ABSTRACT
Metric learning aims to learn a good distance metric that can capture the relationships among instances, and its importance has long been recognized in many fields. In the traditional settings of metric learning, an implicit assumption is that the associated labels of the instances are deterministic. However, in many real-world applications, the associated labels come naturally with probabilities instead of deterministic values. Thus, the existing metric learning methods cannot work well in these applications. To tackle this challenge, in this paper, we study how to effectively learn the distance metric from datasets that contain probabilistic information, and then propose two novel metric learning mechanisms for two types of probabilistic labels, i.e., the instance-wise probabilistic label and the group-wise probabilistic label. Compared with the existing metric learning methods, our proposed mechanisms are capable of learning distance metrics directly from the probabilistic labels with high accuracy. We also theoretically analyze the two proposed mechanisms and provide theoretical bounds on the sample complexity for both of them. Additionally, extensive experiments based on real-world datasets are conducted to verify the desirable properties of the proposed mechanisms.
Supplemental Material
- Mahdieh Soleymani Baghshah and Saeed Bagheri Shouraki . 2009. Semi-Supervised Metric Learning Using Pairwise Constraints Proceedings of the International Joint Conference on Artificial Intelligence. 1217--1222. Google ScholarDigital Library
- Aharon Bar-Hillel, Tomer Hertz, Noam Shental, and Daphna Weinshall . 2005. Learning a mahalanobis metric from equivalence constraints. Journal of Machine Learning Research Vol. 6, Jun (2005), 937--965. Google ScholarDigital Library
- Qiong Cao, Zheng-Chu Guo, and Yiming Ying . 2016. Generalization bounds for metric and similarity learning. Machine Learning Vol. 102, 1 (2016), 115--132. Google ScholarDigital Library
- Olivier Chapelle, Vikas Sindhwani, and Sathiya S Keerthi . 2008. Optimization techniques for semi-supervised support vector machines. Journal of Machine Learning Research Vol. 9, Feb (2008), 203--233. Google ScholarDigital Library
- Jason V Davis, Brian Kulis, Prateek Jain, Suvrit Sra, and Inderjit S Dhillon . 2007. Information-theoretic metric learning. In Proceedings of the 24th international conference on Machine learning. ACM, 209--216. Google ScholarDigital Library
- Matthieu Guillaumin, Jakob Verbeek, and Cordelia Schmid . 2010. Multiple instance metric learning from automatically labeled bags of faces. In Proceedings of the European conference on Computer Vision. Springer, 634--647. Google ScholarDigital Library
- Mengdi Huai, Chenglin Miao, Qiuling Suo, Yaliang Li, Jing Gao, and Aidong Zhang . 2018. Uncorrelated Patient Similarity Learning. In Proceedings of the 2018 SIAM International Conference on Data Mining. SIAM, 270--278.Google ScholarCross Ref
- Yinjie Huang, Cong Li, Michael Georgiopoulos, and Georgios C Anagnostopoulos . 2013. Reduced-rank local distance metric learning. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, 224--239. Google ScholarDigital Library
- Arun Shankar Iyer, J Saketha Nath, and Sunita Sarawagi . 2016. Privacy-preserving class ratio estimation. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 925--934. Google ScholarDigital Library
- Rong Jin, Shijun Wang, and Zhi-Hua Zhou . 2009. Learning a distance metric from multi-instance multi-label data Proceedings of the Conference on Computer Vision and Pattern Recognition. IEEE, 896--902.Google Scholar
- Marc T Law, Yaoliang Yu, Raquel Urtasun, Richard S Zemel, and Eric P Xing . 2017. Efficient multiple instance metric learning using weakly supervised data Proceedings of the Conference on Computer Vision and Pattern Recognition.Google Scholar
- Dewei Li and Yingjie Tian . 2016. Multi-view metric learning for multi-instance image classification. arXiv preprint arXiv:1610.06671 (2016).Google Scholar
- Weiwei Liu and Ivor W Tsang . 2015. Large Margin Metric Learning for Multi-Label Prediction. Proceedings of the AAAI Conference on Artificial Intelligence, Vol. Vol. 15. 2800--2806. Google ScholarDigital Library
- Gang Niu, Bo Dai, Makoto Yamada, and Masashi Sugiyama . 2014. Information-theoretic semi-supervised metric learning via entropy regularization. Neural computation Vol. 26, 8 (2014), 1717--1762. Google ScholarDigital Library
- Giorgio Patrini, Richard Nock, Paul Rivera, and Tiberio Caetano . 2014. (Almost) no label no cry. In Advances in Neural Information Processing Systems. 190--198. Google ScholarDigital Library
