{"type":"ScholarlyArticle","title":"WPML3CP: Wasserstein Partial Multi-Label Learning with Dual Label Correlation Perspectives","authors":[{"name":"Ximing Li","slug":"ximing-li","given":"Ximing","family":"Li","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Yuanchao Dai","slug":"yuanchao-dai","given":"Yuanchao","family":"Dai","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Bing Wang","slug":"bing-wang","given":"Bing","family":"Wang","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Changchun Li","slug":"changchun-li","given":"Changchun","family":"Li","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Renchu Guan","slug":"renchu-guan","given":"Renchu","family":"Guan","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Fangming Gu","slug":"fangming-gu","given":"Fangming","family":"Gu","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"},{"name":"Jihong Ouyang","slug":"jihong-ouyang","given":"Jihong","family":"Ouyang","affiliation":"College of Computer Science and Technology , Jilin University , China; Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education , Jilin University , China"}],"abstract":"Partial multi-label learning (PMLL) refers to a weakly-supervised classification problem, where each instance is associated with a set of candidate labels, covering its ground-truth labels but also with irrelevant ones. The current methodology of PMLL is to estimate the ground-truth confidences of candidate labels, i.e., the likelihood of a candidate label being a ground-truth one, and induce the multi-label predictor with them, rather than the candidate labels. In this paper, we aim to estimate precise ground-truth confidences by leveraging precise label correlations, which are also required to estimate. To this end, we propose to capture label correlations from both measuring and modeling perspectives. Specifically, we measure the loss between ground-truth confidences and predictions by employing the Wasserstein distance involving label correlations; and form a label correlation-aware regularization to constraint predictive parameters. The two techniques are coupled to promote precise estimations of label correlations. Upon these ideas, we propose a novel PMLL method, namely Wasserstein Partial Multi-Label Learning with dual Label Correlation Perspectives (WPML3CP). We conduct extensive experiments on several benchmark datasets. Empirical results demonstrate that WPML3CP can outperform the existing PMLL baselines.","doi":"10.24963/ijcai.2024/495","url":"https://ijcai.org/proceedings/2024/495","pdf":"https://ijcai.org/proceedings/2024/0495.pdf","markdown":"https://ijcai.org/proceedings/2024/495.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/495","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Machine Learning","track":"Main Track","number":495,"pages":{"first":4479,"last":4487},"keywords":["Machine Learning: ML: Weakly supervised learning","Machine Learning: ML: Classification","Machine Learning: ML: Self-supervised Learning"],"videos":[],"references":[{"position":1,"raw":"[Arjovsky et al., 2017] Martín Arjovsky, Soumith Chintala, and Léon Bottou. Wasserstein generative adversarial net- works. In ICML, pages 214-223, 2017.","title":"Martín Arjovsky, Soumith Chintala, and Léon Bottou. Wasserstein generative adversarial networks","authors":"Arjovsky","year":"2017","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Bogachev and Kolesnikov, 2012] Vladimir Igorevich Bo- gachev and Aleksandr Viktorovich Kolesnikov. The monge-kantorovich problem: Achievements, connec- tions, and perspectives. Russian Mathematical Surveys, 67(5):785-890, 2012.","title":"The monge-kantorovich problem: Achievements, connections, and perspectives","authors":"Vladimir Igorevich, Bogachev, Aleksandr Viktorovich, Kolesnikov","year":"2012","venue":"Russian Mathematical Surveys","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Cuturi and Doucet, 2014] Marco Cuturi and Arnaud Doucet. Fast computation of wasserstein barycenters. In