Research map / Adversarial robustness and security / Certified and empirical robustness
Center Smoothing: Certified Robustness for Networks with Structured Outputs
Aounon Kumar, Tom Goldstein
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 |
|---|---|---|---|
| Train the base model with Gaussian noise matching the smoothing noise (ablation) | Center smoothing for total-variation certificates · MNIST and CIFAR-10 dimensionality reduction and image reconstruction | certified output radius | Improved |
| Increase h = ε₁/σ | Center smoothing · Face detection, generative model, and total-variation experiments | certified output radius | Improved |
| Use a higher h such that q exceeds one | Center smoothing certification procedure · Structured-output experiments | certification success | Hurt |
| Use center smoothing instead of the expectation-smoothed ℓ₂ baseline | Autoencoder and image-reconstruction models · MNIST and CIFAR-10 | certified output radius | Improved |
Closest papers by meaning
- Certified Adversarial Robustness via Randomized Smoothing (ICML 2019 · similarity 0.89)
- A FRAMEWORK FOR ROBUSTNESS CERTIFICATION OF SMOOTHED CLASSIFIERS USING F-DIVERGENCES (ICLR 2020 · similarity 0.89)
- Certified Adversarial Robustness via Randomized $\alpha$-Smoothing for Regression Models (NeurIPS 2024 · similarity 0.88)
- Learning Better Certified Models from Empirically-Robust Teachers (NeurIPS 2026 · similarity 0.87)
- Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness (NeurIPS 2024 · similarity 0.87)
- Average Certified Radius is a Poor Metric for Randomized Smoothing (ICML 2025 · similarity 0.87)
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.
