Research map / Adversarial robustness and security / Certified and empirical robustness
When are Non-Parametric Methods Robust?
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 |
|---|---|---|---|
| Use 1-nearest neighbor instead of a histogram classifier on well-separated data | Non-parametric classification · Half Moons, noiseless | empirical astuteness: 0.5 -> 1 | Helped |
| Use a histogram classifier instead of 1-nearest neighbor on well-separated data | Non-parametric classification · Half Moons, noiseless | empirical astuteness: 1 -> 0.5 | Hurt |
| Apply adversarial pruning before 1-nearest-neighbor classification | 1-nearest neighbor · Half Moons, noisy | empirical astuteness: 0.8 | Helped |
| Apply adversarial pruning before histogram classification | Histogram Classifier · Half Moons, noisy | empirical astuteness: 0.7 | Helped |
| Use histogram classification rather than nearest neighbors after adversarial pruning | Adversarially pruned non-parametric classification · Half Moons, noisy | empirical astuteness: 0.8 vs 0.7 | Hurt |
| Require the kernel to decay faster than inverse polynomials and the bandwidth to approach zero | Kernel classifiers on r-separated distributions · r-separated distributions | r-consistency | Helped |
| Use a kernel that does not decay sufficiently fast | Kernel classifier · r-separated distributions | r-consistency: 0.9 | Hurt |
| Use histogram classification rather than a method satisfying the uniform local-weight condition | Non-parametric classification on r-separated data · r-separated distributions | r-consistency: 0.8 | Hurt |
| Preprocess training data with adversarial pruning before k-nearest neighbors | k-nearest neighbors · General distributions | r-consistency | Helped |
| Preprocess training data with adversarial pruning before a sufficiently fast-decaying kernel classifier | Kernel classifier · General distributions | r-consistency | Helped |
Closest papers by meaning
- Consistent Adversarially Robust Linear Classification: Non-Parametric Setting (ICML 2024 · similarity 0.90)
- Consistent Non-Parametric Methods for Maximizing Robustness (NeurIPS 2021 · similarity 0.89)
- Analyzing the Robustness of Nearest Neighbors to Adversarial Examples (ICML 2018 · similarity 0.88)
- Adversarial Robustness of Nonparametric Regression (NeurIPS 2025 · similarity 0.87)
- Robustness of classifiers: from adversarial to random noise (NeurIPS 2016 · similarity 0.86)
- Treatment of Statistical Estimation Problems in Randomized Smoothing for Adversarial Robustness (NeurIPS 2024 · similarity 0.85)
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.