mlsanityData to 7 Oct 2026

Incremental Randomized Smoothing Certification

arXiv 2305.19521 ↗First posted 31 May 2023cs.LG cs.CR cs.PLcited by 12

Not on the map. Why?

Abstract (arXiv)

Randomized smoothing-based certification is an effective approach for obtaining robustness certificates of deep neural networks (DNNs) against adversarial attacks. This method constructs a smoothed DNN model and certifies its robustness through statistical sampling, but it is computationally expensive, especially when certifying with a large number of samples. Furthermore, when the smoothed model is modified (e.g., quantized or pruned), certification guarantees may not hold for the modified DNN, and recertifying from scratch can be prohibitively expensive. We present the first approach for incremental robustness certification for randomized smoothing, IRS. We show how to reuse the certification guarantees for the original smoothed model to certify an approximated model with very few samples. IRS significantly reduces the computational cost of certifying modified DNNs while maintaining strong robustness guarantees. We experimentally demonstrate the effectiveness of our approach, showing up to 3x certification speedup over the certification that applies randomized smoothing of the approximate model from scratch.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 33 of its 66 references that are in the map. First posted on arXiv 31 May 2023.

Builds on

33 of 66 references.

References · 33

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,250
  3. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
  4. 2016End to End Learning for Self-Driving Carscited by 1,000
  5. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
  6. 2017Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weightscited by 211
  7. 2020Randomized Smoothing of All Shapes and Sizescited by 146
  8. 2018Learning to Optimize Tensor Programscited by 138
  9. 2019Branch and Bound for Piecewise Linear Neural Network Verificationcited by 138
  10. 2020Certified Robustness to Label-Flipping Attacks via Randomized Smoothingcited by 103
  11. 2020(De)Randomized Smoothing for Certifiable Defense against Patch Attackscited by 97
  12. 2019Tight Certificates of Adversarial Robustness for Randomly Smoothed Classifierscited by 65
21 more
  1. 2018Deep Neural Network Compression for Aircraft Collision Avoidance Systemscited by 58
  2. 2020Certified Robustness of Graph Neural Networks against Adversarial Structural Perturbationcited by 57
  3. 2020Efficient Robustness Certificates for Discrete Data: Sparsity-Aware Randomized Smoothing for Graphs, Images and Morecited by 56
  4. 2019Certified Robustness for Top-k Predictions against Adversarial Perturbations via Randomized Smoothingcited by 54
  5. 2020Certified Defense to Image Transformations via Randomized Smoothingcited by 51
  6. 2020TSS: Transformation-Specific Smoothing for Robustness Certificationcited by 49
  7. 2021PointGuard: Provably Robust 3D Point Cloud Classificationcited by 49
  8. 2020Detection as Regression: Certified Object Detection by Median Smoothingcited by 47
  9. 2021Boosting Randomized Smoothing with Variance Reduced Classifierscited by 47
  10. 2020Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Frameworkcited by 45
  11. 2020Certifying Confidence via Randomized Smoothingcited by 36
  12. 2020Higher-Order Certification for Randomized Smoothingcited by 36
  13. 2021Scalable Certified Segmentation via Randomized Smoothingcited by 35
  14. 2023Collective Robustness Certificates: Exploiting Interdependence in Graph Neural Networkscited by 16
  15. 2023Incremental Verification of Neural Networkscited by 14
  16. 2020Certified Robustness of Graph Classification against Topology Attack with Randomized Smoothingcited by 12
  17. 2013The cost of using exact confidence intervals for a binomial proportioncited by 11
  18. 2021Shared Certificates for Neural Network Verificationcited by 10
  19. 2021Integer-arithmetic-only Certified Robustness for Quantized Neural Networkscited by 7
  20. 2021Demanded Abstract Interpretation (Extended Version)cited by 4
  21. 2022Sound Randomized Smoothing in Floating-Point Arithmeticscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 2; 2026: 5.

12 papers.

  1. 2024Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustnesscited by 8
  2. 2023Fast Certification of Vision-Language Models Using Incremental Randomized Smoothingcited by 4
  3. 2023Rethinking Randomized Smoothing from the Perspective of Scalabilitycited by 3
  4. 2025BEAVER: An Efficient Deterministic LLM Verifiercited by 2
  5. 2024Integrating uncertainty quantification into randomized smoothing based robustness guaranteescited by 1
  6. 2024ARQ: A Mixed-Precision Quantization Framework for Accurate and Certifiably Robust DNNscited by 1
  7. 2026Certified vs. Empirical Adversarial Robust-ness via Hybrid Convolutions with Attention Stochasticitycited by 1
  8. 2025Support is All You Need for Certified VAE Trainingcited by 0
  9. 2026PDA: Text-Augmented Defense Framework for Robust Vision-Language Models against Adversarial Image Attackscited by 0
  10. 2026Laplace-Bridged Randomized Smoothing for Fast Certified Robustnesscited by 0
  11. 2026RRISE: Robust Radius Inference via a Surrogate Estimatorcited by 0
  12. 2026Uncovering the Limits of Proof Sharing for Neural Networkscited by 0