Abstract
Deep neural networks have achieved significant success in the artificial intelligence community and various downstream tasks. They encode images or texts into dense feature representations and are supervised by a large amount of labeled data. Due to the expensiveness of high-quality labeled data, a huge number of easy-to-access instances are collected to conduct supervised learning. However, they have not been annotated by experts and thus can contain numerous noisy instances, which will degrade the performance. To learn robust feature representations despite misleading noisy labels, we employ supervised contrastive learning to directly perform supervision in the hidden space, rather than in the prediction space like the prevalent cross-entropy loss function. However, cutting-edge noisy label learning methods with supervised contrastive learning always discard the data considered to be noisy, and thus cannot tolerate high-ratio noisy datasets. Therefore, we propose a novel training strategy named Supervised Contrastive Learning with Corrected Labels (Scl \(^2\)) to defend against the attack of noisy labels. Scl \(^2\) corrects the noisy labels with an empirical small-loss assumption and conducts supervised contrastive learning using these corrected data. Specifically, we employ the generated soft labels as supervisory information to facilitate our implementation of supervised contrastive learning. This expansion of contrastive learning ensures the integrity of the supervisory information while effectively enhancing the learning process. In addition, samples sharing the same soft labels are treated as positive sample pairs, while those with different soft labels are considered to be negative sample pairs. With this strategy, the representations from neural networks keep the local discrimination in one mini-batch. Besides, we also employ a prototype contrastive learning technique to ensure global discrimination. Our Scl \(^2\) has demonstrated excellent performance on numerous benchmark datasets, showcasing its effectiveness in various standardized evaluation scenarios. Additionally, our model has proven to be highly valuable when applied to real-world noisy datasets.
Similar content being viewed by others
Data availability and access
Source code for the experiments is available at https://github.com/ChenyangLu922/SCL2.git.
References
Rawat W, Wang Z (2017) Deep convolutional neural networks for image classification: a comprehensive review. Neural Comput 29(9):2352–2449
Krizhevsky A, Sutskever I, Hinton GE (2017) Imagenet classification with deep convolutional neural networks. Commun ACM 60(6):84–90
Khosla P, Teterwak P, Wang C et al (2020) Supervised contrastive learning. Adv Neural Inf Process Syst 33:18661–18673
Chen T, Kornblith S, Norouzi M et al (2020) A simple framework for contrastive learning of visual representations. In: International Conference on Machine Learning. PMLR, pp 1597–1607
Li C, Li X, Ouyang J (2021b) Semi-supervised text classification with balanced deep representation distributions. In: Annual Meeting of the Association for Computational Linguistics. pp 5044–5053
Feng S, Wang B, Yang Z et al (2022) Aspect-based sentiment analysis with attention-assisted graph and variational sentence representation. Knowl-Based Syst 258:109975
Ouyang J, Wang Y, Li X et al (2022) Weakly-supervised text classification with Wasserstein Barycenters regularization. In: International Joint Conference on Artificial Intelligence. pp 3373–3379
Li X, Wang B, Wang Y et al (2023b) Weakly supervised prototype topic model with discriminative seed words: modifying the category prior by self-exploring supervised signals. Soft Comput 27(9):5397–5410
Raza K, Singh NK (2021) A tour of unsupervised deep learning for medical image analysis. Current Medical Imaging 17(9):1059–1077
Paolacci G, Chandler J, Ipeirotis PG (2010) Running experiments on Amazon mechanical Turk. Judgm Decis Mak 5(5):411–419
Arpit D, Jastrzebski S, Ballas N et al (2017) A closer look at memorization in deep networks. In: Proceedings of the 34th International Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, 6–11 August 2017, pp 233–242
Zhang C, Bengio S, Hardt M et al (2021a) Understanding deep learning (still) requires rethinking generalization. Commun ACM 64(3):107–115
Zhang Z, Sabuncu MR (2018) Generalized cross entropy loss for training deep neural networks with noisy labels. pp 8792–8802
Tanaka D, Ikami D, Yamasaki T et al (2018) Joint optimization framework for learning with noisy labels. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp 5552–5560
Ma X, Huang H, Wang Y et al (2020) Normalized loss functions for deep learning with noisy labels. In: International Conference on Machine Learning. pp 6543–6553
Han B, Yao Q, Yu X et al (2018) Co-teaching: robust training of deep neural networks with extremely noisy labels. Adv Neural Inf Process Syst 31:8536–8546
Tan C, Xia J, Wu L et al (2021) Co-learning: Learning from noisy labels with self-supervision. In: Proceedings of the 29th ACM International Conference on Multimedia. pp 1405–1413
Yu X, Han B, Yao J et al (2019) How does disagreement help generalization against label corruption? In: International Conference on Machine Learning. PMLR, pp 7164–7173
Wei H, Feng L, Chen X et al (2020) Combating noisy labels by agreement: A joint training method with co-regularization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 13723–13732
Xiao T, Xia T, Yang Y et al (2015) Learning from massive noisy labeled data for image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp 2691–2699
Wang Y, Liu W, Ma X et al (2018) Iterative learning with open-set noisy labels. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp 8688–8696
Yao Y, Sun Z, Zhang C et al (2021) JO-SRC: a contrastive approach for combating noisy labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 5192–5201
Li S, Xia X, Ge S et al (2022a) Selective-supervised contrastive learning with noisy labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 316–325
Huang B, Lin Y, Xu C (2022) Contrastive label correction for noisy label learning. Inf Sci 611:173–184
He K, Fan H, Wu Y et al (2020) Momentum contrast for unsupervised visual representation learning. In: Proceedings of the IEEE/CVF Conference On Computer Vision and Pattern Recognition. pp 9729–9738
