Research map / Learning theory and algorithms / Matrix and tensor methods
Multivariate Distributionally Robust Convex Regression under Absolute Error Loss
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 distributionally robust absolute-loss convex regression (DRCR) instead of least squares convex regression (LSE) | convex regression · EPA air market data, test | l1 test error: 0.1516 -> 0.1294 | Helped |
| use DRCR instead of linear regression | convex regression · EPA air market data, test | l1 test error: 0.1692 -> 0.1294 | Helped |
| use DRCR instead of LSE and kernel regression | convex regression · synthetic, d=5, Gaussian covariates | empirical l1 and l2 loss | Helped |
| heavy-tailed (Student t, 10 dof) covariates violating the moment assumption | DRCR · synthetic, d=5 | empirical l1 and l2 loss | No clear effect |
| vary the gradient bound constant c (0.8 vs 10) for LSE | LSE convex regression · synthetic, d=5 | empirical l1 and l2 loss | Mixed |
Closest papers by meaning
- High-dimensional Robust Mean Estimation via Gradient Descent (ICML 2020 · similarity 0.88)
- Adversarial Robustness of Nonparametric Regression (NeurIPS 2025 · similarity 0.87)
- Robust Sparse Regression with Non-Isotropic Designs (NeurIPS 2024 · similarity 0.86)
- Risk Bounds For Distributional Regression (NeurIPS 2025 · similarity 0.86)
- Minimax Adaptive Online Nonparametric Regression over Besov spaces (NeurIPS 2025 · similarity 0.86)
- Regularized least squares learning with heavy-tailed noise is minimax optimal (NeurIPS 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.