{"type":"ScholarlyArticle","title":"Exploring Binary Classification Hidden within Partial Label Learning","authors":[{"name":"Hengheng Luo","slug":"hengheng-luo","given":"Hengheng","family":"Luo","affiliation":"Key Lab of Data Engineering and Knowledge Engineering of MOE Renmin University of China"},{"name":"Yabin Zhang","slug":"yabin-zhang","given":"Yabin","family":"Zhang","affiliation":"Gaoling School of Artificial Intelligence , Renmin University of China , Beijing 100872 , China"},{"name":"Suyun Zhao","slug":"suyun-zhao","given":"Suyun","family":"Zhao","affiliation":"Key Lab of Data Engineering and Knowledge Engineering of MOE Renmin University of China"},{"name":"Hong Chen","slug":"hong-chen","given":"Hong","family":"Chen","affiliation":"Key Lab of Data Engineering and Knowledge Engineering of MOE Renmin University of China"},{"name":"Cuiping Li","slug":"cuiping-li","given":"Cuiping","family":"Li","affiliation":"Key Lab of Data Engineering and Knowledge Engineering of MOE Renmin University of China"}],"abstract":"Partial label learning (PLL) is to learn a discriminative model under incomplete supervision, where each instance is annotated with a candidate label set. The basic principle of PLL is that the unknown correct label y of an instance x resides in its candidate label set s, i.e., P(y ∈ s | x) = 1. On which basis, current researches either directly model P(x | y) under different data generation assumptions or propose various surrogate multiclass losses, which all aim to encourage the model-based Pθ(y ∈ s | x)→1 implicitly. In this work, instead, we explicitly construct a binary classification task toward P(y ∈ s | x) based on the discriminative model, that is to predict whether the model-output label of x is one of its candidate labels. We formulate a novel risk estimator with estimation error bound for the proposed PLL binary classification risk. By applying logit adjustment based on disambiguation strategy, the practical approach directly maximizes Pθ(y ∈ s | x) while implicitly disambiguating the correct one from candidate labels simultaneously. Thorough experiments validate that the proposed approach achieves competitive performance against the state-of-the-art PLL methods.","doi":"10.24963/ijcai.2022/456","url":"https://ijcai.org/proceedings/2022/456","pdf":"https://ijcai.org/proceedings/2022/0456.pdf","markdown":"https://ijcai.org/proceedings/2022/456.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/456","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Machine Learning","track":"Main Track","number":456,"pages":{"first":3285,"last":3291},"keywords":["Machine Learning: Weakly Supervised Learning","Machine Learning: Classification"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38987039","length_sec":79,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38985182","length_sec":577,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Briggs et al., 2012] Forrest Briggs, Xiaoli Z Fern, and Ra- viv Raich. Rank-loss support instance machines for miml instance annotation. In SIGKDD, pages 534-542, 2012.","title":"Rank-loss support instance machines for miml instance annotation","authors":"Forrest Briggs, Raviv Xiaoli Z Fern, Raich","year":"2012","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chou et al., 2020] Yu-Ting Chou, Gang Niu, Hsuan-Tien Lin, and Masashi Sugiyama. Unbiased risk estimators can mislead: A case study of learning with complementary la- bels. In ICML, pages 1929-1938. PMLR, 2020.","title":"Unbiased risk estimators can mislead: A case study of learning with complementary labels","authors":"Yu-Ting Chou, Gang Niu, Hsuan-Tien Lin, Masashi Sugiyama","year":"2020","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Clanuwat et al., 2018] Tarin Clanuwat, Mikel Bober-Irizar, Asanobu Kitamoto, Alex Lamb, Kazuaki Yamamoto, and David Ha. Deep learning for classical japanese literature. arXiv preprint arXiv:1812.01718, 2018.","title":"Deep learning for classical japanese literature","authors":"Tarin Clanuwat, Mikel Bober-Irizar, Asanobu Kitamoto, Alex Lamb, Kazuaki Yamamoto, David Ha","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Cour et al., 2009] Timothee Cour, Benjamin Sapp, Chris Jordan, and Ben Taskar. Learning from ambiguously la- beled images. In CVPR, pages 919-926. IEEE, 2009.","title":"Learning from ambiguously labeled images","authors":"Timothee Cour, Benjamin Sapp, Chris Jordan, Ben Taskar","year":"2009","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Cour et al., 2011] T. Cour, B. Sapp, and B. Taskar. Learn- ing from partial labels. Journal of Machine