mlsanityData to 7 Oct 2026

Defending Against Image Corruptions Through Adversarial Augmentations

arXiv 2104.01086 ↗First posted 2 Apr 2021cs.CV cs.LGcited by 28

Not on the map. Why?

Abstract (arXiv)

Modern neural networks excel at image classification, yet they remain vulnerable to common image corruptions such as blur, speckle noise or fog. Recent methods that focus on this problem, such as AugMix and DeepAugment, introduce defenses that operate in expectation over a distribution of image corruptions. In contrast, the literature on ℓp-norm bounded perturbations focuses on defenses against worst-case corruptions. In this work, we reconcile both approaches by proposing AdversarialAugment, a technique which optimizes the parameters of image-to-image models to generate adversarially corrupted augmented images. We theoretically motivate our method and give sufficient conditions for the consistency of its idealized version as well as that of DeepAugment. Our classifiers improve upon the state-of-the-art on common image corruption benchmarks conducted in expectation on CIFAR-10-C and improve worst-case performance against ℓp-norm bounded perturbations on both CIFAR-10 and ImageNet.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). Builds on 68 of its 90 references that are in the map. First posted on arXiv 2 Apr 2021.

Builds on

68 of 90 references.

References · 68

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  7. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  8. 2013Intriguing properties of neural networkscited by 5,114
  9. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  10. 2017Neural Discrete Representation Learningcited by 4,650
  11. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  12. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
38 more
  1. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  2. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  3. 2018Deep contextualized word representationscited by 2,285
  4. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,261
  5. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  6. 2020The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalizationcited by 1,842
  7. 2019Invariant Risk Minimizationcited by 1,832
  8. 2016Adversarial examples in the physical worldcited by 1,762
  9. 2016Understanding deep learning requires rethinking generalizationcited by 1,657
  10. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  11. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  12. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  13. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,507
  14. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  15. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  16. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  17. 2015Spatial Transformer Networkscited by 1,423
  18. 2020Sharpness-Aware Minimization for Efficiently Improving Generalizationcited by 1,404
  19. 2019Natural Adversarial Examplescited by 1,401
  20. 2016End to End Learning for Self-Driving Carscited by 1,000
  21. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  22. 2018AutoAugment: Learning Augmentation Policies from Datacited by 728
  23. 2020Fast is better than free: Revisiting adversarial trainingcited by 723
  24. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  25. 2020RobustBench: a standardized adversarial robustness benchmarkcited by 659
  26. 2017Exploring Generalization in Deep Learningcited by 595
  27. 2017Adversarial Examples Are Not Easily Detected: Bypassing Ten Detection Methodscited by 542
  28. 2020Overfitting in adversarially robust deep learningcited by 504
  29. 2020Measuring Robustness to Natural Distribution Shifts in Image Classificationcited by 452
  30. 2017Nearly-tight VC-dimension and pseudodimension bounds for piecewise linear neural networkscited by 304
  31. 2017Lossy Image Compression with Compressive Autoencoderscited by 289
  32. 2017Size-Independent Sample Complexity of Neural Networkscited by 289
  33. 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
  34. 2020Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examplescited by 275
  35. 2018Generating Adversarial Examples with Adversarial Networkscited by 256
  36. 2018Generalisation in humans and deep neural networkscited by 237
  37. 2018Robustness via curvature regularization, and vice versacited by 174
  38. 2018MixUp as Locally Linear Out-Of-Manifold Regularizationcited by 158

50 most cited of 68 All 68 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 8; 2023: 10; 2024: 3; 2025: 2; 2026: 1.

28 papers.

Sort:

By area:

  1. 2021AugMax: Adversarial Composition of Random Augmentations for Robust Trainingcited by 89
  2. 2022Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptionscited by 78
  3. 2021A Winning Hand: Compressing Deep Networks Can Improve Out-Of-Distribution Robustnesscited by 62
  4. 2021Score-Based Generative Classifierscited by 53
  5. 2021PRIME: A few primitives can boost robustness to common corruptionscited by 40
  6. 2023Quality-Agnostic Deepfake Detection with Intra-model Collaborative Learningcited by 32
  7. 2022GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothingcited by 23
  8. 2022Enhance the Visual Representation via Discrete Adversarial Trainingcited by 23
  9. 2023Seasoning Model Soups for Robustness to Adversarial and Natural Distribution Shiftscited by 23
  10. 2023A Survey on the Robustness of Computer Vision Models against Common Corruptionscited by 20
  11. 2023IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifierscited by 14
  12. 2022Improving Robustness by Enhancing Weak Subnetscited by 12
16 more
  1. 2023HybridAugment++: Unified Frequency Spectra Perturbations for Model Robustnesscited by 12
  2. 2023Distilling Out-of-Distribution Robustness from Vision-Language Foundation Modelscited by 9
  3. 2024Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Modelscited by 9
  4. 2022How Robust are Discriminatively Trained Zero-Shot Learning Models?cited by 6
  5. 2023RobustCLEVR: A Benchmark and Framework for Evaluating Robustness in Object-centric Learningcited by 6
  6. 2023Improving Model Generalization by On-manifold Adversarial Augmentation in the Frequency Domaincited by 5
  7. 2022On Certifying and Improving Generalization to Unseen Domainscited by 4
  8. 2023On the Fly Neural Style Smoothing for Risk-Averse Domain Generalizationcited by 3
  9. 2025A Validation Strategy for Deep Learning Models: Evaluating and Enhancing Robustnesscited by 1
  10. 2026TestifAI: Tomography-Based Testing for Deep Learning Systemscited by 1
  11. 2022Gradient Aligned Attacks via a Few Queriescited by 0
  12. 2022FACM: Intermediate Layer Still Retain Effective Features against Adversarial Examplescited by 0
  13. 2023Improving Robustness via Tilted Exponential Layer: A Communication-Theoretic Perspectivecited by 0
  14. 2024Linearizing Models for Efficient yet Robust Private Inferencecited by 0
  15. 2024Growing Efficient Accurate and Robust Neural Networks on the Edgecited by 0
  16. 2025LayerMix: Enhanced Data Augmentation through Fractal Integration for Robust Deep Learningcited by 0