mlsanityData to 7 Oct 2026

Provably End-to-end Label-Noise Learning without Anchor Points

arXiv 2102.02400 ↗First posted 4 Feb 2021cs.LGcited by 97

Not on the map. Why?

Abstract (arXiv)

In label-noise learning, the transition matrix plays a key role in building statistically consistent classifiers. Existing consistent estimators for the transition matrix have been developed by exploiting anchor points. However, the anchor-point assumption is not always satisfied in real scenarios. In this paper, we propose an end-to-end framework for solving label-noise learning without anchor points, in which we simultaneously optimize two objectives: the cross entropy loss between the noisy label and the predicted probability by the neural network, and the volume of the simplex formed by the columns of the transition matrix. Our proposed framework can identify the transition matrix if the clean class-posterior probabilities are sufficiently scattered. This is by far the mildest assumption under which the transition matrix is provably identifiable and the learned classifier is statistically consistent. Experimental results on benchmark datasets demonstrate the effectiveness and robustness of the proposed method.

Key facts. Cited by 97 papers in the map (data through 7 Oct 2026). Builds on 30 of its 56 references that are in the map. First posted on arXiv 4 Feb 2021.

Builds on

30 of 56 references.

References · 30

  1. 2016Understanding deep learning requires rethinking generalizationcited by 1,657
  2. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,053
  3. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 994
  4. 2017A Closer Look at Memorization in Deep Networkscited by 936
  5. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  6. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
  7. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 663
  8. 2017MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labelscited by 652
  9. 2019How does Disagreement Help Generalization against Label Corruption?cited by 390
  10. 2014Training Deep Neural Networks on Noisy Labels with Bootstrappingcited by 322
  11. 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
  12. 2017WebVision Database: Visual Learning and Understanding from Web Datacited by 262
18 more
  1. 2017Decoupling "when to update" from "how to update"cited by 230
  2. 2019Are Anchor Points Really Indispensable in Label-Noise Learning?cited by 208
  3. 2018Dimensionality-Driven Learning with Noisy Labelscited by 173
  4. 2019SELF: Learning to Filter Noisy Labels with Self-Ensemblingcited by 164
  5. 2019Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Ratescited by 142
  6. 2020Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learningcited by 142
  7. 2020A Survey of Label-noise Representation Learning: Past, Present and Futurecited by 121
  8. 2017Learning with Biased Complementary Labelscited by 93
  9. 2018Masking: A New Perspective of Noisy Supervisioncited by 93
  10. 2020A Second-Order Approach to Learning with Instance-Dependent Label Noisecited by 90
  11. 2020Confidence Scores Make Instance-dependent Label-noise Learning Possiblecited by 63
  12. 2019Learning with Multiple Complementary Labelscited by 53
  13. 2017Learning with Confident Examples: Rank Pruning for Robust Classification with Noisy Labelscited by 52
  14. 2017On Identifiability of Nonnegative Matrix Factorizationcited by 34
  15. 2020Unbiased Risk Estimators Can Mislead: A Case Study of Learning with Complementary Labelscited by 29
  16. 2016Robust Volume Minimization-Based Matrix Factorization for Remote Sensing and Document Clusteringcited by 28
  17. 2018Robustness of Conditional GANs to Noisy Labelscited by 27
  18. 2019Generalization Bounds for Neural Networks via Approximate Description Lengthcited by 15

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 10; 2022: 18; 2023: 20; 2024: 14; 2025: 21; 2026: 14.

97 papers.

Sort:

By area:

