mlsanityData to 14 Oct 2026

Precise and Generalized Robustness Certification for Neural Networks

arXiv 2306.06747 ↗First posted 11 Jun 2023cs.CR cs.CV cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

The objective of neural network (NN) robustness certification is to determine if a NN changes its predictions when mutations are made to its inputs. While most certification research studies pixel-level or a few geometrical-level and blurring operations over images, this paper proposes a novel framework, GCERT, which certifies NN robustness under a precise and unified form of diverse semantic-level image mutations. We formulate a comprehensive set of semantic-level image mutations uniformly as certain directions in the latent space of generative models. We identify two key properties, independence and continuity, that convert the latent space into a precise and analysis-friendly input space representation for certification. GCERT can be smoothly integrated with de facto complete, incomplete, or quantitative certification frameworks. With its precise input space representation, GCERT enables for the first time complete NN robustness certification with moderate cost under diverse semantic-level input mutations, such as weather-filter, style transfer, and perceptual changes (e.g., opening/closing eyes). We show that GCERT enables certifying NN robustness under various common and security-sensitive scenarios like autonomous driving.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 47 of its 74 references that are in the map. First posted on arXiv 11 Jun 2023.

Builds on

47 of 74 references.

References · 47

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  4. 2022Generative Adversarial Networkscited by 5,627
  5. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
  6. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  7. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  8. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  10. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  11. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,545
  12. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
35 more
  1. 2019Natural Adversarial Examplescited by 1,408
  2. 2014Deep Learning in Neural Networks: An Overviewcited by 770
  3. 2017Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkscited by 707
  4. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 468
  5. 2018Efficient Neural Network Robustness Certification with General Activation Functionscited by 445
  6. 2017DeepXplore: Automated Whitebox Testing of Deep Learning Systemscited by 311
  7. 2017DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Carscited by 251
  8. 2017Image reconstruction by domain transform manifold learningcited by 232
  9. 2021Beta-CROWN: Efficient Bound Propagation with Per-neuron Split Constraints for Complete and Incomplete Neural Network Robustness Verificationcited by 228
  10. 2018Adversarial Audio Synthesiscited by 225
  11. 2021High-Fidelity GAN Inversion for Image Attribute Editingcited by 197
  12. 2018Spatially Transformed Adversarial Examplescited by 195
  13. 2020Fast and Complete: Enabling Complete Neural Network Verification with Rapid and Massively Parallel Incomplete Verifierscited by 187
  14. 2018Formal Security Analysis of Neural Networks using Symbolic Intervalscited by 185
  15. 2020SoK: Certified Robustness for Deep Neural Networkscited by 105
  16. 2022Complete Verification via Multi-Neuron Relaxation Guided Branch-and-Boundcited by 92
  17. 2021PRIMA: General and Precise Neural Network Certification via Scalable Convex Hull Approximationscited by 77
  18. 2020Learning Certified Individually Fair Representationscited by 69
  19. 2021Low-Rank Subspaces in GANscited by 53
  20. 2020Certified Defense to Image Transformations via Randomized Smoothingcited by 51
  21. 2020TSS: Transformation-Specific Smoothing for Robustness Certificationcited by 49
  22. 2019Perfectly Parallel Fairness Certification of Neural Networkscited by 42
  23. 2020Explainable Face Recognitioncited by 34
  24. 2019Provable Certificates for Adversarial Examples: Fitting a Ball in the Union of Polytopescited by 25
  25. 2022GSmooth: Certified Robustness against Semantic Transformations via Generalized Randomized Smoothingcited by 23
  26. 2020Robustness Certification of Generative Modelscited by 19
  27. 2020On Noise Injection in Generative Adversarial Networkscited by 19
  28. 2021CC-Cert: A Probabilistic Approach to Certify General Robustness of Neural Networkscited by 19
  29. 2021DeformRS: Certifying Input Deformations with Randomized Smoothingcited by 17
  30. 2021Robustness Certification for Point Cloud Modelscited by 16
  31. 2019Computing Linear Restrictions of Neural Networkscited by 15
  32. 2021Exposing Previously Undetectable Faults in Deep Neural Networkscited by 11
  33. 2021Revisiting Neuron Coverage for DNN Testing: A Layer-Wise and Distribution-Aware Criterioncited by 11
  34. 2020Rethinking conditional GAN training: An approach using geometrically structured latent manifoldscited by 5
  35. 2022Unveiling Hidden DNN Defects with Decision-Based Metamorphic Testingcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 3.

4 papers.

  1. 2024Adversarial Feature Alignment: Balancing Robustness and Accuracy in Deep Learning via Adversarial Trainingcited by 5
  2. 2026Systematic Discovery of Semantic Attacks in Online Map Construction through Conditional Diffusioncited by 1
  3. 2026Semantic Robustness Certification for Vision-Language Modelscited by 1
  4. 2026Towards Compact and Robust DNNs via Compression-aware Sharpness Minimizationcited by 0