ICML, pages 685-693, 2014.","title":"Fast computation of wasserstein barycenters","authors":"Marco Cuturi, Arnaud Doucet","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Cuturi, 2013] Marco Cuturi. Sinkhorn distances: Light- speed computation of optimal transport. In NeurIPS, pages 2292-2300, 2013.","title":"Sinkhorn distances: Lightspeed computation of optimal transport","authors":"Marco Cuturi","year":"2013","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Fang and Zhang, 2019] Jun-Peng Fang and Min-Ling Zhang. Partial multi-label learning via credible label elicitation. In AAAI, pages 3518-3525, 2019.","title":"Partial multi-label learning via credible label elicitation","authors":"Zhang ; Jun-Peng Fang, Min-Ling Fang, Zhang","year":"2019","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Feng et al., 2022] Lei Feng, Jun Huang, Senlin Shu, and Bo An. Regularized matrix factorization for multilabel learning with missing labels. IEEE TCYB, 52(5):3710- 3721, 2022.","title":"Regularized matrix factorization for multilabel learning with missing labels","authors":"Lei Feng, Jun Huang, Senlin Shu, Bo An","year":"2022","venue":"IEEE TCYB","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Frogner et al., 2015] Charlie Frogner, Chiyuan Zhang, Hos- sein Mobahi an Mauricio Araya-Polo, and Tomaso Pogglo. Learning with a wasserstein loss. In NeurIPS, pages 2053- 2061, 2015.","title":"Learning with a wasserstein loss","authors":"Charlie Frogner, Chiyuan Zhang, Hossein Mobahi An, Mauricio Araya-Polo, Tomaso Pogglo","year":"2015","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Hang and Zhang, 2023] Jun-Yi Hang and Min-Ling Zhang. Partial multi-label learning with probabilistic graphical disambiguation. In NeurIPS, 2023.","title":"Partial multi-label learning with probabilistic graphical disambiguation","authors":"Hang, Zhang ; Jun-Yi Hang, Min-Ling Zhang","year":"2023","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Huang et al., 2016] Gao Huang, Chuan Guo, Matt J. Kus- ner, Yu Sun, Kilian Q. Weinberger, and Fei Sha. Super- vised word mover's distance. In NeurIPS, pages 4862- 4870, 2016.","title":"Supervised word mover's distance","authors":"Gao Huang, Chuan Guo, Matt J Kusner, Yu Sun, Kilian Q Weinberger, Fei Sha","year":"2016","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Kusner et al., 2015] Matt J. Kusner, Yu Sun, Nicholas I. Kolkin, and Kilian Q. Weinberger. From word embeddings to document distances. In ICML, pages 957-966, 2015.","title":"From word embeddings to document distances","authors":"Matt J Kusner, Yu Sun, Nicholas I Kolkin, Kilian Q Weinberger","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Li and Wang, 2020] Ximing Li and Yang Wang. Recovering accurate labeling information from partially valid data for effective multi-label learning. In IJCAI, pages 1373-1380, 2020.","title":"Recovering accurate labeling information from partially valid data for effective multi-label learning","authors":"Ximing Li, Yang Wang","year":"2020","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2020/191"},{"position":12,"raw":"[Li et al., 2020] Ziwei Li, Gengyu Lyu, and Songhe Feng. Partial multi-label learning via multi-subspace representa- tion. In IJCAI, pages 2612-2618, 2020.","title":"Partial multi-label learning via multi-subspace representation","authors":"Ziwei Li, Gengyu Lyu, Songhe Feng","year":"2020","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2020/362"},{"position":13,"raw":"[Li et al., 2021] Changchun Li, Ximing Li, and Jihong Ouyang. Semi-supervised text classification with balanced deep representation distributions. In ACL-IJCNLP, pages 5044-5053, 2021.","title":"Semi-supervised text classification with balanced deep representation distributions","authors":"Li","year":"2021","venue":"ACL-IJCNLP","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Li et al., 2022] Changchun Li, Ximing Li, Lei Feng, and Ji- hong Ouyang. Who is your right mixup partner in positive and unlabeled learning. In ICLR, 