Abstract
The long-tailed data distribution in real-world greatly increases the difficulty of training deep neural networks. Oversampling minority classes is one of the commonly used techniques to tackle this problem. In this paper, we first analyze that the commonly used oversampling technique tends to distort the representation learning and harm the network’s generalizability. Then we propose two novel methods to increase the minority feature’s diversity to alleviate such issue. Specifically, from the data perspective, we propose a mixup-based Synthetic Minority Over-sampling TEchnique called mixSMOTE, where tail class samples are synthesized from head classes so that a balanced training distribution can be obtained. Then from the model perspective, we propose Gradient Re-weighting Module (GRM) to re-distribute each instance’s gradient contribution to the representation learning network. Extensive experiments on the long-tailed benchmark CIFAR10-LT, CIFAR100-LT and ImageNet-LT demonstrate the effectiveness of our proposed method.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Buda, M., Maki, A., Mazurowski, M.A.: A systematic study of the class imbalance problem in convolutional neural networks. Neural Netw. 106, 249–259 (2018)
Cao, K., Wei, C., Gaidon, A., Arechiga, N., Ma, T.: Learning imbalanced datasets with label-distribution-aware margin loss. In: Advances in Neural Information Processing Systems, pp. 1565–1576 (2019)
Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P.: SMOTE: synthetic minority over-sampling technique. J. Artif. Intell. Res. 16, 321–357 (2002)
Chawla, N.V., Japkowicz, N., Kotcz, A.: Special issue on learning from imbalanced data sets. ACM SIGKDD Explor. Newsl. 6(1), 1–6 (2004)
Cohen, G., Hilario, M., Sax, H., Hugonnet, S., Geissbuhler, A.: Learning from imbalanced data in surveillance of nosocomial infection. Artif. Intell. Med. 37(1), 7–18 (2006)
Cui, Y., Jia, M., Lin, T.Y., Song, Y., Belongie, S.: Class-balanced loss based on effective number of samples. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9268–9277 (2019)
Drummond, C., Holte, R.C., et al.: C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling. In: Workshop on Learning from Imbalanced Datasets II, vol. 11, pp. 1–8. Citeseer (2003)
Gidaris, S., Komodakis, N.: Dynamic few-shot visual learning without forgetting. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4367–4375 (2018)
Gupta, A., Dollar, P., Girshick, R.: LVIS: a dataset for large vocabulary instance segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5356–5364 (2019)
Haixiang, G., Yijing, L., Jennifer Shang, G., Mingyun, H.Y., Bing, G.: Learning from class-imbalanced data: review of methods and applications. Expert Syst. Appl. 73, 220–239 (2017)
He, H., Garcia, E.A.: Learning from imbalanced data. IEEE Trans. Knowl. Data Eng. 21(9), 1263–1284 (2009)
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)
Jamal, M.A., Brown, M., Yang, M.H., Wang, L., Gong, B.: Rethinking class-balanced methods for long-tailed visual recognition from a domain adaptation perspective. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7610–7619 (2020)
Kang, B., et al.: Decoupling representation and classifier for long-tailed recognition. arXiv preprint arXiv:1910.09217 (2019)
Kim, J., Jeong, J., Shin, J.: M2m: imbalanced classification via major-to-minor translation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13896–13905 (2020)
Krawczyk, B.: Learning from imbalanced data: open challenges and future directions. Prog. Artif. Intell. 5(4), 221–232 (2016). https://doi.org/10.1007/s13748-016-0094-0
Lin, T.-Y., Goyal, P., Girshick, R., He, K., Dollár, P.: Focal loss for dense object detection. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2980–2988 (2017)
Liu, Z., Miao, Z., Zhan, X., Wang, J., Gong, B., Yu, S.X.: Large-scale long-tailed recognition in an open world. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2537–2546 (2019)
Oh Song, H., Xiang, Y., Jegelka, S., Savarese, S.: Deep metric learning via lifted structured feature embedding. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4004–4012 (2016)
Ouyang, W., Wang, X., Zhang, C., Yang, X.: Factors in finetuning deep model for object detection with long-tail distribution. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 864–873 (2016)
Paszke, A., et al.: Automatic differentiation in pytorch (2017)
Peng, M., et al.: Trainable undersampling for class-imbalance learning. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp. 4707–4714 (2019)
Ren, M., Zeng, W., Yang, B., Urtasun, R.: Learning to reweight examples for robust deep learning. arXiv preprint arXiv:1803.09050 (2018)
Shu, J., et al.: Meta-weight-net: learning an explicit mapping for sample weighting. In: Advances in Neural Information Processing Systems, pp. 1919–1930 (2019)
Wang, Y.-X., Ramanan, D., Hebert, M.: Learning to model the tail. In: Advances in Neural Information Processing Systems, pp. 7029–7039 (2017)
Xiang, L., Ding, G.: Learning from multiple experts: Self-paced knowledge distillation for long-tailed classification. arXiv preprint arXiv:2001.01536 (2020)
Xiang, L., Jin, X., Ding, G., Han, J., Li, L.: Incremental few-shot learning for pedestrian attribute recognition. In: IJCAI (2019)
Yan, Y., et al.: Oversampling for imbalanced data via optimal transport. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp. 5605–5612 (2019)
Zhang, H., Cisse, M., Dauphin, Y.N., Lopez-Paz, D.: mixup: beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 (2017)
Zhang, X., Fang, Z., Wen, Y., Li, Z., Qiao, Y.: Range loss for deep face recognition with long-tailed training data. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 5409–5418 (2017)
Acknowledgement
This work was supported by the National Natural Science Foundation of China (No. U1936202, 61925107).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Xiang, L., Ding, G., Han, J. (2021). Increasing Oversampling Diversity for Long-Tailed Visual Recognition. In: Fang, L., Chen, Y., Zhai, G., Wang, J., Wang, R., Dong, W. (eds) Artificial Intelligence. CICAI 2021. Lecture Notes in Computer Science(), vol 13069. Springer, Cham. https://doi.org/10.1007/978-3-030-93046-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-93046-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93045-5
Online ISBN: 978-3-030-93046-2
eBook Packages: Computer ScienceComputer Science (R0)