mlsanityData to 7 Oct 2026

Instance-dependent Label-noise Learning under a Structural Causal Model

arXiv 2109.02986 ↗First posted 7 Sep 2021stat.ML cs.LGcited by 50

Not on the map. Why?

Abstract (arXiv)

Label noise will degenerate the performance of deep learning algorithms because deep neural networks easily overfit label errors. Let X and Y denote the instance and clean label, respectively. When Y is a cause of X, according to which many datasets have been constructed, e.g., SVHN and CIFAR, the distributions of P(X) and P(Y|X) are entangled. This means that the unsupervised instances are helpful to learn the classifier and thus reduce the side effect of label noise. However, it remains elusive on how to exploit the causal information to handle the label noise problem. In this paper, by leveraging a structural causal model, we propose a novel generative approach for instance-dependent label-noise learning. In particular, we show that properly modeling the instances will contribute to the identifiability of the label noise transition matrix and thus lead to a better classifier. Empirically, our method outperforms all state-of-the-art methods on both synthetic and real-world label-noise datasets.

Key facts. Cited by 50 papers in the map (data through 7 Oct 2026). Builds on 22 of its 41 references that are in the map. First posted on arXiv 7 Sep 2021.

Builds on

22 of 41 references.

References · 22

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  2. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  3. 2013Auto-Encoding Variational Bayescited by 3,980
  4. 2016Variational Inference: A Review for Statisticianscited by 1,466
  5. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,053
  6. 2017A Closer Look at Memorization in Deep Networkscited by 936
  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. 2019Causality for Machine Learningcited by 242
  10. 2017Decoupling "when to update" from "how to update"cited by 230
  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
10 more
  1. 2020Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learningcited by 142
  2. 2021Provably End-to-end Label-Noise Learning without Anchor Pointscited by 97
  3. 2017Learning with Biased Complementary Labelscited by 93
  4. 2020A Second-Order Approach to Learning with Instance-Dependent Label Noisecited by 90
  5. 2021Clusterability as an Alternative to Anchor Points When Learning with Noisy Labelscited by 65
  6. 2020Confidence Scores Make Instance-dependent Label-noise Learning Possiblecited by 63
  7. 2017Learning with Bounded Instance- and Label-dependent Label Noisecited by 57
  8. 2020Part-dependent Label Noise: Towards Instance-dependent Label Noisecited by 40
  9. 2020Class2Simi: A Noise Reduction Perspective on Learning with Noisy Labelscited by 25
  10. 2015Distinguishing Cause from Effect Based on Exogeneitycited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 10; 2023: 17; 2024: 10; 2025: 6; 2026: 6.

50 papers.

Sort:

By area:

