Research map / Learning theory and algorithms / Prediction and decision losses
Consistent Robust Regression
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 |
|---|---|---|---|
| Replace subset selection (TORRENT) with iterative hard thresholding on the corruption vector using the full data set (CRR) | robust least squares regression · synthetic Gaussian regression with corruptions | CPU time: up to 5x faster than TORRENT-FC | Helped |
| Use CRR (hard thresholding on corruptions) instead of extended Lasso | robust least squares regression · synthetic Gaussian regression with corruptions | CPU time: up to 20x faster | Helped |
| Use CRR with k >= 2k* corruption index on Gaussian designs | robust least squares regression · theoretical consistency | L2 estimation error | Helped |
| Increase number of data points n with CRR (from 500 to 5000) (ablation) | CRR · synthetic, n=5000 d=100 k=0.37n | convergence speed of ||b_t-b*|| | Helped |
| Increase CRR iterations (iterative correction of corruption estimate) (ablation) | CRR · synthetic, n=2000 d=500 k=0.37n | fraction of corruptions identified | Helped |
Closest papers by meaning
- Robust Sparse Regression with Non-Isotropic Designs (NeurIPS 2024 · similarity 0.88)
- Adversarial Robustness of Nonparametric Regression (NeurIPS 2025 · 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)
- Beyond the Norms: Detecting Prediction Errors in Regression Models (ICML 2024 · similarity 0.85)
- Robust Sparse Estimation for Gaussians with Optimal Error under Huber Contamination (ICML 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.