Abstract
Label noise is often contained in the training data due to various human factors or measurement errors, which significantly causes a negative effect on classifiers. Despite many previous methods that have been proposed to learn robust classifiers, they are mainly based on the single-view feature. On the other hand, although existing multi-view classification methods benefit from the more comprehensive information, they rarely consider label noise. In this paper, we propose a novel label-noise robust classification model with multi-view learning to overcome these limitations. In the proposed model, not only the classifier learning but also the label-noise removal can benefit from the multi-view information. Specifically, we relax the label matrix of the basic multi-view least squares regression model, and develop a nonlinear transformation with a natural probabilistic approximation in the process of labels, which is conveniently optimized and beneficial to improve the discriminative ability of classifiers. Moreover, we preserve the intrinsic manifold structure of multi-view data on the relaxed label matrix, facilitating the process of label relaxation. For optimizing the proposed model with the nonlinear transformation, we derive a lemma about the partial derivation of the softmax related function, and develop an efficient alternating algorithm. Experimental evaluations on six real-world datasets confirm the advantages of the proposed method, compared to the related state-of-the-art methods.
References
Frénay B, Kabán A. A comprehensive introduction to label noise. In: Proceedings of the 2014 European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges, 2014
Oh Y, Kim B, Ham B. Background-aware pooling and noise-aware loss for weakly-supervised semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). virtual, 2021. 6913–6922
Algan G, Ulusoy I. Image classification with deep learning in the presence of noisy labels: A survey. Knowl Based Syst, 2021, 215: 106771
Jiang J, Ma J, Wang Z, et al. Hyperspectral image classification in the presence of noisy labels. IEEE Trans Geosci Remote Sens, 2019, 57: 851–865
Chen P, Ye J, Chen G, et al. Beyond class-conditional assumption: A primary attempt to combat instance-dependent label noise. AAAI, 2021, 35: 11442–11450
Xia X, Liu T, Han B, et al. Part-dependent label noise: Towards instance-dependent label noise. In: Proceedings of the Advances in Neural Information Processing Systems 33 (NeurIPS 2020). Vancouver, 2020
Frenay B, Verleysen M. Classification in the presence of label noise: A survey. IEEE Trans Neural Netw Learn Syst, 2014, 25: 845–869
Miranda ALB, Garcia L P F, Carvalho A C P L F, et al. Use of classification algorithms in noise detection and elimination. In: Proceedings of the Hybrid Artificial Intelligence Systems. Salamanca, 2009. 417–424
Jiang L, Zhou Z, Leung T, et al. Mentornet: Learning data-driven curriculum for very deep neural networks on corrupted labels. In: Proceedings of the 35th International Conference on Machine Learning. Stockholmsmässan, Stockholm, 2018. 2309–2318
Natarajan N, Dhillon I S, Ravikumar P, et al. Learning with noisy labels. In: Proceedings of the Neural Information Processing Systems. Lake Tahoe, 2013. 1196–1204
Gao W, Wang L, Li Y, et al. Risk minimization in the presence of label noise. In: Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. Phoenix, 2016. 1575–1581
Han B, Yao Q, Yu X, et al. Co-teaching: Robust training of deep neural networks with extremely noisy labels. In: Proceedings of the 32nd International Conference on Neural Information Processing Systems. Montreal, 2018. 8536–8546
Han B, Yao J, Niu G, et al. Masking: A new perspective of noisy supervision. In: Proceedings of the Neural Information Processing Systems. Montreal, 2018. 5841–5851
Luo Y, Han B, Gong C. A bi-level formulation for label noise learning with spectral cluster discovery. In: Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence. Yokohama, 2020. 2605–2611
Wei Y, Gong C, Chen S, et al. Harnessing side information for classification under label noise. IEEE Trans Neural Netw Learn Syst, 2020, 31: 3178–3192
Yang Z, Li Q, Wenyin L, et al. Shared multi-view data representation for multi-domain event detection. IEEE Trans Pattern Anal Mach Intell, 2020, 42: 1243–1256
Zhu X, Li X, Zhang S. Block-row sparse multiview multilabel learning for image classification. IEEE Trans Cybern, 2016, 46: 450–461
Zhao D, Gao Q, Lu Y, et al. Consistency and diversity neural network multi-view multi-label learning. Knowl Based Syst, 2021, 218: 106841
Zhang Z, Qin Z, Li P, et al. Multi-view discriminative learning via joint non-negative matrix factorization. In: Proceedings of the Database Systems for Advanced Applications. Gold Coast, 2018. 542–557
Yang M, Deng C, Nie F. Adaptive-weighting discriminative regression for multi-view classification. Pattern Recogn, 2019, 88: 236–245
Tao H, Hou C, Yi D, et al. Multiview classification with cohesion and diversity. IEEE Trans Cybern, 2020, 50: 2124–2137
Bylander T. Learning linear threshold functions in the presence of classification noise. In: Proceedings of the Seventh Annual Conference on Computational Learning Theory. New Brunswick, 1994. 340–347
