mlsanityData to 7 Oct 2026

Curriculum Loss: Robust Learning and Generalization against Label Corruption

arXiv 1905.10045 ↗First posted 24 May 2019cs.LG stat.MLcited by 87

Not on the map. Why?

Abstract (arXiv)

Deep neural networks (DNNs) have great expressive power, which can even memorize samples with wrong labels. It is vitally important to reiterate robustness and generalization in DNNs against label corruption. To this end, this paper studies the 0-1 loss, which has a monotonic relationship with an empirical adversary (reweighted) risk~citephu2016does. Although the 0-1 loss has some robust properties, it is difficult to optimize. To efficiently optimize the 0-1 loss while keeping its robust properties, we propose a very simple and efficient loss, i.e. curriculum loss (CL). Our CL is a tighter upper bound of the 0-1 loss compared with conventional summation based surrogate losses. Moreover, CL can adaptively select samples for model training. As a result, our loss can be deemed as a novel perspective of curriculum sample selection strategy, which bridges a connection between curriculum learning and robust learning. Experimental results on benchmark datasets validate the robustness of the proposed loss.

Key facts. Cited by 87 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 30; 2022: 19; 2023: 19; 2024: 12; 2025: 3; 2026: 4.

87 papers.

Sort:

By area:

  1. 2021Understanding and Improving Early Stopping for Learning with Noisy Labelscited by 112
  2. 2021Sample Selection with Uncertainty of Losses for Learning with Noisy Labelscited by 79
  3. 2023IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturingcited by 64
  4. 2021Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularizationcited by 62
  5. 2021Open-set Label Noise Can Improve Robustness Against Inherent Label Noisecited by 58
  6. 2022Instance-Dependent Label-Noise Learning with Manifold-Regularized Transition Matrix Estimationcited by 49
  7. 2021Asymmetric Loss Functions for Learning with Noisy Labelscited by 46
  8. 2024RIME: Robust Preference-based Reinforcement Learning with Noisy Preferencescited by 37
  9. 2021Estimating Instance-dependent Bayes-label Transition Matrix using a Deep Neural Networkcited by 30
  10. 2021Continual Learning on Noisy Data Streams via Self-Purified Replaycited by 27
  11. 2022CLNode: Curriculum Learning for Node Classificationcited by 24
  12. 2021Learning with Noisy Labels for Robust Point Cloud Segmentationcited by 22
38 more
  1. 2021Adaptive Sample Selection for Robust Learning under Label Noisecited by 20
  2. 2023SILT: Shadow-aware Iterative Label Tuning for Learning to Detect Shadows from Noisy Labelscited by 19
  3. 2023Late Stopping: Avoiding Confidently Learning from Mislabeled Examplescited by 17
  4. 2023Learning to Segment from Noisy Annotations: A Spatial Correction Approachcited by 16
  5. 2021Uncertainty-aware Clustering for Unsupervised Domain Adaptive Object Re-identificationcited by 15
  6. 2022Scalable Penalized Regression for Noise Detection in Learning with Noisy Labelscited by 15
  7. 2021Improving Medical Image Classification with Label Noise Using Dual-uncertainty Estimationcited by 13
  8. 2021Model Generalization on COVID-19 Fake News Detectioncited by 11
  9. 2021On the Robustness of Monte Carlo Dropout Trained with Noisy Labelscited by 11
  10. 2021SENT: Sentence-level Distant Relation Extraction via Negative Trainingcited by 10
  11. 2022PADDLES: Phase-Amplitude Spectrum Disentangled Early Stopping for Learning with Noisy Labelscited by 10
  12. 2021Evaluating Multi-label Classifiers with Noisy Labelscited by 9
  13. 2021CCMN: A General Framework for Learning with Class-Conditional Multi-Label Noisecited by 9
  14. 2022Instance-dependent Label Distribution Estimation for Learning with Label Noisecited by 9
  15. 2023Latent Class-Conditional Noise Modelcited by 9
  16. 2024AIO2: Online Correction of Object Labels for Deep Learning with Incomplete Annotation in Remote Sensing Image Segmentationcited by 9
  17. 2024NLPrompt: Noise-Label Prompt Learning for Vision-Language Modelscited by 9
  18. 2021Sample Prior Guided Robust Model Learning to Suppress Noisy Labelscited by 8
  19. 2023Regularly Truncated M-estimators for Learning with Noisy Labelscited by 8
  20. 2024Tackling Noisy Labels with Network Parameter Additive Decompositioncited by 8
  21. 2024Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learningcited by 7
  22. 2021Truth Discovery in Sequence Labels from Crowdscited by 6
  23. 2022Less is More: Adaptive Curriculum Learning for Thyroid Nodule Diagnosiscited by 6
  24. 2022Robust Point Cloud Segmentation with Noisy Annotationscited by 6
  25. 2024Mitigating Label Noise on Graph via Topological Sample Selectioncited by 6
  26. 2022DiscrimLoss: A Universal Loss for Hard Samples and Incorrect Samples Discriminationcited by 5
  27. 2022A Survey of Dataset Refinement for Problems in Computer Vision Datasetscited by 5
  28. 2021Learning to Combat Noisy Labels via Classification Marginscited by 4
  29. 2021Multiplicative Reweighting for Robust Neural Network Optimizationcited by 4
  30. 2021Co-matching: Combating Noisy Labels by Augmentation Anchoringcited by 4
  31. 2021On Universal Black-Box Domain Adaptationcited by 4
  32. 2022Prototype-Anchored Learning for Learning with Imperfect Annotationscited by 4
  33. 2022Truncate-Split-Contrast: A Framework for Learning from Mislabeled Videoscited by 4
  34. 2021Approximating Instance-Dependent Noise via Instance-Confidence Embeddingcited by 3
  35. 2021How Does Heterogeneous Label Noise Impact Generalization in Neural Nets?cited by 3
  36. 2021Batch Inverse-Variance Weighting: Deep Heteroscedastic Regressioncited by 3
  37. 2021Elucidating Robust Learning with Uncertainty-Aware Corruption Pattern Estimationcited by 3
  38. 2023Improve Noise Tolerance of Robust Loss via Noise-Awarenesscited by 3

50 most cited of 87 All 87 (JSON)