  1. 2021Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotationscited by 253
  2. 2022Robust Training under Label Noise by Over-parameterizationcited by 93
  3. 2021Sample Selection with Uncertainty of Losses for Learning with Noisy Labelscited by 79
  4. 2021Clusterability as an Alternative to Anchor Points When Learning with Noisy Labelscited by 65
  5. 2021Detecting Corrupted Labels Without Training a Model to Predictcited by 57
  6. 2021Instance-dependent Label-noise Learning under a Structural Causal Modelcited by 50
  7. 2022Instance-Dependent Label-Noise Learning with Manifold-Regularized Transition Matrix Estimationcited by 49
  8. 2022ProMix: Combating Label Noise via Maximizing Clean Sample Utilitycited by 48
  9. 2025Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifierscited by 40
  10. 2022Identifiability of Label Noise Transition Matrixcited by 34
  11. 2021Estimating Instance-dependent Bayes-label Transition Matrix using a Deep Neural Networkcited by 30
  12. 2022Beyond Images: Label Noise Transition Matrix Estimation for Tasks with Lower-Quality Featurescited by 27
38 more
  1. 2023Understanding and Mitigating the Label Noise in Pre-training on Downstream Taskscited by 27
  2. 2021Understanding Instance-Level Label Noise: Disparate Impacts and Treatmentscited by 22
  3. 2021Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrectioncited by 20
  4. 2022Towards Robust Adaptive Object Detection under Noisy Annotationscited by 20
  5. 2022SimT: Handling Open-set Noise for Domain Adaptive Semantic Segmentationcited by 17
  6. 2024Label-Noise Robust Diffusion Modelscited by 16
  7. 2023InstanT: Semi-supervised Learning with Instance-dependent Thresholdscited by 13
  8. 2023Deep Learning From Crowdsourced Labels: Coupled Cross-entropy Minimization, Identifiability, and Regularizationcited by 12
  9. 2023Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memorycited by 11
  10. 2023DyGen: Learning from Noisy Labels via Dynamics-Enhanced Generative Modelingcited by 10
  11. 2023Latent Class-Conditional Noise Modelcited by 9
  12. 2024A Conformal Prediction Score that is Robust to Label Noisecited by 9
  13. 2024Slight Corruption in Pre-training Data Makes Better Diffusion Modelscited by 9
  14. 2022Joint Class-Affinity Loss Correction for Robust Medical Image Segmentation with Noisy Labelscited by 8
  15. 2023Regularly Truncated M-estimators for Learning with Noisy Labelscited by 8
  16. 2023Regroup Median Loss for Combating Label Noisecited by 8
  17. 2024Impact of Noisy Supervision in Foundation Model Learningcited by 8
  18. 2021Instance Correction for Learning with Open-set Noisy Labelscited by 7
  19. 2022Reduction from Complementary-Label Learning to Probability Estimatescited by 7
  20. 2024Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learningcited by 7
  21. 2022Distributional Reward Estimation for Effective Multi-Agent Deep Reinforcement Learningcited by 6
  22. 2025ε-Softmax: Approximating One-Hot Vectors for Mitigating Label Noisecited by 6
  23. 2022ProSelfLC: Progressive Self Label Correction Towards A Low-Temperature Entropy Statecited by 5
  24. 2022SplitNet: Learnable Clean-Noisy Label Splitting for Learning with Noisy Labelscited by 5
  25. 2023Deep Clustering with Incomplete Noisy Pairwise Annotations: A Geometric Regularization Approachcited by 5
  26. 2024De-Confusing Pseudo-Labels in Source-Free Domain Adaptationcited by 5
  27. 2023DAC-MR: Data Augmentation Consistency Based Meta-Regularization for Meta-Learningcited by 4
  28. 2023Rectifying Noisy Labels with Sequential Prior: Multi-Scale Temporal Feature Affinity Learning for Robust Video Segmentationcited by 4
  29. 2024Learning under Temporal Label Noisecited by 4
  30. 2024Dirichlet-based Per-Sample Weighting by Transition Matrix for Noisy Label Learningcited by 4
  31. 2024From Biased Selective Labels to Pseudo-Labels: An Expectation-Maximization Framework for Learning from Biased Decisionscited by 4
  32. 2025Imbalanced Medical Image Segmentation with Pixel-dependent Noisy Labelscited by 4
  33. 2025Conformal Prediction of Classifiers with Many Classes based on Noisy Labelscited by 4
  34. 2025Exploring the Noise Robustness of Online Conformal Predictioncited by 4
  35. 2022Multi-class Label Noise Learning via Loss Decomposition and Centroid Estimationcited by 3
  36. 2024Multiclass Learning from Noisy Labels for Non-decomposable Performance Measurescited by 3
  37. 2024Checking the Sufficiently Scattered Condition using a Global Non-Convex Optimization Softwarecited by 3
  38. 2025Benchmarking Real-World Medical Image Classification with Noisy Labels: Challenges, Practice, and Outlookcited by 3

50 most cited of 97 All 97 (JSON)