Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
110 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Extracting Clean and Balanced Subset for Noisy Long-tailed Classification (2404.06795v1)

Published 10 Apr 2024 in cs.LG

Abstract: Real-world datasets usually are class-imbalanced and corrupted by label noise. To solve the joint issue of long-tailed distribution and label noise, most previous works usually aim to design a noise detector to distinguish the noisy and clean samples. Despite their effectiveness, they may be limited in handling the joint issue effectively in a unified way. In this work, we develop a novel pseudo labeling method using class prototypes from the perspective of distribution matching, which can be solved with optimal transport (OT). By setting a manually-specific probability measure and using a learned transport plan to pseudo-label the training samples, the proposed method can reduce the side-effects of noisy and long-tailed data simultaneously. Then we introduce a simple yet effective filter criteria by combining the observed labels and pseudo labels to obtain a more balanced and less noisy subset for a robust model training. Extensive experiments demonstrate that our method can extract this class-balanced subset with clean labels, which brings effective performance gains for long-tailed classification with label noise.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (88)
  1. Self-labelling via simultaneous clustering and representation learning. arXiv preprint arXiv:1911.05371, 2019.
  2. Asuncion, A. Uci machine learning repository, Jan 2007.
  3. The imbalanced training sample problem: Under or over sampling? In Structural, Syntactic, and Statistical Pattern Recognition, Joint IAPR International Workshops, SSPR 2004 and SPR 2004, Lisbon, Portugal, August 18-20, 2004 Proceedings, 2004.
  4. Iterative bregman projections for regularized transportation problems. SIAM Journal on Scientific Computing, 37(2):A1111–A1138, 2015.
  5. Learning imbalanced datasets with label-distribution-aware margin loss. Advances in neural information processing systems, 32, 2019.
  6. Heteroskedastic and imbalanced deep learning with adaptive regularization. arXiv preprint arXiv:2006.15766, 2020.
  7. Deep clustering for unsupervised learning of visual features. In Proceedings of the European conference on computer vision (ECCV), pp.  132–149, 2018.
  8. Unified optimal transport framework for universal domain adaptation. Advances in Neural Information Processing Systems, 35:29512–29524, 2022.
  9. SMOTE: synthetic minority over-sampling technique. J. Artif. Intell. Res., 16:321–357, 2002. doi: 10.1613/jair.953. URL https://doi.org/10.1613/jair.953.
  10. Understanding and utilizing deep neural networks trained with noisy labels. In International Conference on Machine Learning, pp.  1062–1070. PMLR, 2019.
  11. Instance-dependent label-noise learning with manifold-regularized transition matrix estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  16630–16639, 2022.
  12. Scaling algorithms for unbalanced optimal transport problems. Mathematics of Computation, 87(314):2563–2609, 2018.
  13. Class-balanced loss based on effective number of samples. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  9268–9277, 2019.
  14. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pp.  248–255. Ieee, 2009.
  15. Wasserstein adversarial regularization (war) on label noise. arXiv preprint arXiv:1904.03936, 2019.
  16. Enhancing minority classes by mixing: An adaptative optimal transport approach for long-tailed classification. In Thirty-seventh Conference on Neural Information Processing Systems, 2023.
  17. Robust loss functions under label noise for deep neural networks. In Proceedings of the AAAI conference on artificial intelligence, volume 31, 2017.
  18. Learning to re-weight examples with optimal transport for imbalanced classification. Advances in Neural Information Processing Systems, 35:25517–25530, 2022.
  19. On the power of curriculum learning in training deep networks. In International conference on machine learning, pp.  2535–2544. PMLR, 2019.
  20. Co-teaching: Robust training of deep neural networks with extremely noisy labels. Advances in neural information processing systems, 31, 2018.
  21. Sigua: Forgetting may make learning with noisy labels more robust. In International Conference on Machine Learning, pp.  4006–4016. PMLR, 2020.
  22. Deep self-learning from noisy labels. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  5138–5147, 2019.
  23. Learning from imbalanced data. IEEE Transactions on knowledge and data engineering, 21(9):1263–1284, 2009.
  24. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  770–778, 2016.
  25. Momentum contrast for unsupervised visual representation learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  9729–9738, 2020.
  26. Using trusted data to train deep networks on labels corrupted by severe noise. Advances in neural information processing systems, 31, 2018.
  27. Holt, C. C. Forecasting seasonals and trends by exponentially weighted moving averages. International journal of forecasting, 20(1):5–10, 2004.
  28. Subclass-balancing contrastive learning for long-tailed recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  5395–5407, 2023.
  29. Simple and effective regularization methods for training on noisily labeled data with generalization guarantee. arXiv preprint arXiv:1905.11368, 2019a.
  30. Learning data manipulation for augmentation and weighting. Advances in Neural Information Processing Systems, 32, 2019b.
  31. Learning deep representation for imbalanced classification. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  5375–5384, 2016.
  32. O2u-net: A simple noisy label detection approach for deep neural networks. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  3326–3334, 2019.
  33. Uncertainty-aware learning against label noise on imbalanced datasets. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pp.  6960–6969, 2022.
  34. Mentornet: Learning data-driven curriculum for very deep neural networks on corrupted labels. In International conference on machine learning, pp.  2304–2313. PMLR, 2018.
  35. Beyond synthetic noise: Deep learning on controlled noisy labels. In International conference on machine learning, pp.  4804–4815. PMLR, 2020.
  36. Decoupling representation and classifier for long-tailed recognition. arXiv preprint arXiv:1910.09217, 2019.
  37. Unicon: Combating label noise through uniform selection and contrastive learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  9676–9686, 2022.