  1. 2022Harnessing Out-Of-Distribution Examples via Augmenting Content and Stylecited by 43
  2. 2022Identifiability of Label Noise Transition Matrixcited by 34
  3. 2023Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidancecited by 24
  4. 2022From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Modelcited by 21
  5. 2021Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrectioncited by 20
  6. 2023Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labelscited by 20
  7. 2022Instance-Dependent Noisy Label Learning via Graphical Modellingcited by 17
  8. 2023Winning Prize Comes from Losing Tickets: Improve Invariant Learning by Exploring Variant Parameters for Out-of-Distribution Generalizationcited by 12
  9. 2023Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowdscited by 11
  10. 2022PADDLES: Phase-Amplitude Spectrum Disentangled Early Stopping for Learning with Noisy Labelscited by 10
  11. 2023DyGen: Learning from Noisy Labels via Dynamics-Enhanced Generative Modelingcited by 10
  12. 2023Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurationscited by 9
38 more
  1. 2025Jailbreaking the Non-Transferable Barrier via Test-Time Data Disguisingcited by 8
  2. 2024Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learningcited by 7
  3. 2025Toward Robust Non-Transferable Learning: A Survey and Benchmarkcited by 7
  4. 2022Robust Fine-tuning via Perturbation and Interpolation from In-batch Instancescited by 6
  5. 2022Label Noise-Robust Learning using a Confidence-Based Sieving Strategycited by 6
  6. 2023Combating Label Noise With A General Surrogate Model For Sample Selectioncited by 6
  7. 2024Mitigating Label Noise on Graph via Topological Sample Selectioncited by 6
  8. 2023Unleashing the Potential of Regularization Strategies in Learning with Noisy Labelscited by 5
  9. 2023Instance-dependent Noisy-label Learning with Graphical Model Based Noise-rate Estimationcited by 4
  10. 2023PNT-Edge: Towards Robust Edge Detection with Noisy Labels by Learning Pixel-level Noise Transitionscited by 4
  11. 2024Dirichlet-based Per-Sample Weighting by Transition Matrix for Noisy Label Learningcited by 4
  12. 2024From Biased Selective Labels to Pseudo-Labels: An Expectation-Maximization Framework for Learning from Biased Decisionscited by 4
  13. 2025Calibrating Pre-trained Language Classifiers on LLM-generated Noisy Labels via Iterative Refinementcited by 4
  14. 2024ANNE: Adaptive Nearest Neighbors and Eigenvector-based Sample Selection for Robust Learning with Noisy Labelscited by 3
  15. 2024Learning Causal Transition Matrix for Instance-dependent Label Noisecited by 3
  16. 2025Enhanced Sample Selection with Confidence Tracking: Identifying Correctly Labeled yet Hard-to-Learn Samples in Noisy Datacited by 3
  17. 2022Do We Need to Penalize Variance of Losses for Learning with Label Noise?cited by 2
  18. 2026L2CU: Learning to Complement Unseen Userscited by 2
  19. 2022Maximising the Utility of Validation Sets for Imbalanced Noisy-label Meta-learningcited by 1
  20. 2023Asymmetric Co-teaching with Multi-view Consensus for Noisy Label Learningcited by 1
  21. 2023Towards the Identifiability in Noisy Label Learning: A Multinomial Mixture Modelling Approachcited by 1
  22. 2023PASS: Peer-Agreement based Sample Selection for training with Noisy Labelscited by 1
  23. 2023Bridging Generative and Discriminative Noisy-Label Learning via Direction-Agnostic EM Formulationcited by 1
  24. 2024Revisiting Early-Learning Regularization When Federated Learning Meets Noisy Labelscited by 1
  25. 2024NoisyAG-News: A Benchmark for Addressing Instance-Dependent Noise in Text Classificationcited by 1
  26. 2024Toward Fair Graph Neural Networks Via Dual-Teacher Knowledge Distillationcited by 1
  27. 2026Evaluating Feature Dependent Noise in Preference-based Reinforcement Learningcited by 1
  28. 2022Towards Harnessing Feature Embedding for Robust Learning with Noisy Labelscited by 0
  29. 2023Smoothly Giving up: Robustness for Simple Modelscited by 0
  30. 2023Rethinking the Value of Labels for Instance-Dependent Label Noise Learningcited by 0
  31. 2023Multi-Label Noise Transition Matrix Estimation with Label Correlations: Theory and Algorithmcited by 0
  32. 2024EchoAlign: Bridging Generative and Discriminative Learning under Noisy Labelscited by 0
  33. 2025Training Robust Graph Neural Networks by Modeling Noise Dependenciescited by 0
  34. 2025Learning to Clean: Reinforcement Learning for Noisy Label Correctioncited by 0
  35. 2026Variational Rectification Inference for Learning with Noisy Labelscited by 0
  36. 2026Learning Robust Recommenders from Noisy Implicit Feedback via GMM-Weighted Bayesian Transition Matrixcited by 0
  37. 2026Where LLM Annotators Fail: Label-Free Learning on Graphs with LLMscited by 0
  38. 2026Particle Competition and Cooperation for Robust Graph Convolutional Network Learning Under Label Noisecited by 0