Research map / Learning theory and algorithms / Kernels and regression theory
Understanding Randomization in Greedy Model Search
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.
Studies how feature subsampling affects greedy model search beyond the OLS setting, testing whether its effect is only variance reduction like ridge regularization.
Problem
Recent theory suggests feature subsampling in greedy model search acts only as a variance-reduction mechanism analogous to ridge regularization, but these results rely on OLS base learners.
Approach
Investigates the effects of feature subsampling on greedy forward model search with base learners beyond ordinary least squares, challenging the view that randomization only reduces variance.
Experimental setup
Builds on: random forests ridge regularization
Closest papers by meaning
- MABSplit: Faster Forest Training Using Multi-Armed Bandits (NeurIPS 2022 · similarity 0.87)
- Scale-invariant Optimal Sampling for Rare-events Data and Sparse Models (NeurIPS 2024 · similarity 0.86)
- A Provably Accurate Randomized Sampling Algorithm for Logistic Regression (AAAI 2024 · similarity 0.86)
- Towards a statistical theory of data selection under weak supervision (ICLR 2024 · similarity 0.86)
- New Classes of the Greedy-Applicable Arm Feature Distributions in the Sparse Linear Bandit Problem (AAAI 2024 · similarity 0.86)
- Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond (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.