  38. Learning multiple layers of features from tiny images. 2009.
  39. Robust inference via generative classifiers for handling noisy labels. In International conference on machine learning, pp.  3763–3772. PMLR, 2019.
  40. Cleannet: Transfer learning for scalable image classifier training with label noise. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  5447–5456, 2018.
  41. Dividemix: Learning with noisy labels as semi-supervised learning. arXiv preprint arXiv:2002.07394, 2020a.
  42. Mopro: Webly supervised learning with momentum prototypes. arXiv preprint arXiv:2009.07995, 2020b.
  43. Learning from noisy data with robust representation learning. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  9485–9494, 2021a.
  44. Class-balanced pixel-level self-labeling for domain adaptive semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  11593–11603, 2022a.
  45. Coupled-view deep classifier learning from multiple noisy annotators. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pp.  4667–4674, 2020c.
  46. Metasaug: Meta semantic augmentation for long-tailed visual recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  5212–5221, 2021b.
  47. Selective-supervised contrastive learning with noisy labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  316–325, 2022b.
  48. Webvision database: Visual learning and understanding from web data. arXiv preprint arXiv:1708.02862, 2017a.
  49. Learning from noisy labels with distillation. In Proceedings of the IEEE international conference on computer vision, pp.  1910–1918, 2017b.
  50. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, pp.  740–755. Springer, 2014.
  51. Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision, pp.  2980–2988, 2017.
  52. Early-learning regularization prevents memorization of noisy labels. Advances in neural information processing systems, 33:20331–20342, 2020.
  53. Classification with noisy labels by importance reweighting. IEEE Transactions on pattern analysis and machine intelligence, 38(3):447–461, 2015.
  54. Improving the accuracy of learning example weights for imbalance classification. In International Conference on Learning Representations, 2022.
  55. Deep learning face attributes in the wild. In Proceedings of the IEEE international conference on computer vision, pp.  3730–3738, 2015.
  56. Large-scale long-tailed recognition in an open world. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  2537–2546, 2019.
  57. Label-noise learning with intrinsically long-tailed data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  1369–1378, 2023.
  58. Long-tail learning via logit adjustment. arXiv preprint arXiv:2007.07314, 2020.
  59. Influence-balanced loss for imbalanced visual classification. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  735–744, 2021.
  60. Computational optimal transport. Center for Research in Economics and Statistics Working Papers, (2017-86), 2017.
  61. Balanced meta-softmax for long-tailed visual recognition. Advances in neural information processing systems, 33:4175–4186, 2020.
  62. Learning to reweight examples for robust deep learning. In International conference on machine learning, pp.  4334–4343. PMLR, 2018.
  63. Imagenet large scale visual recognition challenge. International journal of computer vision, 115:211–252, 2015.
  64. Meta-weight-net: Learning an explicit mapping for sample weighting. Advances in neural information processing systems, 32, 2019.
  65. Joint optimization framework for learning with noisy labels. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  5552–5560, 2018.
  66. Long-tailed recognition by routing diverse distribution-aware experts. arXiv preprint arXiv:2010.01809, 2020.
  67. Dynamic curriculum learning for imbalanced data classification. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  5017–5026, 2019.
  68. Learning to model the tail. Advances in neural information processing systems, 30, 2017.
  69. Robust long-tailed learning under label noise. arXiv preprint arXiv:2108.11569, 2021.
  70. Ngc: A unified framework for learning with open-world noisy data. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  62–71, 2021.
  71. Are anchor points really indispensable in label-noise learning? Advances in neural information processing systems, 32, 2019.
  72. Robust early-learning: Hindering the memorization of noisy labels. In International conference on learning representations, 2020a.
  73. Part-dependent label noise: Towards instance-dependent label noise. Advances in Neural Information Processing Systems, 33:7597–7610, 2020b.
  74. 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.
  75. A survey on long-tailed visual recognition. International Journal of Computer Vision, 130(7):1837–1872, 2022.
  76. Searching to exploit memorization effect in learning from corrupted labels. arXiv preprint arXiv:1911.02377, 2019.
  77. Identifying hard noise in long-tailed sample distribution. In European Conference on Computer Vision, pp.  739–756. Springer, 2022.
  78. Understanding deep learning (still) requires rethinking generalization. Communications of the ACM, 64(3):107–115, 2021a.
  79. When noisy labels meet long tail dilemmas: A representation calibration method. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  15890–15900, 2023.
  80. Prototypical pseudo label denoising and target structure learning for domain adaptive semantic segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  12414–12424, 2021b.
  81. Learning with feature-dependent label noise: A progressive approach. arXiv preprint arXiv:2103.07756, 2021c.
  82. Generalized cross entropy loss for training deep neural networks with noisy labels. Advances in neural information processing systems, 31, 2018.
  83. Contrast to divide: Self-supervised pre-training for learning with noisy labels. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp.  1657–1667, 2022.
  84. Error-bounded correction of noisy labels. In International Conference on Machine Learning, pp.  11447–11457. PMLR, 2020.
  85. Improving calibration for long-tailed recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  16489–16498, 2021.
  86. Places: A 10 million image database for scene recognition. IEEE transactions on pattern analysis and machine intelligence, 40(6):1452–1464, 2017.
  87. Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Aug 2020. doi: 10.1109/cvpr42600.2020.00974. URL http://dx.doi.org/10.1109/cvpr42600.2020.00974.
  88. Balanced contrastive learning for long-tailed visual recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  6908–6917, 2022.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (6)
  1. Zhuo Li (164 papers)
  2. He Zhao (117 papers)
  3. Zhen Li (334 papers)
  4. Tongliang Liu (251 papers)
  5. Dandan Guo (19 papers)
  6. Xiang Wan (94 papers)
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets