{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,21]],"date-time":"2026-02-21T18:36:05Z","timestamp":1771698965367,"version":"3.50.1"},"reference-count":36,"publisher":"Institute of Electronics, Information and Communications Engineers (IEICE)","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEICE Trans. Inf. & Syst."],"published-print":{"date-parts":[[2022,2,1]]},"DOI":"10.1587\/transinf.2021edp7035","type":"journal-article","created":{"date-parts":[[2022,1,31]],"date-time":"2022-01-31T22:16:58Z","timestamp":1643667418000},"page":"364-376","source":"Crossref","is-referenced-by-count":8,"title":["Learning from Noisy Complementary Labels with Robust Loss Functions"],"prefix":"10.1587","volume":"E105.D","author":[{"given":"Hiroki","family":"ISHIGURO","sequence":"first","affiliation":[{"name":"University of Tokyo"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Takashi","family":"ISHIDA","sequence":"additional","affiliation":[{"name":"University of Tokyo"},{"name":"RIKEN"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Masashi","family":"SUGIYAMA","sequence":"additional","affiliation":[{"name":"University of Tokyo"},{"name":"RIKEN"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"532","reference":[{"key":"1","unstructured":"[1] A. Krizhevsky, I. Sutskever, and G.E. Hinton, \u201cImageNet classification with deep convolutional neural networks,\u201d Advances in Neural Information Processing Systems, pp.1097-1105, Curran Associates, Inc., 2012."},{"key":"2","unstructured":"[2] J. Howe, Crowdsourcing: Why the Power of the Crowd Is Driving the Future of Business, 1 ed., Crown Publishing Group, USA, 2008."},{"key":"3","unstructured":"[3] T. Ishida, G. Niu, W. Hu, and M. Sugiyama, \u201cLearning from complementary labels,\u201d Advances in Neural Information Processing Systems, pp.5639-5649, Curran Associates, Inc., 2017."},{"key":"4","doi-asserted-by":"crossref","unstructured":"[4] X. Yu, T. Liu, M. Gong, and D. Tao, \u201cLearning with biased complementary labels,\u201d Proc. European Conference on Computer Vision, 2018. 10.1007\/978-3-030-01246-5_5","DOI":"10.1007\/978-3-030-01246-5_5"},{"key":"5","unstructured":"[5] T. Ishida, G. Niu, A. Menon, and M. Sugiyama, \u201cComplementary-label learning for arbitrary losses and models,\u201d Proc. 36th International Conference on Machine Learning, pp.2971-2980, PMLR, 2019."},{"key":"6","unstructured":"[6] Y.T. Chou, G. Niu, H.T. Lin, and M. Sugiyama, \u201cUnbiased risk estimators can mislead: A case study of learning with complementary labels,\u201d Proc. 37th International Conference on Machine Learning, pp.1929-1938, PMLR, 2020."},{"key":"7","unstructured":"[7] T. Zhang, \u201cStatistical analysis of some multi-category large margin classification methods,\u201d J. Mach. Learn. Res., vol.5, pp.1225-1251, 2004."},{"key":"8","unstructured":"[8] L. Feng, T. Kaneko, B. Han, G. Niu, B. An, and M. Sugiyama, \u201cLearning with multiple complementary labels,\u201d Proc. 37th International Conference on Machine Learning, pp.3072-3081, PMLR, 2020."},{"key":"9","doi-asserted-by":"publisher","unstructured":"[9] Y. Xu, M. Gong, J. Chen, T. Liu, K. Zhang, and K. Batmanghelich, \u201cGenerative-discriminative complementary learning,\u201d Proc. Thirty-Fourth AAAI Conference on Artificial Intelligence, pp.6526-6533, 2020. 10.1609\/aaai.v34i04.6126","DOI":"10.1609\/aaai.v34i04.6126"},{"key":"10","doi-asserted-by":"crossref","unstructured":"[10] Y. Kim, J. Yim, J. Yun, and J. Kim, \u201cNLNL: Negative learning for noisy labels,\u201d IEEE International Conference on Computer Vision, pp.101-110, 2019. 