2022.","title":"Who is your right mixup partner in positive and unlabeled learning","authors":"Li","year":"2022","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Li et al., 2023] Ximing Li, Yuanzhi Jiang, Changchun Li, Yiyuan Wang, and Jihong Ouyang. Learning with partial labels from semi-supervised perspective. In AAAI, pages 8666-8674, 2023.","title":"Learning with partial labels from semi-supervised perspective","authors":"Ximing Li, Yuanzhi Jiang, Changchun Li, Yiyuan Wang, Jihong Ouyang","year":"2023","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Lin et al., 2011] Zhouchen Lin, Risheng Liu, and Zhixun Su. Linearized alternating direction method with adaptive penalty for low-rank representation. In NeurIPS, pages 612-620, 2011.","title":"Linearized alternating direction method with adaptive penalty for low-rank representation","authors":"Lin","year":"2011","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Liu et al., 2010] Guangcan Liu, Zhouchen Lin, and Yong Yu. Robust subspace segmentation by low-rank represen- tation. In ICML, pages 663-670, 2010.","title":"Robust subspace segmentation by low-rank representation","authors":"Liu","year":"2010","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Liu et al., 2023] Bing-Qing Liu, Bin-Bin Jia, and Min-Ling Zhang. Towards enabling binary decomposition for partial multi-label learning. IEEE TPAMI, 2023.","title":"Towards enabling binary decomposition for partial multi-label learning","authors":"Liu","year":"2023","venue":"IEEE TPAMI","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Pele and Werman, 2009] Ofir Pele and Michael Werman. Fast and robust earth mover's distances. In ICCV, pages 460-467, 2009.","title":"Fast and robust earth mover's distances","authors":"Werman Pele, Ofir Pele, Michael Werman","year":"2009","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Rubner et al., 2000] Yossi Rubner, Carlo Tomasi, and Leonidas J. Guibas. The earth mover's distance as a metric for image retrieval. IJCV, 40(2):99-121, 2000.","title":"The earth mover's distance as a metric for image retrieval","authors":"Yossi Rubner, Carlo Tomasi, Leonidas J Guibas","year":"2000","venue":"IJCV","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Sun et al., 2019] Lijuan Sun, Songhe Feng, Tao Wang, Con- gyan Lang, and Yi Jin. Partial multi-label learning with low-rank and sparse decomposition. In AAAI, pages 5016- 5023, 2019.","title":"Partial multi-label learning with low-rank and sparse decomposition","authors":"Lijuan Sun, Songhe Feng, Tao Wang, Congyan Lang, Yi Jin","year":"2019","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Sun et al., 2022] Lijuan Sun, Songhe Feng, Jun Liu, Gengyu Lyu, and Congyan Lang. Global-local label corre- lation for partial multi-label learning. IEEE Transactions on Multimedia, 24:581-593, 2022.","title":"Global-local label correlation for partial multi-label learning","authors":"Lijuan Sun, Songhe Feng, Jun Liu, Gengyu Lyu, Congyan Lang","year":"2022","venue":"IEEE Transactions on Multimedia","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Villani, 2008] Cédric Villani. Optimal Transport: Old and New. Springer Berlin Heidelberg, 2008.","title":"Optimal Transport: Old and New","authors":"Cédric Villani","year":"2008","venue":null,"doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Wang et al., 2023] Haobo Wang, Shisong Yang, Gengyu Lyu, Weiwei Liu, Tianlei Hu, Ke Chen, Songhe Feng, and Gang Chen. Deep partial multi-label learning with graph disambiguation. arXiv preprint arXiv:2305.05882, 2023.","title":"Deep partial multi-label learning with graph disambiguation","authors":"Wang","year":"2023","venue":null,"doi":null,"url":null,"internal_url":"/proceedings/2023/479"},{"position":25,"raw":"[Xie and Huang, 2018a] Ming-Kun Xie and Sheng-Jun Huang. Partial multi-label learning. In AAAI, pages 4302-4309, 2018.","title":"Partial multi-label learning","authors":"Huang Xie, Ming-Kun Xie, Sheng-Jun Huang","year":"2018","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Xie