mlsanityData to 7 Oct 2026

NCSAM Noise-Compensated Sharpness-Aware Minimization for Noisy Label Learning

arXiv 2601.19947 ↗First posted 24 Jan 2026cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Learning from Noisy Labels (LNL) remains a fundamental challenge in deep learning because real-world datasets often contain corrupted annotations. Most existing methods rely on label correction or sample selection mechanisms. In contrast, we study LNL from an optimization perspective by establishing a theoretical connection between label noise and the flatness-seeking behavior of Sharpness-Aware Minimization (SAM). Based on this analysis, we propose Noise-Compensated Sharpness-Aware Minimization (NCSAM), which uses a noise-compensated perturbation to counteract the optimization bias induced by noisy labels. By correcting distorted SAM perturbations, NCSAM mitigates the memorization of noisy labels during training while preserving the simplicity of optimization-based learning. Experiments on synthetic and real-world noisy-label benchmarks show that NCSAM consistently improves over SAM-based optimization baselines and remains competitive with representative noisy-label learning methods.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 42 of its 62 references that are in the map. First posted on arXiv 24 Jan 2026.

Builds on

42 of 62 references.

References · 42

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  4. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  5. 2020Sharpness-Aware Minimization for Efficiently Improving Generalizationcited by 1,404
  6. 2016On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minimacited by 1,337
  7. 2017Visualizing the Loss Landscape of Neural Netscited by 1,126
  8. 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,054
  9. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,053
  10. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 994
  11. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  12. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
30 more
  1. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 663
  2. 2017Exploring Generalization in Deep Learningcited by 595
  3. 2017Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Datacited by 527
  4. 2019Symmetric Cross Entropy for Robust Learning with Noisy Labelscited by 499
  5. 2017Sharp Minima Can Generalize For Deep Netscited by 462
  6. 2020Early-Learning Regularization Prevents Memorization of Noisy Labelscited by 414
  7. 2019How does Disagreement Help Generalization against Label Corruption?cited by 390
  8. 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
  9. 2020Combating noisy labels by agreement: A joint training method with co-regularizationcited by 289
  10. 2019Understanding and Utilizing Deep Neural Networks Trained with Noisy Labelscited by 209
  11. 2019Are Anchor Points Really Indispensable in Label-Noise Learning?cited by 208
  12. 2020Learning with Instance-Dependent Label Noise: A Sample Sieve Approachcited by 158
  13. 2019Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Ratescited by 142
  14. 2019Meta Label Correction for Noisy Label Learningcited by 108
  15. 2021Learning with Feature-Dependent Label Noise: A Progressive Approachcited by 102
  16. 2019Asymmetric Valleys: Beyond Sharp and Flat Local Minimacited by 90
  17. 2020A Second-Order Approach to Learning with Instance-Dependent Label Noisecited by 90
  18. 2022A Closer Look at Smoothness in Domain Adversarial Trainingcited by 85
  19. 2020Beyond Class-Conditional Assumption: A Primary Attempt to Combat Instance-Dependent Label Noisecited by 82
  20. 2022UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learningcited by 77
  21. 2020Error-Bounded Correction of Noisy Labelscited by 74
  22. 2021Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labelscited by 74
  23. 2013Classification with Asymmetric Label Noise: Consistency and Maximal Denoisingcited by 67
  24. 2022Escaping Saddle Points for Effective Generalization on Class-Imbalanced Datacited by 28
  25. 2016Loss factorization, weakly supervised learning and label noise robustnesscited by 24
  26. 2022Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labelscited by 16
  27. 2020Noisy Concurrent Training for Efficient Learning under Label Noisecited by 13
  28. 2023Unleashing the Potential of Regularization Strategies in Learning with Noisy Labelscited by 5
  29. 2024Bilateral Sharpness-Aware Minimization for Flatter Minimacited by 2
  30. 2025Balanced Sharpness-Aware Minimization for Imbalanced Regressioncited by 1

Built on by

Not cited yet.