Research map / Adversarial robustness and security / Certified and empirical robustness
Certified Adversarial Robustness with Additive Noise
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 |
|---|---|---|---|
| STN (Gaussian noise + stability training) vs Wong et al. cascade | certified defense · CIFAR-10 | certified robust accuracy: 58.7% -> 65.6% | Helped |
| STN vs Wong et al. cascade | certified defense · MNIST | certified robust accuracy: 74.6% -> 69.0% | Hurt |
| STN vs Wong et al. cascade | certified defense · MNIST | natural accuracy: 81.4% -> 98.9% | Helped |
| add stability training (ablation) | Gaussian-noise certified classifier · MNIST and CIFAR-10 | certified accuracy lower bound | Helped |
| STN vs PixelDP | certified defense · MNIST, CIFAR-10, ImageNet | certified accuracy lower bound | Helped |
| increase noise sigma (ablation) | STN · MNIST and CIFAR-10 | empirical robust accuracy | Mixed |
| add test-time Gaussian noise (sigma>0 vs sigma=0) (ablation) | STN · MNIST and CIFAR-10 | empirical robust accuracy | Helped |
| STN vs TRADES | adversarial defense · MNIST and CIFAR-10 | empirical robust accuracy | Mixed |
| stability training instead of Gaussian augmentation (ablation) | noise-based certified classifier · certified bound | certified accuracy | Helped |
Closest papers by meaning
- On the Scalability of Certified Adversarial Robustness with Generated Data (NeurIPS 2024 · similarity 0.94)
- Theoretically Principled Trade-off between Robustness and Accuracy (ICML 2019 · similarity 0.92)
- Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing (ICML 2025 · similarity 0.92)
- Adversarial Training and Robustness for Multiple Perturbations (NeurIPS 2019 · similarity 0.92)
- Stable Unlearnable Example: Enhancing the Robustness of Unlearnable Examples via Stable Error-Minimizing Noise (AAAI 2024 · similarity 0.91)
- Adversarial Robustness via Deformable Convolution with Stochasticity (ICML 2025 · similarity 0.90)
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.