Research map / Adversarial robustness and security / Certified and empirical robustness
Black-Box Certification with Randomized Smoothing: A Functional Optimization Based Framework
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace Laplace smoothing with non-Laplacian centripetal distribution | l1 randomized smoothing · CIFAR-10 l1, radius 1.0 | certified accuracy (%): 30 -> 34 | Helped |
| replace Laplace smoothing with non-Laplacian centripetal distribution | l1 randomized smoothing · ImageNet l1, radius 3.0 | certified accuracy (%): 18 -> 22 | Helped |
| replace Gaussian smoothing with non-Gaussian l2-based distribution | l2 randomized smoothing · CIFAR-10 l2, radius 0.5 | certified accuracy (%): 43 -> 46 | Helped |
| replace Gaussian smoothing with non-Gaussian l2-based distribution | l2 randomized smoothing · ImageNet l2, radius 1.0 | certified accuracy (%): 37 -> 39 | Helped |
| replace Gaussian smoothing with mixed-norm smoothing distribution | linf randomized smoothing · CIFAR-10 linf, radius 8/255 | certified accuracy (%): 25 -> 32 | Helped |
| use l2-based family instead of mixed-norm family for linf certification (ablation) | linf certification smoothing family · toy example d=5 | accuracy-robustness Pareto frontier | Hurt |
| apply non-Gaussian smoothing to ImageNet linf | linf smoothing · ImageNet linf | certified accuracy | No clear effect |
Closest papers by meaning
- Incremental Randomized Smoothing Certification (ICLR 2024 · similarity 0.92)
- Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness (NeurIPS 2024 · similarity 0.91)
- Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing (ICML 2025 · similarity 0.91)
- Adaptive Randomized Smoothing: Certified Adversarial Robustness for Multi-Step Defences (NeurIPS 2024 · similarity 0.90)
- Certified Adversarial Robustness via Randomized $\alpha$-Smoothing for Regression Models (NeurIPS 2024 · similarity 0.90)
- DRF: Improving Certified Robustness via Distributional Robustness Framework (AAAI 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.