mlsanityData to 7 Oct 2026

NoisyMix: Boosting Model Robustness to Common Corruptions

arXiv 2202.01263 ↗First posted 2 Feb 2022cs.LG stat.MLcited by 17

Not on the map. Why?

Abstract (arXiv)

For many real-world applications, obtaining stable and robust statistical performance is more important than simply achieving state-of-the-art predictive test accuracy, and thus robustness of neural networks is an increasingly important topic. Relatedly, data augmentation schemes have been shown to improve robustness with respect to input perturbations and domain shifts. Motivated by this, we introduce NoisyMix, a novel training scheme that promotes stability as well as leverages noisy augmentations in input and feature space to improve both model robustness and in-domain accuracy. NoisyMix produces models that are consistently more robust and that provide well-calibrated estimates of class membership probabilities. We demonstrate the benefits of NoisyMix on a range of benchmark datasets, including ImageNet-C, ImageNet-R, and ImageNet-P. Moreover, we provide theory to understand implicit regularization and robustness of NoisyMix.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 45 of its 76 references that are in the map. First posted on arXiv 2 Feb 2022.

Builds on

45 of 76 references.

References · 45

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  5. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2017On Calibration of Modern Neural Networkscited by 4,860
  8. 2016Wide Residual Networkscited by 3,057
  9. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  10. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  11. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  12. 2016Identity Mappings in Deep Residual Networkscited by 1,960
33 more
  1. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  2. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  3. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  4. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  5. 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 950
  6. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  7. 2019Adversarial Training for Free!cited by 638
  8. 2020Overfitting in adversarially robust deep learningcited by 504
  9. 2020Measuring Robustness to Natural Distribution Shifts in Image Classificationcited by 452
  10. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  11. 2020Do Adversarially Robust ImageNet Models Transfer Better?cited by 315
  12. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  13. 2020Puzzle Mix: Exploiting Saliency and Local Statistics for Optimal Mixupcited by 260
  14. 2018Adversarial Attacks and Defences: A Surveycited by 259
  15. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 201
  16. 2020How Does Mixup Help With Robustness and Generalization?cited by 193
  17. 2021On the Origin of Implicit Regularization in Stochastic Gradient Descentcited by 183
  18. 2021Revisiting ResNets: Improved Training and Scaling Strategiescited by 158
  19. 2016Improving the Robustness of Deep Neural Networks via Stability Trainingcited by 156
  20. 2013Stabilitycited by 119
  21. 2016Understanding How Image Quality Affects Deep Neural Networkscited by 118
  22. 2017Regularization for Deep Learning: A Taxonomycited by 91
  23. 2021AugMax: Adversarial Composition of Random Augmentations for Robust Trainingcited by 89
  24. 2021Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labelscited by 74
  25. 2019Adversarially robust transfer learningcited by 66
  26. 2021A Winning Hand: Compressing Deep Networks Can Improve Out-Of-Distribution Robustnesscited by 62
  27. 2020The Implicit Regularization of Stochastic Gradient Flow for Least Squarescited by 58
  28. 2016Noisy Activation Functionscited by 53
  29. 2019Interpolated Adversarial Training: Achieving Robust Neural Networks without Sacrificing Too Much Accuracycited by 49
  30. 2021Noisy Recurrent Neural Networkscited by 42
  31. 2020Explicit Regularisation in Gaussian Noise Injectionscited by 37
  32. 2019From GAN to WGANcited by 34
  33. 2021Noisy Feature Mixupcited by 27

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 5; 2024: 4; 2025: 5; 2026: 1.

17 papers.

Sort:

  1. 2023Robustness of SAM: Segment Anything Under Corruptions and Beyondcited by 30
  2. 2023A Survey on the Robustness of Computer Vision Models against Common Corruptionscited by 20
  3. 2024Can Biases in ImageNet Models Explain Generalization?cited by 20
  4. 2023Classification robustness to common optical aberrationscited by 13
  5. 2023Tailoring Mixup to Data for Calibrationcited by 5
  6. 2022Dynamic Test-Time Augmentation via Differentiable Functionscited by 4
  7. 2024How Do Training Methods Influence the Utilization of Vision Models?cited by 3
  8. 2022Universum-inspired Supervised Contrastive Learningcited by 2
  9. 2024Assessing Visually-Continuous Corruption Robustness of Neural Networks Relative to Human Performancecited by 2
  10. 2025Examining the Impact of Optical Aberrations to Image Classification and Object Detection Modelscited by 2
  11. 2025Curvature Enhanced Data Augmentation for Regressioncited by 1
  12. 2025Improving Mixup Calibration with Wasserstein Distributionally Robust Optimizationcited by 1
5 more
  1. 2026One Model, Many Behaviors: Training-Induced Effects on Out-of-Distribution Detectioncited by 1
  2. 2023Don't Look into the Sun: Adversarial Solarization Attacks on Image Classifierscited by 0
  3. 2024Global Average Feature Augmentation for Robust Semantic Segmentation with Transformerscited by 0
  4. 2025Combined Image Data Augmentations diminish the benefits of Adaptive Label Smoothingcited by 0
  5. 2025Stylized Synthetic Augmentation further improves Corruption Robustnesscited by 0