Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A Survey of Deep Long-Tail Classification Advancements (2404.15593v1)

Published 24 Apr 2024 in cs.LG

Abstract: Many data distributions in the real world are hardly uniform. Instead, skewed and long-tailed distributions of various kinds are commonly observed. This poses an interesting problem for machine learning, where most algorithms assume or work well with uniformly distributed data. The problem is further exacerbated by current state-of-the-art deep learning models requiring large volumes of training data. As such, learning from imbalanced data remains a challenging research problem and a problem that must be solved as we move towards more real-world applications of deep learning. In the context of class imbalance, state-of-the-art (SOTA) accuracies on standard benchmark datasets for classification typically fall less than 75%, even for less challenging datasets such as CIFAR100. Nonetheless, there has been progress in this niche area of deep learning. To this end, in this survey, we provide a taxonomy of various methods proposed for addressing the problem of long-tail classification, focusing on works that happened in the last few years under a single mathematical framework. We also discuss standard performance metrics, convergence studies, feature distribution and classifier analysis. We also provide a quantitative comparison of the performance of different SOTA methods and conclude the survey by discussing the remaining challenges and future research direction.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (89)
  1. Long-tailed recognition via weight balancing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 6897–6907.
  2. Long-Tail Learning with Rebalanced Contrastive Loss. arXiv:2312.01753 [cs.LG]
  3. Sercan Ö Arik and Tomas Pfister. 2021. Tabnet: Attentive interpretable tabular learning. In Proceedings of the AAAI conference on artificial intelligence, Vol. 35. 6679–6687.
  4. Spectrally-normalized margin bounds for neural networks. Advances in neural information processing systems 30 (2017).
  5. Siam-IDS: Handling class imbalance problem in intrusion detection systems using siamese neural network. Procedia Computer Science 171 (2020), 780–789.
  6. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258 (2021).
  7. Language models are few-shot learners. Advances in neural information processing systems 33 (2020), 1877–1901.
  8. A systematic study of the class imbalance problem in convolutional neural networks. Neural networks 106 (2018), 249–259.
  9. Jonathon Byrd and Zachary Lipton. 2019. What is the effect of importance weighting in deep learning?. In International conference on machine learning. PMLR, 872–881.
  10. Heteroskedastic and imbalanced deep learning with adaptive regularization. arXiv preprint arXiv:2006.15766 (2020).
  11. Learning imbalanced datasets with label-distribution-aware margin loss. Advances in neural information processing systems 32 (2019).
  12. SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research (2002).
  13. Revisiting Zero-Shot Abstractive Summarization in the Era of Large Language Models from the Perspective of Position Bias. arXiv preprint arXiv:2401.01989 (2024).
  14. Reviving threshold-moving: a simple plug-in bagging ensemble for binary and multiclass imbalanced data. arXiv preprint arXiv:1606.08698 (2016).
  15. Class-balanced loss based on effective number of samples. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 9268–9277.
  16. Long-Tail Learning with Rebalanced Contrastive Loss. arXiv preprint arXiv:2312.01753 (2023).
  17. Pml: Progressive margin loss for long-tailed age classification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 10503–10512.
  18. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).
  19. Improving Electric Fraud Detection using Class Imbalance Strategies.. In ICPRAM (2). 135–141.
  20. Improving pedestrian detection from a long-tailed domain perspective. In Proceedings of the 29th ACM International Conference on Multimedia. 2918–2926.
  21. Yingxiao Du and Jianxin Wu. 2023. No One Left Behind: Improving the Worst Categories in Long-Tailed Learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 15804–15813.
  22. A multiple resampling method for learning from imbalanced data sets. Computational intelligence 20, 1 (2004), 18–36.
  23. Learning with average top-k loss. Advances in neural information processing systems 30 (2017).
  24. Revisiting Long-tailed Image Classification: Survey and Benchmarks with New Evaluation Metrics. arXiv preprint arXiv:2302.01507 (2023).
  25. Long-tailed visual recognition with deep models: A methodological survey and evaluation. Neurocomputing (2022).
  26. An investigation into neural net optimization via hessian eigenvalue density. In International Conference on Machine Learning. PMLR, 2232–2241.
  27. On calibration of modern neural networks. In International conference on machine learning. PMLR, 1321–1330.
  28. Lvis: A dataset for large vocabulary instance segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 5356–5364.
  29. Boran Han. 2023. Wrapped Cauchy Distributed Angular Softmax for Long-Tailed Visual Recognition. arXiv preprint arXiv:2305.18732 (2023).
  30. Haibo He and Edwardo A Garcia. 2009. Learning from imbalanced data. IEEE Transactions on knowledge and data engineering 21, 9 (2009), 1263–1284.
  31. Disentangling label distribution for long-tailed visual recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 6626–6636.
  32. Deep imbalanced learning for face recognition and attribute prediction. IEEE transactions on pattern analysis and machine intelligence 42, 11 (2019), 2781–2794.
  33. iNaturalist 2018 competition dataset 2018. iNaturalist 2018 competition dataset.  https://github.com/visipedia/inat_comp/tree/master/2018.
  34. Cost-sensitive support vector machines. Neurocomputing 343 (2019), 50–64.
  35. Self-damaging contrastive learning. In International Conference on Machine Learning. PMLR, 4927–4939.
  36. ELM: Embedding and Logit Margins for Long-Tail Learning. arXiv preprint arXiv:2204.13208 (2022).
  37. Justin M Johnson and Taghi M Khoshgoftaar. 2019. Survey on deep learning with class imbalance. Journal of Big Data 6, 1 (2019), 1–54.
  38. Exploring balanced feature spaces for representation learning. In International Conference on Learning Representations.
