Research map / Learning theory and algorithms / Sample complexity
Consistent regression when oblivious outliers overwhelm
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 the Huber loss estimator under Gaussian design with oblivious outliers | Robust linear regression · Gaussian design | squared parameter error: n ≳ (d/α)^2 -> n = ω(d/α^2) | Helped |
| Apply Huber loss estimation to designs whose column spans are well-spread | Robust linear regression · General deterministic design | squared prediction error | Helped |
| Use coordinate-wise medians with iterative bootstrapping | Robust linear regression · Gaussian design | squared parameter error: O((d/(α^2 n))·log d) | Helped |
| Use the median-based algorithm when the parameter is k-sparse | Robust linear regression · Sparse regression, Gaussian design | squared parameter error: O((k/(α^2 n))·log d) | Helped |
| Require a well-spread column span for the general-design Huber guarantee | Huber loss estimator · General deterministic design | squared prediction error | Mixed |
Closest papers by meaning
- Robust Sparse Regression with Non-Isotropic Designs (NeurIPS 2024 · similarity 0.92)
- Information-Computation Tradeoffs for Noiseless Linear Regression with Oblivious Contamination (NeurIPS 2025 · similarity 0.87)
- Robust Sparse Estimation for Gaussians with Optimal Error under Huber Contamination (ICML 2024 · similarity 0.86)
- $H$-Consistency Guarantees for Regression (ICML 2024 · similarity 0.86)
- Robustness Auditing for Linear Regression: To Singularity and Beyond (ICLR 2025 · similarity 0.85)
- All-Purpose Mean Estimation over R: Optimal Sub-Gaussianity with Outlier Robustness and Low Moments Performance (ICML 2025 · 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.