Learning Re- search, 12(4):1501-1536, 2011.","title":"Learning from partial labels","authors":"T Cour, B Sapp, B Taskar","year":"2011","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Feng and An, 2018] Lei Feng and Bo An. Leveraging la- tent label distributions for partial label learning. In IJCAI, pages 2107-2113, 2018.","title":"Leveraging latent label distributions for partial label learning","authors":"Lei Feng, Bo An","year":"2018","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2018/291"},{"position":7,"raw":"[Feng and An, 2019] Lei Feng and Bo An. Partial label learning by semantic difference maximization. In IJCAI, pages 2294-2300, 2019.","title":"Lei Feng and Bo An. Partial label learning by semantic difference maximization","authors":"An Feng","year":"2019","venue":"IJCAI","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Feng et al., 2020] Lei Feng, Jiaqi Lv, Bo Han, Miao Xu, Gang Niu, Xin Geng, Bo An, and Masashi Sugiyama. Provably consistent partial-label learning. In NeurIPS, 2020.","title":"Provably consistent partial-label learning","authors":"Lei Feng, Jiaqi Lv, Bo Han, Miao Xu, Gang Niu, Xin Geng, Bo An, Masashi Sugiyama","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Grathwohl et al., 2020] Will Grathwohl, Kuan-Chieh Wang, Jörn-Henrik Jacobsen, David Duvenaud, Moham- mad Norouzi, and Kevin Swersky. Your classifier is secretly an energy based model and you should treat it like one. In ICLR, 2020.","title":"Your classifier is secretly an energy based model and you should treat it like one","authors":"Grathwohl","year":"2020","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Guillaumin et al., 2010] Matthieu Guillaumin, Jakob Ver- beek, and Cordelia Schmid. Multiple instance metric learning from automatically labeled bags of faces. In ECCV, pages 634-647. Springer, 2010.","title":"Multiple instance metric learning from automatically labeled bags of faces","authors":"Matthieu Guillaumin, Jakob Verbeek, Cordelia Schmid","year":"2010","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Hüllermeier and Beringer, 2006] Eyke Hüllermeier and Jürgen Beringer. Learning from ambiguously labeled examples. Intelligent Data Analysis, 10(5):419-439, 2006.","title":"Eyke Hüllermeier and Jürgen Beringer. Learning from ambiguously labeled examples","authors":"Beringer Hüllermeier","year":"2006","venue":"Intelligent Data Analysis","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Jin and Ghahramani, 2002] Rong Jin and Zoubin Ghahra- mani. Learning with multiple labels. In NeurIPS, vol- ume 2, pages 897-904. Citeseer, 2002.","title":"Rong Jin and Zoubin Ghahramani","authors":"Jin, Ghahramani","year":"2002","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[LeCun et al., 1998] Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning ap- plied to document recognition. Proceedings of the IEEE, 86(11):2278-2324, 1998.","title":"Gradient-based learning applied to document recognition","authors":"Yann Lecun, Léon Bottou, Yoshua Bengio, Patrick Haffner","year":"1998","venue":"Proceedings of the IEEE","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Liu and Dietterich, 2012] Liping Liu and Thomas G Diet- terich. A conditional multinomial mixture model for su- perset label learning. In Advances in neural information processing systems, pages 548-556. Citeseer, 2012.","title":"A conditional multinomial mixture model for superset label learning","authors":"Dietterich Liu, Liping Liu, Thomas G Dietterich","year":"2012","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu and Dietterich, 2014] Li-Ping Liu and Thomas G. Diet- terich. Learnability of the superset label learning problem. In ICML, 2014.","title":"Dietterich. Learnability of the superset label learning problem","authors":"Dietterich ; ] Li-Ping Liu, Thomas G Liu","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Lv et al., 2020] Jiaqi Lv, Miao Xu, Lei Feng, Gang Niu, Xin Geng, and Masashi Sugiyama. Progressive identification of true labels for partial-label learning. In ICML, pages 6500-6510. PMLR, 2020.","title":"Progressive identification of true labels for partial-label learning","authors":"Jiaqi Lv, Miao Xu, Lei Feng, Gang Niu, Xin Geng, Masashi Sugiyama","year":"2020","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Mohri et al., 2012] Mehryar Mohri, Afshin Rostamizadeh, and Ameet S. Talwalkar. Foundations of machine learning. In Adaptive computation and machine learning, 2012.","title":"Foundations of machine learning","authors":"Mehryar Mohri, Afshin Rostamizadeh, Ameet S Talwalkar","year":"2012","venue":"Adaptive