Wang B, Ding L, Zhong Q et al (2022) A contrastive cross-channel data augmentation framework for aspect-based sentiment analysis. In: International Conference on Computational Linguistics. pp 6691–6704
Ortego D, Arazo E, Albert P et al (2021) Multi-objective interpolation training for robustness to label noise. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 6606–6615
Natarajan N, Dhillon IS, Ravikumar P et al (2013) Learning with noisy labels. Adv Neural Inf Proces Syst 26
Patrini G, Rozza A, Menon AK et al (2017) Making deep neural networks robust to label noise: a loss correction approach. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp 2233–2241
Xia X, Liu T, Wang N et al (2019) Are anchor points really indispensable in label-noise learning? pp 6835–6846
Goldberger J, Ben-Reuven E (2017) Training deep neural-networks using a noise adaptation layer. In: International Conference on Learning Representations
Ghosh A, Kumar H, Sastry PS (2017) Robust loss functions under label noise for deep neural networks. In: Proceedings of the AAAI Conference on Artificial Intelligence. pp 1919–1925
Wang Y, Ma X, Chen Z et al (2019) Symmetric cross entropy for robust learning with noisy labels. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp 322–330
Jiang L, Zhou Z, Leung T et al (2018) Mentornet: learning data-driven curriculum for very deep neural networks on corrupted labels. In: International Conference on Machine Learning. PMLR, pp 2309–2318
Chen P, Liao B, Chen G et al (2019) Understanding and utilizing deep neural networks trained with noisy labels. In: International Conference on Machine Learning. PMLR, pp 1062–1070
Li X, Jiang Y, Li C et al (2023a) Learning with partial labels from semi-supervised perspective. In: Proceedings of the AAAI Conference on Artificial Intelligence. pp 8666–8674
Li X, Wang Y (2020) Recovering accurate labeling information from partially valid data for effective multi-label learning. In: Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI 2020. pp 1373–1380
Li C, Li X, Feng L et al (2021a) Who is your right mixup partner in positive and unlabeled learning. In: International Conference on Learning Representations
Yu K, Lin TR, Ma H et al (2021) A multi-stage semi-supervised learning approach for intelligent fault diagnosis of rolling bearing using data augmentation and metric learning. Mech Syst Signal Process 146:107043
Li X, Lu P, Hu L et al (2022b) A novel self-learning semi-supervised deep learning network to detect fake news on social media. Multimed Tools Appl 81(14):19341–19349
Yan Y, Xu Z, Tsang IW et al (2016) Robust semi-supervised learning through label aggregation. In: Proceedings of the AAAI Conference on Artificial Intelligence. pp 2244–2250
Nguyen DT, Mummadi CK, Ngo T et al (2020) Self: learning to filter noisy labels with self-ensembling. In: 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020
Ding Y, Wang L, Fan D et al (2018) A semi-supervised two-stage approach to learning from noisy labels. In: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, pp 1215–1224
Li J, Socher R, Hoi SCH (2020) Dividemix: learning with noisy labels as semi-supervised learning. In: 8th International Conference on Learning Representations, ICLR 2020,Addis Ababa, Ethiopia, April 26–30, 2020
Berthelot D, Carlini N, Goodfellow IJ et al (2019) Mixmatch: a holistic approach to semi-supervised learning. Adv Neural Inf Proces Syst 32:5050–5060
Grill J, Strub F, Altché F et al (2020) Bootstrap your own latent-a new approach to self-supervised learning. Adv Neural Inf Process Syst 33:21271–21284
Chen X, He K (2021) Exploring simple Siamese representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 15750–15758
Yi L, Liu S, She Q et al (2022) On learning contrastive representations for learning with noisy labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp 16661–16670
Reed SE, Lee H, Anguelov D et al (2015) Training deep neural networks on noisy labels with bootstrapping. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7–9, 2015, Workshop Track Proceedings
Lu Y, He W (2022) SELC: self-ensemble label correction improves learning with noisy labels. In: Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria, 23–29 July 2022. pp 3278–3284
Krizhevsky A (2009) Learning multiple layers of features from tiny images. Master’s thesis, University of Tront
Song H, Kim M, Lee J (2019) Selfie: Refurbishing unclean samples for robust deep learning. In: International Conference on Machine Learning. PMLR, pp 5907–5915
Liu S, Niles-Weed J, Razavian N et al (2020) Early-learning regularization prevents memorization of noisy labels. Adv Neural Inf Proces Syst 33:20331–20342
Chen P, Ye J, Chen G et al (2021) Beyond class-conditional assumption: A primary attempt to combat instance-dependent label noise. In: Proceedings of the AAAI Conference on Artificial Intelligence. pp 11442–11450
Zhang Y, Zheng S, Wu P et al (2021b) Learning with feature-dependent label noise: a progressive approach. In: International Conference on Learning Representations
Funding
This work was partially supported by the National Natural Science Foundation of China (NSFC) [No.62006094, No.62276113] and the Project Funded by China Postdoctoral Science Foundation (No.2022M721321).
Author information
Authors and Affiliations
Contributions
Jihong Ouyang: Resources, Project administration, Supervision. Chenyang Lu: Methodology, Software, Investigation, Writing - Original Draft, Data Curation, Validation. Bing Wang: Writing - Review & Editing, Visualization. Changchun Li: Conceptualization, Supervision.
Corresponding author
Ethics declarations
Ethical and informed consent for data used
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. We the undersigned declare that this manuscript entitled “Supervised Contrastive Learning with Corrected Labels for Noisy Label Learning” is original, has not been published before, and is not currently being considered for publication elsewhere.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Ouyang, J., Lu, C., Wang, B. et al. Supervised contrastive learning with corrected labels for noisy label learning. Appl Intell 53, 29378–29392 (2023). https://doi.org/10.1007/s10489-023-05018-0
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-023-05018-0