and Huang, 2018b] Ming-Kun Xie and Sheng-Jun Huang. Partial multi-label learning. In AAAI, pages 4302-4309, 2018.","title":"Partial multi-label learning","authors":"Huang Xie, Ming-Kun Xie, Sheng-Jun Huang","year":"2018","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Xie and Huang, 2021] Ming-Kun Xie and Sheng-Jun Huang. Partial multi-label learning with noisy label identification. IEEE TPAMI, 44(7):3676-3687, 2021.","title":"Partial multi-label learning with noisy label identification","authors":"Huang Xie, Ming-Kun Xie, Sheng-Jun Huang","year":"2021","venue":"IEEE TPAMI","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Xie et al., 2021] Ming-Kun Xie, Feng Sun, and Sheng-Jun Huang. Partial multi-label learning with meta disambigua- tion. In ACM SIGKDD, pages 1904-1912, 2021.","title":"Partial multi-label learning with meta disambiguation","authors":"Ming-Kun Xie, Feng Sun, Sheng-Jun Huang","year":"2021","venue":"ACM SIGKDD","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Xie et al., 2023] Ming-Kun Xie, Jia-Hao Xiao, Hao-Zhe Liu, Gang Niu, Masashi Sugiyama, and Sheng-Jun Huang. Class-distribution-aware pseudo-labeling for semi-supervised multi-label learning. In NeurIPS, 2023.","title":"Class-distribution-aware pseudo-labeling for semi-supervised multi-label learning","authors":"Ming-Kun Xie, Jia-Hao Xiao, Hao-Zhe Liu, Gang Niu, Masashi Sugiyama, Sheng-Jun Huang","year":"2023","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Xu et al., 2020] Ning Xu, Yun-Peng Liu, and Xin Geng. Partial multi-label learning with label distribution. In AAAI, pages 6510-6517, 2020.","title":"Partial multi-label learning with label distribution","authors":"Ning Xu, Yun-Peng Liu, Xin Geng","year":"2020","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Yu et al., 2018] Guoxian Yu, Xia Chen, Carlotta Domeni- coni, Jun Wang, Zhao Li, Zili Zhang, and Xindong Wu. Feature-induced partial multi-label learning. In IEEE ICDM, pages 1398-1403, 2018.","title":"Feature-induced partial multi-label learning","authors":"Guoxian Yu, Xia Chen, Carlotta Domeniconi, Jun Wang, Zhao Li, Zili Zhang, Xindong Wu","year":"2018","venue":"IEEE ICDM","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Zhang and Fang, 2021] Min-Ling Zhang and Jun-Peng Fang. Partial multi-label learning via credible label elic- itation. IEEE TPAMI, 43(10):3587-3599, 2021.","title":"Partial multi-label learning via credible label elicitation","authors":"Fang Zhang, Min-Ling Zhang, Jun-Peng Fang","year":"2021","venue":"IEEE TPAMI","doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Zhang and Zhou, 2014a] Min-Ling Zhang and Zhi-Hua Zhou. A review on multi-label learning algorithms. IEEE TKDE, 26(8):1819-1837, 2014.","title":"A review on multi-label learning algorithms","authors":"Zhou Zhang, Min-Ling Zhang, Zhi-Hua Zhou","year":"2014","venue":"IEEE TKDE","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Zhang and Zhou, 2014b] Min-Ling Zhang and Zhi-Hua Zhou. A review on multi-label learning algorithms. IEEE TKDE, 26(8):1819-1837, 2014.","title":"A review on multi-label learning algorithms","authors":"Zhou Zhang, Min-Ling Zhang, Zhi-Hua Zhou","year":"2014","venue":"IEEE TKDE","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[Zhang et al., 2017] Yongqiang Zhang, Daming Shi, Junbin Gao, and Dansong Cheng. Low-rank-sparse subspace rep- resentation for robust regression. In IEEE CVPR, pages 2972-2981, 2017.","title":"Low-rank-sparse subspace representation for robust regression","authors":"Yongqiang Zhang, Daming Shi, Junbin Gao, Dansong Cheng","year":"2017","venue":"IEEE CVPR","doi":null,"url":null,"internal_url":null},{"position":36,"raw":"[Zhao and Zhou, 2018] Peng Zhao and Zhi-Hua Zhou. Label distribution learning by optimal transport. In AAAI, pages 4506-4513, 2018.","title":"Label distribution learning by optimal transport","authors":"Peng Zhao, Zhi-Hua Zhou","year":"2018","venue":"AAAI","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}