Patrini G, Nielsen F, Nock R, et al. Loss factorization, weakly supervised learning and label noise robustness. In: Proceedings of the 33rd International Conference on International Conference on Machine Learning. New York, 2016. 708–717
Tanaka D, Ikami D, Yamasaki T, et al. Joint optimization framework for learning with noisy labels. In: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, 2018. 5552–5560
Xia X, Liu T, Wang N, et al. Are anchor points really indispensable in label-noise learning? In: Proceedings of the Neural Information Processing Systems. Vancouver, 2019. 6835–6846
Collier M, Mustafa B, Kokiopoulou E, et al. Correlated input-dependent label noise in large-scale image classification. In: Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). virtual, 2021. 1551–1560
Cheng J L, Bie L, Zhao X B, et al. Visual information quantification for object recognition and retrieval. Sci China Tech Sci, 2021, 64: 2618–2626
Bickel S, Scheffer T. Multi-view clustering. In: Proceedings of the Fourth IEEE International Conference on Data Mining. Brighton, 2004. 19–26
Wang R, Li L, Tao X, et al. Trio-based collaborative multi-view graph clustering with multiple constraints. Inf Process Manage, 2021, 58: 102466
Liang N, Yang Z, Li L, et al. Incomplete multiview clustering with cross-view feature transformation. IEEE Trans Artif Intell, 2022, 3: 749–762
Yang Z, Liang N, Yan W, et al. Uniform distribution non-negative matrix factorization for multiview clustering. IEEE Trans Cybern, 2021, 51: 3249–3262
Liang N, Yang Z, Li Z, et al. Multi-view clustering by non-negative matrix factorization with co-orthogonal constraints. Knowl Based Syst, 2020, 194: 105582
Li Z, Yang Z, Zhao H, et al. Direct-optimization-based DC dictionary learning with the MCP regularizer. IEEE Trans Neural Netw Learn Syst, 2021, doi: https://doi.org/10.1109/TNNLS.2021.3114400
Li Z, Li Y, Tan B, et al. Structured sparse coding with the group log-regularizer for key frame extraction. IEEE CAA J Autom Sin, 2022, 9: 1818–1830
Zhao H, Li Z, Chen W, et al. Accelerated partially shared dictionary learning with differentiable scale-invariant sparsity for multiview clustering. IEEE Trans Neural Netw Learn Syst, 2022, doi: https://doi.org/10.1109/TNNLS.2022.3153310
Qiu Y N, Zhou G X, Chen X Q, et al. Semi-supervised non-negative Tucker decomposition for tensor data representation. Sci China Tech Sci, 2021, 64: 1881–1892
Cai H, Liu B, Xiao Y, et al. Semi-supervised multi-view clustering based on orthonormality-constrained nonnegative matrix factorization. Inf Sci, 2020, 536: 171–184
Liang N, Yang Z, Li Z, et al. Semi-supervised multi-view clustering with graph-regularized partially shared non-negative matrix factorization. Knowl Based Syst, 2020, 190: 105185
Zhang D P, Luo Y H, Yu Y Y, et al. Semi-supervised multi-view clustering with dual hypergraph regularized partially shared non-negative matrix factorization. Sci China Tech Sci, 2022, 65: 1349–1365
Nie F, Cai G, Li X. Multi-view clustering and semi-supervised classification with adaptive neighbours. In: Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco, 2017. 2408–2414
Nie F, Tian L, Wang R, et al. Multiview semi-supervised learning model for image classification. IEEE Trans Knowl Data Eng, 2020, 32: 2389–2400
Zhang B, Qiang Q, Wang F, et al. Fast multi-view semi-supervised learning with learned graph. IEEE Trans Knowl Data Eng, 2022, 34: 286–299
Abudu S, Cui C L, King J P, et al. Modeling of daily pan evaporation using partial least squares regression. Sci China Tech Sci, 2011, 54: 163–174
Jamil W, Bouchachia A. Competitive normalized least-squares regression. IEEE Trans Neural Netw Learn Syst, 2021, 32: 3262–3267
Du X, Zhang L, Gao Y. A class of augmented Lagrangians for equality constraints in nonlinear programming problems. Appl Math Comput, 2006, 172: 644–663
Li K, Qi G J, Ye J, et al. Linear subspace ranking hashing for cross-modal retrieval. IEEE Trans Pattern Anal Mach Intell, 2017, 39: 1825–1838
Liang N, Yang Z, Li Z, et al. Semi-supervised multi-view learning by using label propagation based non-negative matrix factorization. Knowl Based Syst, 2021, 228: 107244
Liang N, Yang Z, Xie S. Incomplete multi-view clustering with sample-level auto-weighted graph fusion. IEEE Trans Knowl Data Eng, 2022, doi: https://doi.org/10.1109/TKDE.2022.3171911
Ding C H Q, Tao Li C H Q, Jordan M I. Convex and semi-nonnegative matrix factorizations. IEEE Trans Pattern Anal Mach Intell, 2010, 32: 45–55
Park J Y, Kim J H. Incremental class learning for hierarchical classification. IEEE Trans Cybern, 2020, 50: 178–189
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the Key-Area Research and Development Program of Guangdong Province (Grant No. 2019B010154002), the Guangdong Natural Science Foundation (Grant No. 2022A1515010688), and the National Natural Science Foundation of China (Grant No. 61722304).
Rights and permissions
About this article
Cite this article
Liang, N., Yang, Z., Li, L. et al. Label-noise robust classification with multi-view learning. Sci. China Technol. Sci. 66, 1841–1854 (2023). https://doi.org/10.1007/s11431-021-2139-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11431-021-2139-0