Research map / Adversarial robustness and security / Certified and empirical robustness
Higher-Order Certification For Randomized Smoothing
Jeet Mohapatra, Ching-Yun Ko, Tsui-Wei Weng, Pin-Yu Chen, Sijia Liu, Luca Daniel
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Add first-order (gradient) information to the certificate instead of zeroth-order only | Gaussian-smoothed classifier on CIFAR10 · CIFAR10 certified accuracy, l2 | certified accuracy | Improved |
| Use first-order certificate for l1 radius | Gaussian-smoothed classifier on CIFAR10 · CIFAR10 certified accuracy, l1 | certified accuracy | Improved |
| Use first-order certificate for subspace l2 radius (red channel) | Gaussian-smoothed classifier on CIFAR10 · CIFAR10 certified accuracy, subspace l2 | certified accuracy | Improved |
| Use first-order certificate with 4,000,000 samples for l_inf radius | Gaussian-smoothed classifier on CIFAR10 · CIFAR10 certified accuracy, l_inf | certified accuracy | Improved |
| Compare Gaussian first-order certificate against uniform-smoothing state of the art for l1 alone | Gaussian-smoothed classifier on CIFAR10 · CIFAR10 certified accuracy, l1 | certified accuracy | Hurt |
| Evaluate multiple threat models jointly with the proposed method versus single-measure baselines | Gaussian-smoothed classifier · CIFAR10 joint threat models | certified accuracy | Improved |
Closest papers by meaning
- Incremental Randomized Smoothing Certification (ICLR 2024 · similarity 0.90)
- Average Certified Radius is a Poor Metric for Randomized Smoothing (ICML 2025 · similarity 0.90)
- Effects of Exponential Gaussian Distribution on (Double Sampling) Randomized Smoothing (ICML 2024 · similarity 0.90)
- The Lipschitz-Variance-Margin Tradeoff for Enhanced Randomized Smoothing (ICLR 2024 · similarity 0.89)
- Certified Adversarial Robustness via Randomized $\alpha$-Smoothing for Regression Models (NeurIPS 2024 · similarity 0.89)
- Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness (NeurIPS 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.
