Data to 14 Oct 2026

Learning with Symmetric Label Noise: The Importance of Being Unhinged

arXiv 1505.07634 ↗First posted 28 May 2015cs.LGcited by 120

Not on the map. Why?

Abstract (arXiv)

Convex potential minimisation is the de facto approach to binary classification. However, Long and Servedio [2010] proved that under symmetric label noise (SLN), minimisation of any convex potential over a linear function class can result in classification performance equivalent to random guessing. This ostensibly shows that convex losses are not SLN-robust. In this paper, we propose a convex, classification-calibrated loss and prove that it is SLN-robust. The loss avoids the Long and Servedio [2010] result by virtue of being negatively unbounded. The loss is a modification of the hinge loss, where one does not clamp at zero; hence, we call it the unhinged loss. We show that the optimal unhinged solution is equivalent to that of a strongly regularised SVM, and is the limiting solution for any convex potential; this implies that strong l2 regularisation makes most standard learners SLN-robust. Experiments confirm the SLN-robustness of the unhinged loss.

Key facts. Cited by 120 papers in the map (data through 14 Oct 2026). First posted on arXiv 28 May 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2020: 1; 2021: 42; 2022: 28; 2023: 19; 2024: 8; 2025: 16; 2026: 6.

120 papers.

Sort:

By area:

2021Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarkscited by 395
2022Decoupled Knowledge Distillationcited by 388
2021ASAM: Adaptive Sharpness-Aware Minimization for Scale-Invariant Learning of Deep Neural Networkscited by 293
2021Learning with Feature-Dependent Label Noise: A Progressive Approachcited by 102
2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmarkcited by 99
2022Fisher SAM: Information Geometry and Sharpness Aware Minimisationcited by 84
2021FINE Samples for Learning with Noisy Labelscited by 82
2021Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularizationcited by 62
2021Open-set Label Noise Can Improve Robustness Against Inherent Label Noisecited by 59
2022DrugOOD: Out-of-Distribution (OOD) Dataset Curator and Benchmark for AI-aided Drug Discovery -- A Focus on Affinity Prediction Problems with Noise Annotationscited by 54
2023Towards Label-free Scene Understanding by Vision Foundation Modelscited by 54
2021Asymmetric Loss Functions for Learning with Noisy Labelscited by 46

38 more

2022Investigating Why Contrastive Learning Benefits Robustness Against Label Noisecited by 46
2021Contrastive Learning Improves Model Robustness Under Label Noisecited by 36
2021Unsupervised Domain Adaptation of Black-Box Source Modelscited by 32
2021Meta-learning with an Adaptive Task Schedulercited by 27
2020Robust Imitation Learning from Noisy Demonstrationscited by 26
2022Towards Federated Learning against Noisy Labels via Local Self-Regularizationcited by 26
2021Noise-resistant Deep Metric Learning with Ranking-based Instance Selectioncited by 23
2022Few-shot Learning with Noisy Labelscited by 23
2021Noise-robust Graph Learning by Estimating and Leveraging Pairwise Interactionscited by 21
2022Is BERT Robust to Label Noise? A Study on Learning with Noisy Labels in Text Classificationcited by 21
2025Early Stopping Against Label Noise Without Validation Datacited by 21
2021Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrectioncited by 20
2021Understanding the Interaction of Adversarial Training with Noisy Labelscited by 19
2022Informative Pseudo-Labeling for Graph Neural Networks with Few Labelscited by 19
2021Pervasive AI for IoT applications: A Survey on Resource-efficient Distributed Artificial Intelligencecited by 18
2024Geometric-Averaged Preference Optimization for Soft Preference Labelscited by 18
2023Late Stopping: Avoiding Confidently Learning from Mislabeled Examplescited by 17
2021Delving into Sample Loss Curve to Embrace Noisy and Imbalanced Datacited by 16
2023BadLabel: A Robust Perspective on Evaluating and Enhancing Label-noise Learningcited by 16
2024Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustnesscited by 13
2021Tackling Instance-Dependent Label Noise via a Universal Probabilistic Modelcited by 12
2021Connecting Low-Loss Subspace for Personalized Federated Learningcited by 12
2022FedNoiL: A Simple Two-Level Sampling Method for Federated Learning with Noisy Labelscited by 12
2021Rethinking Importance Weighting for Transfer Learningcited by 11
2021Label Distributionally Robust Losses for Multi-class Classification: Consistency, Robustness and Adaptivitycited by 11
2023Robust Nonparametric Regression under Poisoning Attackcited by 11
2023ERASE: Error-Resilient Representation Learning on Graphs for Label Noise Tolerancecited by 11
2025Hide and Seek in Noise Labels: Noise-Robust Collaborative Active Learning with LLM-Powered Assistancecited by 11
2021Estimating the electrical power output of industrial devices with end-to-end time-series classification in the presence of label noisecited by 10
2022PADDLES: Phase-Amplitude Spectrum Disentangled Early Stopping for Learning with Noisy Labelscited by 10
2023Counterfactual Prediction Under Outcome Measurement Errorcited by 10
2023Are Neurons Actually Collapsed? On the Fine-Grained Structure in Neural Representationscited by 10
2023Complementary Domain Adaptation and Generalization for Unsupervised Continual Domain Shift Learningcited by 9
2021NoiLIn: Improving Adversarial Training and Correcting Stereotype of Noisy Labelscited by 8
2023Regroup Median Loss for Combating Label Noisecited by 8
2024Debiased Recommendation with Noisy Feedbackcited by 8
2021Knodle: Modular Weakly Supervised Learning with PyTorchcited by 7
2022Compressing Features for Learning with Noisy Labelscited by 7

50 most cited of 120 All 120 (JSON)