Research map / Learning theory and algorithms / Sample complexity
Sample Complexity of Robust Linear Classification on Separated Data
Robi Bhattacharjee, Somesh Jha, Kamalika Chaudhuri
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 |
|---|---|---|---|
| Use max-margin (hard margin SVM) algorithm on linearly separated data with aspect ratio rho instead of a robust learner | Standard linear classification on well-separated distributions · Well-separated distributions with constant aspect ratio (F_{r,rho}, rho > 20) | expected standard loss | Improved |
| Any learning algorithm trained for robust loss on well-separated data (lower bound construction) | Robust linear classification · Linearly r-separated distributions with constant aspect ratio | expected robust loss: lower bound Omega(d/n) | Hurt |
Closest papers by meaning
- Adversarially Robust Generalization Requires More Data (NeurIPS 2018 · similarity 0.91)
- Robustness of classifiers: from adversarial to random noise (NeurIPS 2016 · similarity 0.89)
- Robust Density Estimation from Batches: The Best Things in Life are (Nearly) Free (ICML 2021 · similarity 0.88)
- Probably Approximately Global Robustness Certification (ICML 2025 · similarity 0.88)
- Probabilistically Robust Learning: Balancing Average and Worst-case Performance (ICML 2022 · similarity 0.88)
- Consistent Adversarially Robust Linear Classification: Non-Parametric Setting (ICML 2024 · 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.
