin multiple-instance dictionary learning","authors":"Wang","year":"2013","venue":"International Conference on Machine Learning","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Weston and Watkins, 1998] J Weston and C Watkins. Multi- class support vector machines. Technical Report, 1998.","title":"Multiclass support vector machines","authors":"Weston, Watkins ; J Weston, C Watkins","year":"1998","venue":null,"doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Yao et al., 2017] Y Yao, J Zhang, F Shen, X Hua, J Xu, Z Tang. Exploiting Web Images for Dataset Construction: A Domain Robust Approach. IEEE Transactions on Multi- media, 19(8): 1771-1784, 2017.","title":"Exploiting Web Images for Dataset Construction: A Domain Robust Approach","authors":"J Yao, F Zhang, X Shen, J Hua, Z Xu, Tang","year":"2017","venue":"IEEE Transactions on Multimedia","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Yao et al., 2016] Y Yao, X Hua, F Shen, and Z Tang. A do- main robust approach for image dataset construction. ACM Conference on Multimedia, 212-216, 2016.","title":"A domain robust approach for image dataset construction","authors":"X Yao, F Hua, Shen, Tang","year":"2016","venue":"ACM Conference on Multimedia","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Krizhevsky et al., 2012] A Krizhevsky, I Sutskever, and G Hinton. Imagenet classification with deep convolutional neural networks. Advances in Neural Information Process- ing Systems, 1097-1105, 2012.","title":"Imagenet classification with deep convolutional neural networks","authors":"Krizhevsky, G Sutskever, Hinton","year":"2012","venue":"Advances in Neural Information Processing Systems","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Andrews et al., 2003] S Andrews, I Tsochantaridis, and T Hofmann. Support vector machines for multiple-instance learning. Advances in neural information processing sys- tems, 577-584, 2003.","title":"Support vector machines for multiple-instance learning","authors":"I Andrews, Tsochantaridis, Hofmann","year":"2003","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Yao et al., 2016] Y Yao, J Zhang, F Shen, X Hua, J Xu, Z Tang. Automatic Image Dataset Construction with Mul- tiple Textual Metadata. IEEE Conference on Multimedia and Expo, 1-6, 2016.","title":"Automatic Image Dataset Construction with Multiple Textual Metadata","authors":"J Yao, F Zhang, X Shen, J Hua, Z Xu, Tang","year":"2016","venue":"IEEE Conference on Multimedia and Expo","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Bunescu and Mooney, 2007] R Bunescu and R Mooney. Multiple instance learning for sparse positive bags. ACM International conference on Machine learning, 105-112, 2007.","title":"Multiple instance learning for sparse positive bags","authors":"R Bunescu, Mooney","year":"2007","venue":"ACM International conference on Machine learning","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Coates et al., 2011] A Coates, A Ng, and H Lee. An analy- sis of single-layer networks in unsupervised feature learn- ing. International conference on artificial intelligence, 215-223, 2011.","title":"An analysis of single-layer networks in unsupervised feature learning","authors":"Coates, H Ng, Lee","year":"2011","venue":"International conference on artificial intelligence","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Divvala et al., 2014] S Divvala, A Farhadi, and C Guestrin. Learning everything about anything: Webly-supervised vi- sual concept learning. IEEE Conference on Computer Vi- sion and Pattern Recognition, 3270-3277, 2014.","title":"Learning everything about anything: Webly-supervised visual concept learning","authors":"Divvala, C Farhadi, Guestrin","year":"2014","venue":"IEEE Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Li et al., 2014] W Li, L Niu, and D Xu. Exploiting priv- ileged information from web data for image categoriza- tion. European Conference on Computer Vision, 437-452, 2014.","title":"Exploiting privileged information from web data for image categorization","authors":"L Li, Niu, Xu","year":"2014","venue":"European Conference on Computer Vision","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Niu et al., 2017] L Niu, W Li, D Xu. Visual recognition by learning from web data via weakly supervised domain generalization. IEEE transactions on neural networks and learning systems, 28(9):1985-1999, 2017.","title":"Visual recognition by learning from web data via weakly supervised domain generalization","authors":"W Niu, D Li, Xu","year":"2017","venue":"IEEE transactions on neural networks and learning systems","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Platt et al., 1999] J. Platt, \"Fast training of support vector machines using sequential minimal optimization,\" Ad- vances in Kernel Methods, 185-208, 1999.","title":"Fast training of support vector machines using sequential minimal optimization","authors":"J Platt","year":"1999","venue":"Advances in Kernel Methods","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Wang and Ji, 2015] Z Wang and Q Ji. Classifier learning with hidden information. IEEE Conference on Computer Vision and Pattern Recognition, 4969-4977, 2015.","title":"Classifier learning with hidden information","authors":"Wang, Ji ; ] Z Wang, Q Ji","year":"2015","venue":"IEEE Conference on Computer Vision and Pattern Recognition","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Brefeld et al., 2006] U Brefeld, T Scheffer, and S Wrobel. Efficient co-regularised least squares regression. ACM conference on Machine learning, 137-144, 2006.","title":"Efficient co-regularised least squares regression","authors":"Brefeld, Scheffer, Wrobel","year":"2006","venue":"ACM conference on Machine learning","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Shen et al., 2017] F Shen, X Gao, L Liu, Y Yang, H Shen. Deep Asymmetric Pairwise Hashing. ACM conference on Multimedia, 1522-1530, 2017.","title":"Deep Asymmetric Pairwise Hashing","authors":"X Shen, L Gao, Y Liu, H Yang, Shen","year":"2017","venue":"ACM conference on Multimedia","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Yao et al., 2018] Y Yao, J Zhang, F Shen, W Yang, P Huang, Z Tang. Discovering and Distinguishing Multiple Visual Senses for Polysemous Words. AAAI Conference on Artificial Intelligence, 2018.","title":"Discovering and Distinguishing Multiple Visual Senses for Polysemous Words","authors":"J Yao, F Zhang, W Shen, P Yang, Z Huang, Tang","year":"2018","venue":"AAAI Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}