  39. Decoupling representation and classifier for long-tailed recognition. arXiv preprint arXiv:1910.09217 (2019).
  40. Byungju Kim and Junmo Kim. 2020. Adjusting decision boundary for class imbalanced learning. IEEE Access 8 (2020), 81674–81685.
  41. Label-imbalanced and group-sensitive classification under overparameterization. Advances in Neural Information Processing Systems 34 (2021), 18970–18983.
  42. Bartosz Krawczyk. 2016. Learning from imbalanced data: open challenges and future directions. Progress in Artificial Intelligence 5, 4 (2016), 221–232.
  43. A survey on addressing high-class imbalance in big data. Journal of Big Data 5, 1 (2018), 1–30.
  44. Gradient harmonized single-stage detector. In Proceedings of the AAAI conference on artificial intelligence.
  45. FCC: Feature Clusters Compression for Long-Tailed Visual Recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 24080–24089.
  46. Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision. 2980–2988.
  47. Charles X Ling and Victor S Sheng. 2008. Cost-sensitive learning and the class imbalance problem. Encyclopedia of machine learning 2011 (2008).
  48. Self-supervised learning is more robust to dataset imbalance. arXiv preprint arXiv:2110.05025 (2021).
  49. Addressing the class imbalance problem in twitter spam detection using ensemble learning. Computers & Security 69 (2017), 35–49.
  50. Sphereface: Deep hypersphere embedding for face recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 212–220.
  51. Exploratory undersampling for class-imbalance learning. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics) 39, 2 (2008), 539–550.
  52. Large-scale long-tailed recognition in an open world. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2537–2546.
  53. Dhruv Mahajan et al. 2018. Exploring the limits of weakly supervised pretraining. In Proceedings of the European conference on computer vision.
  54. David Masko and Paulina Hensman. 2015. The impact of imbalanced training data for convolutional neural networks.
  55. On the statistical consistency of algorithms for binary classification under class imbalance. In International Conference on Machine Learning. PMLR, 603–611.
  56. Long-tail learning via logit adjustment. arXiv preprint arXiv:2007.07314 (2020).
  57. Supplementary-I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification. ([n. d.]).
  58. Optimal transport for long-tailed recognition with learnable cost matrix. In International Conference on Learning Representations.
  59. Luis Perez and Jason Wang. 2017. The effectiveness of data augmentation in image classification using deep learning. preprint arXiv:1712.04621 (2017).
  60. John Platt et al. 1999. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in large margin classifiers 10, 3 (1999), 61–74.
  61. Learning transferable visual models from natural language supervision. In International conference on machine learning. PMLR.
  62. Escaping Saddle Points for Effective Generalization on Class-Imbalanced Data. arXiv preprint arXiv:2212.13827 (2022).
  63. Balanced meta-softmax for long-tailed visual recognition. Advances in neural information processing systems 33 (2020), 4175–4186.
  64. Learning to reweight examples for robust deep learning. In International conference on machine learning. PMLR, 4334–4343.
  65. Nils Rethmeier and Isabelle Augenstein. 2020. Self-supervised contrastive zero to few-shot learning from small, long-tailed text data. (2020).
  66. Nils Rethmeier and Isabelle Augenstein. 2022. Long-tail zero and few-shot learning via contrastive pretraining on and for small data. In Computer Sciences & Mathematics Forum, Vol. 3. MDPI, 10.
  67. From generalized zero-shot learning to long-tail with class descriptors. In Proceedings of the IEEE/CVF winter conference on applications of computer vision. 286–295.
  68. Connor Shorten and Taghi M Khoshgoftaar. 2019. A survey on image data augmentation for deep learning. Journal of big data 6, 1 (2019), 1–48.
  69. Meta-weight-net: Learning an explicit mapping for sample weighting. Advances in neural information processing systems 32 (2019).
  70. Equalization loss for long-tailed object recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 11662–11671.
  71. Experimental perspectives on learning from imbalanced data. In Proceedings of the 24th international conference on Machine learning. 935–942.
  72. Controlling the sensitivity of support vector machines. In Proceedings of the international joint conference on AI, Vol. 55. Stockholm, 60.
  73. Additive margin softmax for face verification. IEEE Signal Processing Letters 25, 7 (2018).
  74. Seesaw loss for long-tailed instance segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 9695–9704.
  75. Long-tailed recognition by routing diverse distribution-aware experts. arXiv preprint arXiv:2010.01809 (2020).
  76. Dynamic curriculum learning for imbalanced data classification. In Proceedings of the IEEE/CVF international conference on computer vision. 5017–5026.
  77. Learning to model the tail. Advances in neural information processing systems 30 (2017).
  78. On the margin theory of feedforward neural networks. (2018).
  79. Robust long-tailed learning under label noise. arXiv preprint arXiv:2108.11569 (2021).
  80. A survey on long-tailed visual recognition. International Journal of Computer Vision 130, 7 (2022).
  81. Identifying and compensating for feature deviation in imbalanced deep learning. arXiv preprint arXiv:2001.01385 (2020).
  82. Distribution alignment: A unified framework for long-tail visual recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2361–2370.
  83. Self-supervised aggregation of diverse experts for test-agnostic long-tailed recognition. arXiv preprint arXiv:2107.09249 (2021).
  84. Deep long-tailed learning: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023).
  85. Zizhao Zhang and Tomas Pfister. 2021. Learning fast sample re-weighting without reward data. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 725–734.
  86. Unequal-training for deep face recognition with long-tailed noisy data. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 7812–7821.
  87. Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 9719–9728.
  88. Prompt-aligned gradient for prompt tuning. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 15659–15669.
  89. Balanced contrastive learning for long-tailed visual recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 6908–6917.
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