Abstract
Clean data is critical to the success of deep learning training. In practice, noisy labels are often included in the data. When the network learns these noisy labels, it will cause the model to degrade. In order to alleviate this problem, the sample reweighting method based on meta-learning has been proposed in recent years. It can reduce the negative impact of noisy labels by adjusting the model’s learning degree to the sample. It also make the model pay more attention to the cleaner data in the dataset. Based on this, we propose Meta-Learning with Normalized Projection Loss Reweighting (MLNP), which is a sample reweighting method based on meta learning. This strategy directs the classification network to identify potentially clean data in the dataset with a higher weight. The weight is based on the Euclidean distance between features and projection similarity of the meta set sample and the training set sample. Furthermore, learning data with less weight is more likely to be noisy. Through experiments, we show the robustness of MLNP and achieves advanced performance on a range of datasets.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Figa_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-024-03591-6/MediaObjects/11760_2024_3591_Fig5_HTML.png)
Similar content being viewed by others
References
Sun, Z., Yao, Y., Wei, X.-S., Zhang, Y., Shen, F., Wu, J., Zhang, J., Shen, H.T.: Webly supervised fine-grained recognition: benchmark datasets and an approach. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 10602–10611 (2021)
Wah, C., Branson, S., Welinder, P., Perona, P., Belongie, S.: The caltech-ucsd birds-200-2011 dataset (2011)
Maji, S., Rahtu, E., Kannala, J., Blaschko, M.B., Vedaldi, A.: Fine-grained visual classification of aircraft. CoRR abs/1306.5151 (2013)
Krause, J., Stark, M., Deng, J., Fei-Fei, L.: 3d object representations for fine-grained categorization. In: 2013 IEEE International Conference on Computer Vision Workshops, pp. 554–561 (2013). https://doi.org/10.1109/ICCVW.2013.77
Li, D., Li, X., Wang, B.: Texture direction recognition of wooden beams and columns based on improved meta-learning. Signal, Image Video Process. 17(8), 4447–4454 (2023). https://doi.org/10.1007/s11760-023-02678-w
Zhao, Z., Chen, G., Lin, Y.: Temporal-masked skeleton-based action recognition with supervised contrastive learning. Signal Image Video Process. 17(5), 2267–2275 (2023)
Fu, B., Dong, Y., Fu, S., Wu, Y., Ren, Y., Thanh, D.N.H.: Multistage supervised contrastive learning for hybrid-degraded image restoration. Signal, Image Video Process. 17, 573–581 (2023). https://doi.org/10.1007/s11760-022-02262-8
Dong, H., Zhang, T., Zhang, T., Wei, L.: Supervised learning-based retinal vascular segmentation by m-unet full convolutional neural network. Signal, Image Video Process. 16, 1755–1761 (2022). https://doi.org/10.1007/s11760-022-02132-3
Lin, C., Mao, X., Qiu, C., Zou, L.: Dtcnet: transformer-cnn distillation for super-resolution of remote sensing image. IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. 17, 11117–11133 (2024). https://doi.org/10.1109/JSTARS.2024.3409808
Ye, A., Xiao, X., Xiao, H., Jiang, C., Lin, C.: Acgnd: towards lower complexity and fast solution for dynamic tensor inversion. Complex Intell. Syst. 31, 1–15 (2024)
Lin, C., Qiu, C., Jiang, H., Zou, L.: A deep neural network based on prior-driven and structural preserving for SAR image despeckling. IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. 16, 6372–6392 (2023). https://doi.org/10.1109/JSTARS.2023.3292325
Zhang, C., Bengio, S., Hardt, M., Recht, B., Vinyals, O.: Understanding deep learning (still) requires rethinking generalization. Commun. ACM 64(3), 107–115 (2021). https://doi.org/10.1145/3446776
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv: Learning, (2015)
Shorten, C., Khoshgoftaar, T.M.: A survey on image data augmentation for deep learning. J. Big Data (2019). https://doi.org/10.1186/s40537-019-0197-0
Loshchilov, I., Hutter, F.: Decoupled weight decay regularization. Learning, Learning (2017)
Xiao, T., Xia, T., Yang, Y., Huang, C., Wang, X.: Learning from massive noisy labeled data for image classification. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2691–2699 (2015). https://doi.org/10.1109/CVPR.2015.7298885