10.1109\/iccv.2019.00019","DOI":"10.1109\/ICCV.2019.00019"},{"key":"11","unstructured":"[11] A. Krizhevsky, \u201cLearning multiple layers of features from tiny images,\u201d Master's thesis, University of Toronto, 2009."},{"key":"12","unstructured":"[12] C. Zhang, S. Bengio, M. Hardt, B. Recht, and O. Vinyals, \u201cUnderstanding deep learning requires rethinking generalization,\u201d International Conference on Learning Representations, pp.1-15, 2017."},{"key":"13","unstructured":"[13] D. Arpit, S. Jastrz\u0119bski, N. Ballas, D. Krueger, E. Bengio, M.S. Kanwal, T. Maharaj, A. Fischer, A. Courville, Y. Bengio, and S. Lacoste-Julien, \u201cA closer look at memorization in deep networks,\u201d Proc. 34th International Conference on Machine Learning, pp.233-242, PMLR, 2017."},{"key":"14","doi-asserted-by":"publisher","unstructured":"[14] T. Liu and D. Tao, \u201cClassification with noisy labels by importance reweighting,\u201d IEEE Trans. Pattern Anal. Mach. Intell., vol.38, no.3, pp.447-461, 2016. 10.1109\/tpami.2015.2456899","DOI":"10.1109\/TPAMI.2015.2456899"},{"key":"15","doi-asserted-by":"crossref","unstructured":"[15] G. Patrini, A. Rozza, A.K. Menon, R. Nock, and L. Qu, \u201cMaking deep neural networks robust to label noise: A loss correction approach,\u201d IEEE Conference on Computer Vision and Pattern Recognition, pp.2233-2241, 2017. 10.1109\/cvpr.2017.240","DOI":"10.1109\/CVPR.2017.240"},{"key":"16","unstructured":"[16] B. Han, J. Yao, G. Niu, M. Zhou, I. Tsang, Y. Zhang, and M. Sugiyama, \u201cMasking: A new perspective of noisy supervision,\u201dAdvances in Neural Information Processing Systems, pp.5836-5846, Curran Associates, Inc., 2018."},{"key":"17","unstructured":"[17] D. Hendrycks, M. Mazeika, D. Wilson, and K. Gimpel, \u201cUsing trusted data to train deep networks on labels corrupted by severe noise,\u201d Advances in Neural Information Processing Systems, pp.10456-10465, Curran Associates, Inc., 2018."},{"key":"18","unstructured":"[18] X. Xia, T. Liu, N. Wang, B. Han, C. Gong, G. Niu, and M. Sugiyama, \u201cAre anchor points really indispensable in label-noise learning?,\u201d Advances in Neural Information Processing Systems, pp.6838-6849, Curran Associates, Inc., 2019."},{"key":"19","unstructured":"[19] Y. Yao, T. Liu, B. Han, M. Gong, J. Deng, G. Niu, and M. Sugiyama, \u201cDual T: Reducing estimation error for transition matrix in label-noise learning,\u201d Advances in Neural Information Processing Systems, pp.7260-7271, Curran Associates, Inc., 2020."},{"key":"20","unstructured":"[20] X. Xia, T. Liu, B. Han, N. Wang, M. Gong, H. Liu, G. Niu, D. Tao, and M. Sugiyama, \u201cPart-dependent label noise: Towards instance-dependent label noise,\u201d Advances in Neural Information Processing Systems, pp.7597-7610, Curran Associates, Inc., 2020."},{"key":"21","unstructured":"[21] B. Han, G. Niu, X. Yu, Q. Yao, M. Xu, I. Tsang, and M. Sugiyama, \u201cSIGUA: Forgetting may make learning with noisy labels more robust,\u201d Proc. 37th International Conference on Machine Learning, pp.4006-4016, PMLR, 2020."},{"key":"22","unstructured":"[22] M.D. Reid and R.C. Williamson, \u201cComposite binary losses,\u201d J. Mach. Learn. Res., vol.11, pp.2387-2422, 2010."},{"key":"23","unstructured":"[23] R.C. Williamson, E. Vernet, and M.D. Reid, \u201cComposite multiclass losses,\u201d J. Mach. Learn. Res., vol.17, no.1, pp.7860-7911, 2016."},{"key":"24","doi-asserted-by":"publisher","unstructured":"[24] N. Manwani and P.S. Sastry, \u201cNoise tolerance under risk minimization,\u201d IEEE Trans. Cybern., vol.43, no.3, pp.1146-1151, 2013. 10.1109\/tsmcb.2012.2223460","DOI":"10.1109\/TSMCB.2012.2223460"},{"key":"25","doi-asserted-by":"publisher","unstructured":"[25] A. Ghosh, N. Manwani, and P.S. Sastry, \u201cMaking risk minimization tolerant to label noise,\u201d Neurocomputing, vol.160, pp.93-107, 2015. 