mlsanityData to 7 Oct 2026

GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothing

arXiv 2206.04310 ↗First posted 9 Jun 2022cs.LG cs.AI cs.CRcited by 23

Not on the map. Why?

Abstract (arXiv)

Certified defenses such as randomized smoothing have shown promise towards building reliable machine learning systems against ℓp-norm bounded attacks. However, existing methods are insufficient or unable to provably defend against semantic transformations, especially those without closed-form expressions (such as defocus blur and pixelate), which are more common in practice and often unrestricted. To fill up this gap, we propose generalized randomized smoothing (GSmooth), a unified theoretical framework for certifying robustness against general semantic transformations via a novel dimension augmentation strategy. Under the GSmooth framework, we present a scalable algorithm that uses a surrogate image-to-image network to approximate the complex transformation. The surrogate model provides a powerful tool for studying the properties of semantic transformations and certifying robustness. Experimental results on several datasets demonstrate the effectiveness of our approach for robustness certification against multiple kinds of semantic transformations and corruptions, which is not achievable by the alternative baselines.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 33 of its 44 references that are in the map. First posted on arXiv 9 Jun 2022.

Builds on

33 of 44 references.

References · 33

  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. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  6. 2013Intriguing properties of neural networkscited by 5,114
  7. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  8. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  9. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  10. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  11. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  12. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
21 more
  1. 2017Evasion Attacks against Machine Learning at Test Timecited by 686
  2. 2017Provable defenses against adversarial examples via the convex outer adversarial polytopecited by 564
  3. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 462
  4. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 322
  5. 2019Adversarial Examples Improve Image Recognitioncited by 283
  6. 2019ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed datacited by 260
  7. 2019Towards Stable and Efficient Training of Verifiably Robust Neural Networkscited by 204
  8. 2018Spatially Transformed Adversarial Examplescited by 195
  9. 2018Physical Adversarial Examples for Object Detectorscited by 163
  10. 2020Randomized Smoothing of All Shapes and Sizescited by 146
  11. 2019Implicit Semantic Data Augmentation for Deep Networkscited by 106
  12. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  13. 2018Semantic Adversarial Examplescited by 86
  14. 2020Certified Defense to Image Transformations via Randomized Smoothingcited by 51
  15. 2020TSS: Transformation-Specific Smoothing for Robustness Certificationcited by 49
  16. 2021Defending Against Image Corruptions Through Adversarial Augmentationscited by 28
  17. 2020Model-Based Robust Deep Learning: Generalizing to Natural, Out-of-Distribution Datacited by 27
  18. 2017Towards Practical Verification of Machine Learning: The Case of Computer Vision Systemscited by 19
  19. 2022Toward Certified Robustness Against Real-World Distribution Shiftscited by 16
  20. 2020Extensions and limitations of randomized smoothing for robustness guaranteescited by 15
  21. 2021Understanding Adversarial Attacks on Observations in Deep Reinforcement Learningcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 8; 2024: 4; 2025: 4; 2026: 5.

23 papers.

Sort:

By area:

  1. 2023Robust Classification via a Single Diffusion Modelcited by 79
  2. 2023Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attackscited by 32
  3. 2024Provably Unlearnable Data Examplescited by 18
  4. 2023RS-Del: Edit Distance Robustness Certificates for Sequence Classifiers via Randomized Deletioncited by 17
  5. 2024Mitigating the Curse of Dimensionality for Certified Robustness via Dual Randomized Smoothingcited by 15
  6. 2022Provable Defense Against Geometric Transformationscited by 14
  7. 2023Towards Verifying the Geometric Robustness of Large-scale Neural Networkscited by 11
  8. 2023Precise and Generalized Robustness Certification for Neural Networkscited by 4
  9. 2024Quantum-Inspired Analysis of Neural Network Vulnerabilities: The Role of Conjugate Variables in System Attackscited by 4
  10. 2022Robustness Certification of Visual Perception Models via Camera Motion Smoothingcited by 3
  11. 2025Probabilistic Robustness Analysis in High Dimensional Space: Application to Semantic Segmentation Networkcited by 3
  12. 2023Pixel-wise Smoothing for Certified Robustness against Camera Motion Perturbationscited by 2
11 more
  1. 2025Reconcile Certified Robustness and Accuracy for DNN-based Smoothed Majority Vote Classifiercited by 2
  2. 2023Is Certifying ℓp Robustness Still Worthwhile?cited by 1
  3. 2024Effects of Exponential Gaussian Distribution on (Double Sampling) Randomized Smoothingcited by 1
  4. 2025CAMP in the Odyssey: Provably Robust Reinforcement Learning with Certified Radius Maximizationcited by 1
  5. 2026Certified vs. Empirical Adversarial Robust-ness via Hybrid Convolutions with Attention Stochasticitycited by 1
  6. 2023General Lipschitz: Certified Robustness Against Resolvable Semantic Transformations via Transformation-Dependent Randomized Smoothingcited by 0
  7. 2025Principal Eigenvalue Regularization for Improved Worst-Class Certified Robustness of Smoothed Classifierscited by 0
  8. 2026Feature-Space Smoothing: Certified Robustness of Deep Representationscited by 0
  9. 2026Faults That Fortify: CNN Adversarial Robustness via GPU Undervoltingcited by 0
  10. 2026NNV3: Expanding Neural Network Verification to New Architectures and Domainscited by 0
  11. 2026Interpretable but Fragile? Robustness of Concept Bottlenecks under Geometric-Semantic Perturbationscited by 0