Chen, X., Gupta, A.: Webly supervised learning of convolutional networks. In: 2015 IEEE International Conference on Computer Vision (ICCV), pp. 1431–1439 (2015). https://doi.org/10.1109/ICCV.2015.168
Sukhbaatar, S., Bruna, J., Paluri, M., Bourdev, L., Fergus, R.: Training convolutional networks with noisy labels. arXiv preprint arXiv:1406.2080 (2014)
Liu, T., Tao, D.: Classification with noisy labels by importance reweighting. IEEE Trans. Pattern Anal. Mach. Intell. 38(3), 447–461 (2016). https://doi.org/10.1109/TPAMI.2015.2456899
Zhang, H., Xing, X., Liu, L.: Dualgraph: a graph-based method for reasoning about label noise. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2021). https://doi.org/10.1109/cvpr46437.2021.00953
Shu, J., Xie, Q., Yi, L., Zhao, Q., Zhou, S., Xu, Z., Meng, D.: Meta-weight-net: learning an explicit mapping for sample weighting. Neural Information Processing Systems, Neural Information Processing Systems (2019)
Belton, N., Hagos, M.T., Lawlor, A., Curran, K.M.: Fewsome: One-class few shot anomaly detection with siamese networks. In: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 2978–2987 (2023). https://doi.org/10.1109/CVPRW59228.2023.00299
Yue, C., Huang, R., Towey, D., Xian, Z., Wu, G.: An entropy-based group decision-making approach for software quality evaluation. Expert. Syst. Appl. 238, 121979 (2024)
Yue, C., Huang, R., Towey, D., Xian, Z., Wu, G.: An entropy-based group decision-making approach for software quality evaluation. Expert. Syst. Appl. 238, 121979 (2024). https://doi.org/10.1016/j.eswa.2023.121979
Oza, P., Patel, V.M.: One-class convolutional neural network. IEEE Signal Process. Lett. 26, 277–281 (2019). https://doi.org/10.1109/lsp.2018.2889273
Jewell, J., Khazaie, V., Mohsenzadeh, Y.: Oled: one-class learned encoder-decoder network with adversarial context masking for novelty detection. arXiv: Computer Vision and Pattern Recognition (2021)
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., Houlsby, N.: An image is worth 16x16 words: transformers for image recognition at scale. arXiv: Computer Vision and Pattern Recognition (2020)
Cohen, M., Avidan, S.: Transformaly–two (feature spaces) are better than one
Li, C.-L., Sohn, K., Yoon, J., Pfister, T.: Cutpaste: self-supervised learning for anomaly detection and localization. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2021). https://doi.org/10.1109/cvpr46437.2021.00954
Wang, G., Wang, Y., Qin, J., Zhang, D., Bao, X., Huang, D.: Unilaterally aggregated contrastive learning with hierarchical augmentation for anomaly detection. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6888–6897 (2023)
Yue, Z., Jia, Y.: A direct projection-based group decision-making methodology with crisp values and interval data. Soft Comput. 21, 2395–2405 (2017). https://doi.org/10.1007/s00500-015-1953-5
Wang, Q.: A two-tuple linguistics muti-attribute group decision making method based on bi-directional projection operator. Mathematics in Practice and Theory (2015)
Fu, C., Gao, X., Liu, M., Liu, X., Han, L., Chen, J.: Grap: grey risk assessment based on projection in ad hoc networks. J. Parallel Distrib. Comput. 71(9), 1249–1260 (2011). https://doi.org/10.1016/j.jpdc.2010.11.012
Zheng, G., Jing, Y., Huang, H., Gao, Y.: Application of improved grey relational projection method to evaluate sustainable building envelope performance. Appl. Energy 87(2), 710–720 (2010). https://doi.org/10.1016/j.apenergy.2009.08.020
Xu, Z., Hu, H.: Projection models for intuitionistic fuzzy multiple attribute decision making. Int. J. Inf. Technol. Decis. Mak. 09(02), 267–280 (2010). https://doi.org/10.1142/s0219622010003816
Shu, J., Yuan, X., Meng, D., Xu, Z.: Cmw-net: learning a class-aware sample weighting mapping for robust deep learning. CoRR abs/2202.05613 (2022) 2202.05613
Yi, M., Hou, L., Shang, L., Jiang, X., Li, Q., Ma, Z.: Reweighting augmented samples by minimizing the maximal expected loss. International Conference on Learning Representations,International Conference on Learning Representations (2021)
Krizhevsky, A., Hinton, G., et al.: Learning multiple layers of features from tiny images (2009)
Shu, J., Yuan, X., Meng, D., Xu, Z.: Cmw-net: learning a class-aware sample weighting mapping for robust deep learning. IEEE Trans. Pattern Anal. Mach. Intell. 45(10), 11521–11539 (2023). https://doi.org/10.1109/TPAMI.2023.3271451