computation and machine learning","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Nguyen and Caruana, 2008] Nam Nguyen and Rich Caru- ana. Classification with partial labels. In SIGKDD, pages 551-559, 2008.","title":"Classification with partial labels","authors":"Caruana Nguyen","year":"2008","venue":"SIGKDD","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Paszke et al., 2019] Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high- performance deep learning library. Advances in neural in- formation processing systems, 32:8026-8037, 2019.","title":"Pytorch: An imperative style, highperformance deep learning library","authors":"Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga","year":"2019","venue":"Advances in neural information processing systems","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Qian et al., 2019] Qi Qian, Lei Shang, Baigui Sun, Juhua Hu, H. Li, and Rong Jin. Softtriple loss: Deep metric learning without triplet sampling. ICCV, pages 6449- 6457, 2019.","title":"Softtriple loss: Deep metric learning without triplet sampling","authors":"Lei Qi Qian, Baigui Shang, Juhua Sun, H Hu, Rong Li, Jin","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Seo and Huh, 2021] Junghoon Seo and Joon Suk Huh. On the power of deep but naive partial label learning. In ICASSP, pages 3820-3824. IEEE, 2021.","title":"On the power of deep but naive partial label learning","authors":"Huh Seo, Junghoon Seo, Joon Suk, Huh","year":"2021","venue":"ICASSP","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Wen et al., 2021] Hongwei Wen, Jingyi Cui, Hanyuan Hang, Jiabin Liu, Yisen Wang, and Zhouchen Lin. Lever- aged weighted loss for partial label learning. In ICML, volume 139, pages 11091-11100. PMLR, 2021.","title":"Leveraged weighted loss for partial label learning","authors":"Hongwei Wen, Jingyi Cui, Hanyuan Hang, Jiabin Liu, Yisen Wang, Zhouchen Lin","year":"2021","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Wu and Zhang, 2018] X. Wu and Min-Ling Zhang. To- wards enabling binary decomposition for partial label learning. In IJCAI, 2018.","title":"Towards enabling binary decomposition for partial label learning","authors":"Zhang Wu, X Wu, Min-Ling Zhang","year":"2018","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2018/398"},{"position":24,"raw":"[Xiao et al., 2017] Han Xiao, Kashif Rasul, and Roland Vollgraf. Fashion-mnist: a novel image dataset for bench- marking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017.","title":"Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms","authors":"Han Xiao, Kashif Rasul, Roland Vollgraf","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yan and Guo, 2020] Yan Yan and Yuhong Guo. Partial label learning with batch label correction. In AAAI, volume 34, pages 6575-6582, 2020.","title":"Partial label learning with batch label correction","authors":"Yan Yan, Yuhong Guo","year":"2020","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zeng et al., 2013] Zinan Zeng, Shijie Xiao, Kui Jia, Tsung- Han Chan, Shenghua Gao, Dong Xu, and Yi Ma. Learn- ing by associating ambiguously labeled images. In CVPR, pages 708-715, 2013.","title":"Learning by associating ambiguously labeled images","authors":"Zinan Zeng, Shijie Xiao, Kui Jia, Tsung-Han Chan, Shenghua Gao, Dong Xu, Yi Ma","year":"2013","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang and Yu, 2015] Min-Ling Zhang and Fei Yu. Solving the partial label learning problem: An instance-based ap- proach. In IJCAI, 2015.","title":"Solving the partial label learning problem: An instance-based approach","authors":"Yu ; ] Min-Ling Zhang, Fei Zhang, Yu","year":"2015","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2015/568"},{"position":28,"raw":"[Zhang et al., 2017] Min-Ling Zhang, Fei Yu, and Cai- Zhi Tang. Disambiguation-free partial label learning. IEEE Transactions on Knowledge and Data Engineering, 29(10):2155-2167, 2017.","title":"Disambiguation-free partial label learning","authors":"Zhang","year":"2017","venue":"IEEE Transactions on Knowledge and Data Engineering","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhang et al., 2020] Yabin Zhang, Guang Yang, Suyun Zhao, Peng Ni, Hairong Lian, Hong Chen, and Cuiping Li. Partial label learning via generative adversarial nets. In ECAI 2020, pages 1674-1681. IOS Press, 2020.","title":"Partial label learning via generative adversarial nets","authors":"Yabin Zhang, Guang Yang, Suyun Zhao, Peng Ni, Hairong Lian, Hong Chen, Cuiping Li","year":"2020","venue":"ECAI 2020","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}