Research map / Learning theory and algorithms / Prediction and decision losses
On the Consistency of Feature Selection With Lasso for Non-linear Targets
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 |
|---|---|---|---|
| Lasso fit with a linear hypothesis on data generated by nonlinear link functions (polynomial, sine, mixed) with Gaussian features | Lasso feature selection · synthetic nonlinear targets, Gaussian X, n up to 4000 | probability of successful relevant feature selection | Helped |
| Increase sample size n for Lasso under nonlinear link (polynomial g1 with m=3,5) versus linear case | Lasso feature selection · synthetic, w in R^100 with 10 nonzeros | sample size needed for near-certain selection | Hurt |
| Use a false link function g4(u)=1/u that violates the expectation assumption on g' (ablation) | Lasso feature selection · synthetic, Gaussian X with Sigma=I | selection success probability | Hurt |
| Replace Gaussian features with Beta(2,2) or Student-t(nu=1) distributed features (ablation) | Lasso feature selection with g5 · synthetic, g5 target | selection consistency | Hurt |
| Use uniform or Weibull(1,1.5) features instead of Gaussian (ablation) | Lasso feature selection with g5 · synthetic, g5 target | selection success probability | No clear effect |
Closest papers by meaning
- A Probabilistic Derivation of LASSO and L12-Norm Feature Selections (AAAI 2019 · similarity 0.89)
- Optimal Sparse Recovery with Decision Stumps (AAAI 2023 · similarity 0.87)
- Post-selection inference with HSIC-Lasso (ICML 2021 · similarity 0.86)
- A Comparison of Hamming Errors of Representative Variable Selection Methods (ICLR 2022 · similarity 0.86)
- Feature Selection using Stochastic Gates (ICML 2020 · similarity 0.86)
- Best Subset Selection: Optimal Pursuit for Feature Selection and Elimination (ICML 2025 · similarity 0.84)
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.