mlsanityData to 14 Oct 2026

Do We Really Need Gold Samples for Sample Weighting Under Label Noise?

arXiv 2104.09045 ↗First posted 19 Apr 2021cs.LG cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Learning with labels noise has gained significant traction recently due to the sensitivity of deep neural networks under label noise under common loss functions. Losses that are theoretically robust to label noise, however, often makes training difficult. Consequently, several recently proposed methods, such as Meta-Weight-Net (MW-Net), use a small number of unbiased, clean samples to learn a weighting function that downweights samples that are likely to have corrupted labels under the meta-learning framework. However, obtaining such a set of clean samples is not always feasible in practice. In this paper, we analytically show that one can easily train MW-Net without access to clean samples simply by using a loss function that is robust to label noise, such as mean absolute error, as the meta objective to train the weighting network. We experimentally show that our method beats all existing methods that do not use clean samples and performs on-par with methods that use gold samples on benchmark datasets across various noise types and noise rates.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 44 of its 70 references that are in the map. First posted on arXiv 19 Apr 2021.

Builds on

44 of 70 references.

References · 44

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  3. 2016Wide Residual Networkscited by 3,070
  4. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  5. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,059
  6. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 997
  7. 2017A Closer Look at Memorization in Deep Networkscited by 942
  8. 2017Focal Loss for Dense Object Detectioncited by 752
  9. 2018Learning to Reweight Examples for Robust Deep Learningcited by 675
  10. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 665
  11. 2017MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labelscited by 653
  12. 2016Learning to learn by gradient descent by gradient descentcited by 620
32 more
  1. 2019Symmetric Cross Entropy for Robust Learning with Noisy Labelscited by 500
  2. 2017Robust Loss Functions under Label Noise for Deep Neural Networkscited by 443
  3. 2018Bilevel Programming for Hyperparameter Optimization and Meta-Learningcited by 405
  4. 2019How does Disagreement Help Generalization against Label Corruption?cited by 390
  5. 2014Training Deep Neural Networks on Noisy Labels with Bootstrappingcited by 322
  6. 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
  7. 2020Normalized Loss Functions for Deep Learning with Noisy Labelscited by 257
  8. 2018Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noisecited by 216
  9. 2019Probabilistic End-to-end Noise Correction for Learning with Noisy Labelscited by 210
  10. 2017CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noisecited by 181
  11. 2018Dimensionality-Driven Learning with Noisy Labelscited by 173
  12. 2017Learning from Noisy Labels with Distillationcited by 161
  13. 2018Learning to Learn from Noisy Labeled Datacited by 160
  14. 2019NLNL: Negative Learning for Noisy Labelscited by 151
  15. 2017Active Bias: Training More Accurate Neural Networks by Emphasizing High Variance Samplescited by 147
  16. 2019Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Ratescited by 142
  17. 2015Learning with Symmetric Label Noise: The Importance of Being Unhingedcited by 120
  18. 2017Learning From Noisy Large-Scale Datasets With Minimal Supervisioncited by 115
  19. 2018Iterative Learning with Open-set Noisy Labelscited by 108
  20. 2019Image Classification with Deep Learning in the Presence of Noisy Labels: A Surveycited by 101
  21. 2015Webly Supervised Learning of Convolutional Networkscited by 95
  22. 2018Masking: A New Perspective of Noisy Supervisioncited by 93
  23. 2013Classification with Asymmetric Label Noise: Consistency and Maximal Denoisingcited by 68
  24. 2014Making Risk Minimization Tolerant to Label Noisecited by 54
  25. 2019On Symmetric Losses for Learning from Corrupted Labelscited by 53
  26. 2020Calibrated Surrogate Losses for Adversarially Robust Classificationcited by 36
  27. 2019L_DMI: An Information-theoretic Noise-robust Loss Functioncited by 26
  28. 2016Loss factorization, weakly supervised learning and label noise robustnesscited by 24
  29. 2019Learning Sound Event Classifiers from Web Audio with Noisy Labelscited by 22
  30. 2019Noise-Tolerant Paradigm for Training Face Recognition CNNscited by 20
  31. 2019A Novel Self-Supervised Re-labeling Approach for Training with Noisy Labelscited by 10
  32. 2013Double Ramp Loss Based Reject Option Classifiercited by 9

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 0; 2023: 1; 2024: 0; 2025: 1.

3 papers.

  1. 2021Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrectioncited by 20
  2. 2023CIFAR-10-Warehouse: Broad and More Realistic Testbeds in Model Generalization Analysiscited by 12
  3. 2025CLID-MU: Cross-Layer Information Divergence Based Meta Update Strategy for Learning with Noisy Labelscited by 0