Research map / Adversarial robustness and security / Certified and empirical robustness
Adversarial examples from computational constraints
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 a possibly exponential-time search when a robust classifier exists | Robust learning with few samples · Robustly feasible distribution families | sample complexity | Helped |
| Use a robust SQ learner for the constructed task | Statistical query learning · Constructed classification task, ℓ2 robust learning | SQ query complexity: 2^{d^{\Omega(1)}} | Hurt |
| Add a first-coordinate class marker | Hard pair of distributions · Constructed classification task, non-robust classification | learnability: polynomial time | Helped |
| Use samples to learn a robust classifier while ignoring the first coordinate | Constructed classification task · Constructed classification task, ℓ2 robustness | training samples: O(d) | Helped |
| Replace ℓ2 inputs with their Hadamard transform | Constructed classification task · Constructed classification task, ℓ∞ robustness | robustness parameter | Helped |
Closest papers by meaning
- Adversarial Examples Might be Avoidable: The Role of Data Concentration in Adversarial Robustness (NeurIPS 2023 · similarity 0.90)
- Adversarial vulnerability for any classifier (NeurIPS 2018 · similarity 0.90)
- Adversarially Robust Generalization Requires More Data (NeurIPS 2018 · similarity 0.90)
- Consistent Adversarially Robust Linear Classification: Non-Parametric Setting (ICML 2024 · similarity 0.89)
- Gradient Flow Provably Learns Robust Classifiers for Orthonormal GMMs (ICML 2025 · similarity 0.88)
- Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data (ICLR 2025 · 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.