Abstract
Image classification has lead to the revolution of artificial intelligence in the past five years. However, image classification algorithms are significantly affected by the inherent variance in sensory input and noise in the labelled data in a real-world situation. Also, the class count of items in the real world is significantly larger than that of the typical experiment setup. How to speedily train/fine-tune a large scale model in giant label space with considerable noise is a recent interest of the machine learning community. This paper proposed a multi-stage training algorithm to fine-tune a pre-trained EfficientNet model on AliProducts large scale product classification dataset, which has a large label space (50030 classes) and severe label noise. Our method can generalize well on such a dataset while keeping the prior knowledge gained in the large-scale pretraining stage. With our novel Dynamically Induced Cross-Entropy(DICE) network loss and several other methods to tackle unbalanced datasets and improve model convergence, the model achieved 76.67% of Top-1 Accuracy and 85.42% of Top-5 accuracy, which are 3.63%/2.9% higher than symmetric cross-entropy and are significantly higher than the usual fine-tuning method with categorical cross-entropy loss(CCE).
This material is based upon work supported by TTAD-2021-03.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Angluin, D., Laird, P.: Learning from noisy examples. Mach. Learn. 2(4), 343–370 (1988)
Berthon, A., Han, B., Niu, G., Liu, T., Sugiyama, M.: Confidence scores make instance-dependent label-noise learning possible. In: International Conference on Machine Learning, pp. 825–836. PMLR (2021)
Feng, L., Shu, S., Lin, Z., Lv, F., Li, L., An, B.: Can cross entropy loss be robust to label noise? In: IJCAI, pp. 2206–2212 (2020)
Gao, B.B., Xing, C., Xie, C.W., Wu, J., Geng, X.: Deep label distribution learning with label ambiguity. IEEE Trans. Image Process. 26(6), 2825–2838 (2017)
Ghosh, A., Kumar, H., Sastry, P.: Robust loss functions under label noise for deep neural networks. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 31 (2017)
Ghosh, A., Manwani, N., Sastry, P.: Making risk minimization tolerant to label noise. Neurocomputing 160, 93–107 (2015)
Goldberger, J., Ben-Reuven, E.: Training deep neural-networks using a noise adaptation layer (2016)
Han, B., et al.: Co-teaching: robust training of deep neural networks with extremely noisy labels. arXiv preprint arXiv:1804.06872 (2018)
Jiang, L., Zhou, Z., Leung, T., Li, L.J., Fei-Fei, L.: MentorNet: learning data-driven curriculum for very deep neural networks on corrupted labels. In: International Conference on Machine Learning, pp. 2304–2313. PMLR (2018)
Menon, A.K., Rawat, A.S., Reddi, S.J., Kumar, S.: Can gradient clipping mitigate label noise? In: International Conference on Learning Representations (2019)
Mikolov, T., Chen, K., Corrado, G., Dean, J.: Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781 (2013)
Natarajan, N., Dhillon, I.S., Ravikumar, P.K., Tewari, A.: Learning with noisy labels. Adv. Neural. Inf. Process. Syst. 26, 1196–1204 (2013)
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)
Sukhbaatar, S., Fergus, R.: Learning from noisy labels with deep neural networks. arXiv preprint arXiv:1406.2080 (2014). vol. 2(3), p. 4
Sun, C., Shrivastava, A., Singh, S., Gupta, A.: Revisiting unreasonable effectiveness of data in deep learning era. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 843–852 (2017)
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2818–2826 (2016)
Wang, Y., Ma, X., Chen, Z., Luo, Y., Yi, J., Bailey, J.: Symmetric cross entropy for robust learning with noisy labels. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 322–330 (2019)
Wei, H., Feng, L., Chen, X., An, B.: 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. 13726–13735 (2020)
Xia, X., et al.: Are anchor points really indispensable in label-noise learning? Adv. Neural. Inf. Process. Syst. 32, 6838–6849 (2019)
Xiao, T., Xia, T., Yang, Y., Huang, C., Wang, X.: Learning from massive noisy labeled data for image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2691–2699 (2015)
Yao, J., et al.: Deep learning from noisy image labels with quality embedding. IEEE Trans. Image Process. 28(4), 1909–1922 (2018)
Yao, Q., Yang, H., Han, B., Niu, G., Kwok, J.: Searching to exploit memorization effect in learning from corrupted labels. arXiv preprint arXiv:1911.02377 (2019)
Yi, K., Wu, J.: Probabilistic end-to-end noise correction for learning with noisy labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7017–7025 (2019)
Zhang, Z., Sabuncu, M.R.: Generalized cross entropy loss for training deep neural networks with noisy labels. In: 32nd Conference on Neural Information Processing Systems (NeurIPS) (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Liu, T. (2022). DICE: Dynamically Induced Cross Entropy for Robust Learning with Noisy Labels. In: Sun, F., Hu, D., Wermter, S., Yang, L., Liu, H., Fang, B. (eds) Cognitive Systems and Information Processing. ICCSIP 2021. Communications in Computer and Information Science, vol 1515. Springer, Singapore. https://doi.org/10.1007/978-981-16-9247-5_9
Download citation
DOI: https://doi.org/10.1007/978-981-16-9247-5_9
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-9246-8
Online ISBN: 978-981-16-9247-5
eBook Packages: Computer ScienceComputer Science (R0)