Set a Thief to Catch a Thief: Combating Label Noise through Noisy Meta Learning
Not on the map. Why?
Abstract (arXiv)
Learning from noisy labels (LNL) aims to train high-performance deep models using noisy datasets. Meta learning based label correction methods have demonstrated remarkable performance in LNL by designing various meta label rectification tasks. However, extra clean validation set is a prerequisite for these methods to perform label correction, requiring extra labor and greatly limiting their practicality. To tackle this issue, we propose a novel noisy meta label correction framework STCT, which counterintuitively uses noisy data to correct label noise, borrowing the spirit in the saying ``Set a Thief to Catch a Thief''. The core idea of STCT is to leverage noisy data which is i.i.d. with the training data as a validation set to evaluate model performance and perform label correction in a meta learning framework, eliminating the need for extra clean data. By decoupling the complex bi-level optimization in meta learning into representation learning and label correction, STCT is solved through an alternating training strategy between noisy meta correction and semi-supervised representation learning. Extensive experiments on synthetic and real-world datasets demonstrate the outstanding performance of STCT, particularly in high noise rate scenarios. STCT achieves 96.9% label correction and 95.2% classification performance on CIFAR-10 with 80% symmetric noise, significantly surpassing the current state-of-the-art.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 52 of its 69 references that are in the map. First posted on arXiv 22 Feb 2025.
Builds on
52 of 69 references.
References · 52
- 2017Attention Is All You Needcited by 60,361
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2016Wide Residual Networkscited by 3,070
- 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 997
- 2020Meta-Learning in Neural Networks: A Surveycited by 950
- 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 692
- 2018Learning to Reweight Examples for Robust Deep Learningcited by 675
- 2020Learning from Noisy Labels with Deep Neural Networks: A Surveycited by 582
- 2020Early-Learning Regularization Prevents Memorization of Noisy Labelscited by 416
- 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
38 more
- 2020Identifying Mislabeled Data using the Area Under the Margin Rankingcited by 261
- 2020Normalized Loss Functions for Deep Learning with Noisy Labelscited by 257
- 2021Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotationscited by 256
- 2019Probabilistic End-to-end Noise Correction for Learning with Noisy Labelscited by 210
- 2019Understanding and Utilizing Deep Neural Networks Trained with Noisy Labelscited by 209
- 2017CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noisecited by 181
- 2020CoMatch: Semi-supervised Learning with Contrastive Graph Regularizationcited by 167
- 2018Learning to Learn from Noisy Labeled Datacited by 160
- 2020Learning with Instance-Dependent Label Noise: A Sample Sieve Approachcited by 158
- 2020Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learningcited by 142
- 2019Deep Self-Learning From Noisy Labelscited by 122
- 2022SimMatch: Semi-supervised Learning with Similarity Matchingcited by 118
- 2022Selective-Supervised Contrastive Learning with Noisy Labelscited by 112
- 2018Iterative Learning with Open-set Noisy Labelscited by 108
- 2021Learning with Feature-Dependent Label Noise: A Progressive Approachcited by 102
- 2022Robust Training under Label Noise by Over-parameterizationcited by 93
- 2020Multi-Objective Interpolation Training for Robustness to Label Noisecited by 86
- 2021Jo-SRC: A Contrastive Approach for Combating Noisy Labelscited by 80
- 2021Contrast to Divide: Self-Supervised Pre-Training for Learning with Noisy Labelscited by 69
- 2021Learning Fast Sample Re-weighting Without Reward Datacited by 69
- 2021Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularizationcited by 62
- 2021Asymmetric Loss Functions for Learning with Noisy Labelscited by 46
- 2022Mitigating Memorization of Noisy Labels by Clipping the Model Predictioncited by 40
- 2020Learning to Purify Noisy Labels via Meta Soft Label Correctorcited by 39
- 2021Faster Meta Update Strategy for Noise-Robust Deep Learningcited by 38
- 2021Contrastive Learning Improves Model Robustness Under Label Noisecited by 36
- 2022Identifiability of Label Noise Transition Matrixcited by 34
- 2023Twin Contrastive Learning with Noisy Labelscited by 28
- 2023Noisy Correspondence Learning with Meta Similarity Correctioncited by 26
- 2023Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learningcited by 22
- 2022On Learning Contrastive Representations for Learning with Noisy Labelscited by 21
- 2023Learning from Noisy Labels with Decoupled Meta Label Purifiercited by 21
- 2020Robustness of Accuracy Metric and its Inspirations in Learning with Noisy Labelscited by 20
- 2021Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrectioncited by 20
- 2023FedDiv: Collaborative Noise Filtering for Federated Learning with Noisy Labelscited by 13
- 2022CrossSplit: Mitigating Label Noise Memorization through Data Splittingcited by 12
- 2022L2B: Learning to Bootstrap Robust Models for Combating Label Noisecited by 10
- 2024Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learningcited by 7
50 most cited of 52 All 52 (JSON)
Built on by
Not cited yet.