- Peng Peng, Raymond Chi-Wing Wong, and Phillp S Yu . 2014. Learning on probabilistic labels. In Proceedings of the 2014 SIAM International Conference on Data Mining. SIAM, 307--315.Google ScholarCross Ref
- Filipe Rodrigues, Francisco Pereira, and Bernardete Ribeiro . 2014. Gaussian process classification and active learning with multiple annotators International Conference on Machine Learning. 433--441. Google ScholarDigital Library
- Stefan Rueping . 2010. SVM classifier estimation from group probabilities Proceedings of the 27th international conference on machine learning (ICML-10). 911--918. Google ScholarDigital Library
- Shai Shalev-Shwartz and Shai Ben-David . 2014. Understanding machine learning: From theory to algorithms. Cambridge university press. Google ScholarDigital Library
- Kihyuk Sohn . 2016. Improved deep metric learning with multi-class n-pair loss objective Advances in Neural Information Processing Systems. 1857--1865. Google ScholarDigital Library
- Jimeng Sun, Fei Wang, Jianying Hu, and Shahram Edabollahi . 2012. Supervised patient similarity measure of heterogeneous patient records. ACM SIGKDD Explorations Newsletter Vol. 14, 1 (2012), 16--24. Google ScholarDigital Library
- Tao Sun, Dan Sheldon, and Brendan O'Connor . 2017. A Probabilistic Approach for Learning with Label Proportions Applied to the US Presidential Election. In Proceedings of the 2017 IEEE International Conference on Data Mining (ICDM). IEEE, 445--454.Google ScholarCross Ref
- Qiuling Suo, Fenglong Ma, Ye Yuan, Mengdi Huai, Weida Zhong, Aidong Zhang, and Jing Gao . 2017. Personalized Disease Prediction Using a CNN-Based Similarity Learning Method Proceedings of the 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).Google Scholar
- Tian Tian and Jun Zhu . 2015. Max-margin majority voting for learning from crowds Advances in Neural Information Processing Systems. 1621--1629. Google ScholarDigital Library
- Dong Wang and Xiaoyang Tan . 2014. Robust Distance Metric Learning in the Presence of Label Noise Proceedings of the AAAI Conference on Artificial Intelligence. 1321--1327. Google ScholarDigital Library
- Kilian Q Weinberger, John Blitzer, and Lawrence K Saul . 2006. Distance metric learning for large margin nearest neighbor classification Advances in neural information processing systems. 1473--1480. Google ScholarDigital Library
- Kilian Q Weinberger and Lawrence K Saul . 2009. Distance metric learning for large margin nearest neighbor classification. Journal of Machine Learning Research Vol. 10, Feb (2009), 207--244. Google ScholarDigital Library
- Eric P Xing, Michael I Jordan, Stuart J Russell, and Andrew Y Ng . 2003. Distance metric learning with application to clustering with side-information Advances in neural information processing systems. 521--528. Google ScholarDigital Library
- Felix X Yu, Dong Liu, Sanjiv Kumar, Tony Jebara, and Shih-Fu Chang . 2013. SVM for learning with label proportions. In Internalization conference on machine learning (2013). Google ScholarDigital Library
- Pourya Zadeh, Reshad Hosseini, and Suvrit Sra . 2016. Geometric mean metric learning. In International Conference on Machine Learning. 2464--2471. Google ScholarDigital Library
- Mengting Zhan, Shilei Cao, Buyue Qian, Shiyu Chang, and Jishang Wei . 2016. Low-rank sparse feature selection for patient similarity learning Proceeding of the 2016 IEEE 16th International Conference on Data Mining (ICDM). IEEE.Google Scholar
- Dengyong Zhou, Qiang Liu, John Platt, and Christopher Meek . 2014. Aggregating ordinal labels from crowds by minimax conditional entropy International conference on machine learning. 262--270. Google ScholarDigital Library
Index Terms
- Metric Learning from Probabilistic Labels
Recommendations
Learning Distance Metrics from Probabilistic Information
Special Issue on KDD 2018, Regular Papers and Survey PaperThe goal of metric learning is to learn a good distance metric that can capture the relationships among instances, and its importance has long been recognized in many fields. An implicit assumption in the traditional settings of metric learning is that ...
Joint learning of labels and distance metric
Special issue on game theoryMachine learning algorithms frequently suffer from the in sufficiency of training data and the usage of inappropriate distance metric. In this paper, we propose a joint learning of labels and distance metric (JLLDM) approach, which is able to ...
Co-metric: a metric learning algorithm for data with multiple views
We address the problem of metric learning for multi-view data. Many metric learning algorithms have been proposed, most of them focus just on single view circumstances, and only a few deal with multi-view data. In this paper, motivated by the co-...
Comments