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Lin, T.-Y., RoyChowdhury, A., Maji, S.: Bilinear cnn models for fine-grained visual recognition. In: 2015 IEEE International Conference on Computer Vision (ICCV) (2015). https://doi.org/10.1109/iccv.2015.170
Dubey, A., Gupta, O., Guo, P., Raskar, R., Farrell, R., Naik, N.: Pairwise confusion for fine-grained visual classification. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 70–86 (2018)
Zhang, C., Lin, G., Wang, Q., Shen, F., Yao, Y., Tang, Z.: Guided by meta-set: a data-driven method for fine-grained visual recognition. IEEE Trans. Multimed. 25, 4691–4703 (2022)
Liu, Y., Wu, Z., Lo, S.-L., Chen, Z., Ke, G., Yue, C.: Data reweighting net for web fine-grained image classification. Multimed. Tools Appl. 2, 1–21 (2024)
Malach, E., Shalev-Shwartz, S.: Decoupling“when to update”from“how to update”. Advances in neural information processing systems 30 (2017)
Han, B., Yao, Q., Yu, X., Niu, G., Xu, M., Hu, W., Tsang, I., Sugiyama, M.: Co-teaching: robust training of deep neural networks with extremely noisy labels. Advances in neural information processing systems 31 (2018)
Shu, J., Xie, Q., Yi, L., Zhao, Q., Zhou, S., Xu, Z., Meng, D.: Meta-weight-net: learning an explicit mapping for sample weighting. In: Wallach, H.M., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E.B., Garnett, R. (eds.) Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, December 8-14, 2019, Vancouver, BC, Canada, pp. 1917–1928 (2019). https://proceedings.neurips.cc/paper/2019/hash/e58cc5ca94270acaceed13bc82dfedf7-Abstract.html
Zhang, Z., Sabuncu, M.: Generalized cross entropy loss for training deep neural networks with noisy labels. arXiv: Learning (2018)
Ren, M., Zeng, W., Yang, B., Urtasun, R.: Learning to reweight examples for robust deep learning. International Conference on Machine Learning, (2018)
Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., Lerer, A.: Automatic differentiation in pytorch (2017)
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M.S., Berg, A.C., Fei-Fei, L.: Imagenet large scale visual recognition challenge. Int. J. Comput. Vis. 115(3), 211–252 (2015). https://doi.org/10.1007/s11263-015-0816-y
Zhang, H., Cisse, M., Dauphin, Y.N., Lopez-Paz, D.: mixup: beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 (2017)
He, K., Zhang, X., Ren, S., Sun, J.: Identity mappings in deep residual networks, pp. 630–645 (2016). https://doi.org/10.1007/978-3-319-46493-0_38
Malach, E., Shalev-Shwartz, S.: Decoupling when to update from how to update. Neural Information Processing Systems, (2017)
Han, B., Yao, Q., Yu, X., Niu, G., Xu, M., Hu, W., Tsang, I., Sugiyama, M.: Co-teaching: robust training of deep neural networks with extremely noisy labels. Neural Information Processing Systems, (2018) https://doi.org/10.5555/3327757.3327944
Patrini, G., Rozza, A., Krishna Menon, A., Nock, R., Qu, L.: 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. 1944–1952 (2017)
Yue, C.: Picture fuzzy normalized projection and extended vikor approach to software reliability assessment. Appl. Soft Comput. 88, 106056 (2020). https://doi.org/10.1016/j.asoc.2019.106056
Yue, C.: A projection-based approach to software quality evaluation from the users’perspectives. Int. J. Mach. Learn. Cybern. 10, 2341–2353 (2018)
Yue, C.: Projection-based approach to group decision-making with hybrid information representations and application to software quality evaluation. Comput. Ind. Eng. 132, 98–113 (2019)
Funding
This research had no funding.
Author information
Authors and Affiliations
Contributions
Liu: Designed and wrote of full paper experiments, images and tables, text. Yue: Proposed Normalized Projection idea and Reviewed the paper. Lo: Supervisor. Reviewed the paper and proposed revision suggestions. Wu: Proposed suggestions for the writing of ablation experiments. Deng: Proposed text writing and idea suggestions.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no Conflict of interest.
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
Liu, Y., Yue, C., Lo, Sl. et al. Meta-learning with normalized projection loss reweighting for webly supervised fine-grained recognition. SIViP 19, 67 (2025). https://doi.org/10.1007/s11760-024-03591-6
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11760-024-03591-6