10.1016\/j.neucom.2014.09.081","DOI":"10.1016\/j.neucom.2014.09.081"},{"key":"26","unstructured":"[26] A. Ghosh, H. Kumar, and P.S. Sastry, \u201cRobust loss functions under label noise for deep neural networks,\u201d Proc. Thirty-First AAAI Conference on Artificial Intelligence, pp.1919-1925, 2017."},{"key":"27","unstructured":"[27] Z. Zhang and M. Sabuncu, \u201cGeneralized cross entropy loss for training deep neural networks with noisy labels,\u201d Advances in Neural Information Processing Systems, pp.8778-8788, Curran Associates, Inc., 2018."},{"key":"28","doi-asserted-by":"crossref","unstructured":"[28] Y. Wang, X. Ma, Z. Chen, Y. Luo, J. Yi, and J. Bailey, \u201cSymmetric cross entropy for robust learning with noisy labels,\u201d IEEE International Conference on Computer Vision, pp.322-330, 2019. 10.1109\/iccv.2019.00041","DOI":"10.1109\/ICCV.2019.00041"},{"key":"29","doi-asserted-by":"publisher","unstructured":"[29] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, \u201cGradient-based learning applied to document recognition,\u201d Proc. IEEE, vol.86, no.11, pp.2278-2324, 1998. 10.1109\/5.726791","DOI":"10.1109\/5.726791"},{"key":"30","unstructured":"[30] H. Xiao, K. Rasul, and R. Vollgraf, \u201cFashion-MNIST: a novel image dataset for benchmarking machine learning algorithms,\u201d arXiv preprint arXiv:1708.07747, 2017."},{"key":"31","unstructured":"[31] T. Clanuwat, M. Bober-Irizar, A. Kitamoto, A. Lamb, K. Yamamoto, and D. Ha, \u201cDeep learning for classical Japanese literature,\u201d Neural Information Processing Systems Workshop on Machine Learning for Creativity and Design, pp.1-8, 2018."},{"key":"32","unstructured":"[32] M. Lin, Q. Chen, and S. Yan, \u201cNetwork in network,\u201d arXiv preprint arXiv:1312.4400, 2013."},{"key":"33","unstructured":"[33] S. Ioffe and C. Szegedy, \u201cBatch normalization: Accelerating deep network training by reducing internal covariate shift,\u201d Proc. 32nd International Conference on Machine Learning, pp.448-456, PMLR, 2015."},{"key":"34","unstructured":"[34] D.P. Kingma and J. Ba, \u201cAdam: A method for stochastic optimization,\u201d International Conference on Learning Representations, pp.1-15, 2015."},{"key":"35","unstructured":"[35] R. Kiryo, G. Niu, M.C. du Plessis, and M. Sugiyama, \u201cPositive-unlabeled learning with non-negative risk estimator,\u201d Advances in Neural Information Processing Systems, pp.1675-1685, Curran Associates, Inc., 2017."},{"key":"36","unstructured":"[36] N. Lu, T. Zhang, G. Niu, and M. Sugiyama, \u201cMitigating overfitting in supervised classification from two unlabeled datasets: A consistent risk correction approach,\u201d International Conference on Artificial Intelligence and Statistics, pp.1115-1125, PMLR, 2020."}],"container-title":["IEICE Transactions on Information and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/transinf\/E105.D\/2\/E105.D_2021EDP7035\/_pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,9]],"date-time":"2024-05-09T04:55:41Z","timestamp":1715230541000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/transinf\/E105.D\/2\/E105.D_2021EDP7035\/_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,1]]},"references-count":36,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2022]]}},"URL":"https:\/\/doi.org\/10.1587\/transinf.2021edp7035","relation":{},"ISSN":["0916-8532","1745-1361"],"issn-type":[{"value":"0916-8532","type":"print"},{"value":"1745-1361","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,2,1]]},"article-number":"2021EDP7035"}}