Research map / Learning theory and algorithms / Sample complexity
Generalized Resubstitution for Regression Error Estimation
Diego Marcondes, Ulisses Braga-Neto
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 |
|---|---|---|---|
| smooth the empirical measure in both X and Y directions (XY-bolstering) instead of X only | Gaussian bolstered resubstitution · Synthetic polynomial regression, d = 1,2,3 | bias and RMSE: least RMSE in 39 out of 108 scenarios | Improved |
| smooth only in the X direction (X-bolstering) for complex data-generating functions and higher dimension | Gaussian bolstered resubstitution · Synthetic polynomial regression, p_g = 2,3, d >= 2 | bias and RMSE | Hurt |
| use method of moments instead of maximum pseudo-likelihood for the bolstering kernel (ablation) | X-Gaussian bolstering · Synthetic polynomial regression, d >= 2 | bias and RMSE | Hurt |
| use posterior-probability estimator (smoothing in Y) instead of other estimators | generalized resubstitution · Synthetic polynomial regression (108 scenarios) | absolute bias: least absolute bias in 83 out of 108 scenarios | Improved |
| combine Gaussian bolstering with posterior-probability estimation (ablation) | posterior-probability estimator · Synthetic polynomial regression, p_g = 2,3 | bias and RMSE | Hurt |
Closest papers by meaning
- Generalized Resubstitution for Classification Error Estimation (JMLR 2022 · similarity 0.88)
- Robust Generalized Method of Moments: A Finite Sample Viewpoint (NeurIPS 2022 · similarity 0.86)
- Beyond the Norms: Detecting Prediction Errors in Regression Models (ICML 2024 · similarity 0.85)
- The Adaptive Doubly Robust Estimator and a Paradox Concerning Logging Policy (NeurIPS 2021 · similarity 0.84)
- Prediction Risk and Estimation Risk of the Ridgeless Least Squares Estimator under General Assumptions on Regression Errors (ICLR 2025 · similarity 0.84)
- Geometry-Aware Instrumental Variable Regression (ICML 2024 · similarity 0.83)
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.
