Research map / Adversarial robustness and security / Certified and empirical robustness
Adversarial Learning Guarantees for Linear Hypotheses and Neural Networks
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 |
|---|---|---|---|
| Choose an ℓ_p weight regularizer with 1/r + 1/p ≥ 1 | Linear hypotheses under ℓ_r perturbations · Adversarial Rademacher complexity | dimension dependence | Helped |
| Analyze the original network directly rather than a surrogate | One-hidden-layer neural networks · Adversarial Rademacher complexity | bound applicability | Helped |
Closest papers by meaning
- Rademacher Complexity for Adversarially Robust Generalization (ICML 2019 · similarity 0.96)
- Adversarially Robust Generalization Requires More Data (NeurIPS 2018 · similarity 0.90)
- On Achieving Optimal Adversarial Test Error (ICLR 2023 · similarity 0.90)
- On Robustness to Adversarial Examples and Polynomial Optimization (NeurIPS 2019 · similarity 0.89)
- Probably Approximately Global Robustness Certification (ICML 2025 · similarity 0.88)
- Expressive Losses for Verified Robustness via Convex Combinations (ICLR 2024 · similarity 0.88)
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.