Research map / Learning theory and algorithms / Kernels and regression theory
Debiasing Sketched Ridge Regression: A Functional Estimation Perspective
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Analyzes sketched Ridge regression as a plug-in estimator of a nonlinear covariance functional and applies iterative bootstrap and linear aggregation to reduce its bias.
Problem
Sketched Ridge regression carries a bias from the random sketch, which degrades its estimation accuracy.
Approach
Views the sketched solution as a plug-in estimator of a nonlinear function of a second-moment matrix, and applies classical bias-reduction principles (iterative bootstrap and linear aggregation of multiple estimators) to cancel low-order bias terms.
Key results
- Iterative bootstrap and linear aggregation of multiple sketched estimators cancel low-order bias terms and yield higher-order bias reduction (abstract truncated; no numeric results stated).
Limitations
- Only the abstract is available; full experimental details are not visible.
Experimental setup
Builds on: iterative bootstrap linear aggregation of estimators
Closest papers by meaning
- Asymptotically Free Sketched Ridge Ensembles: Risks, Cross-Validation, and Tuning (ICLR 2024 · similarity 0.89)
- Distributed Least Squares in Small Space via Sketching and Bias Reduction (NeurIPS 2024 · similarity 0.89)
- Fundamental Bias in Inverting Random Sampling Matrices with Application to Sub-sampled Newton (ICML 2025 · similarity 0.86)
- Optimal Sketching for Residual Error Estimation for Matrix and Vector Norms (ICLR 2024 · similarity 0.86)
- Matrix Product Sketching via Coordinated Sampling (ICLR 2025 · similarity 0.85)
- Sketchy Moment Matching: Toward Fast and Provable Data Selection for Finetuning (NeurIPS 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.