Research map / Probabilistic and scientific modeling / Molecules and proteins
Variable Selection via Penalized Neural Network: a Drop-Out-One Loss Approach
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 Drop-Out-One loss selection (GEPNN) instead of Spinn weight shrinkage | penalized neural network · Simulation case 1 | MSPE: 6.68(0.76) -> 2.34(0.25) | Helped |
| Use GEPNN instead of Spinn | penalized neural network · Simulation case 1 | number of selected variables: 8.2(3.66) -> 5.2(0.4) | Helped |
| Use GEPNN instead of Spinn for group selection | penalized neural network · Simulation case 2 | MSPE: 6.62(1.49) -> 4.20(0.80) | Helped |
| Use GEPNN instead of Spinn for group selection | penalized neural network · Simulation case 2 | number of selected groups: 9.83(6.81) -> 3.2(0.42) | Helped |
| Use GEPNN instead of Spinn | penalized neural network · CCLE | number of selected variables: 28.9(7.71) -> 7.3(2.26) | Helped |
| Use GEPNN instead of l1-NN | penalized neural network · CCLE | MSPE: 1.20(0.13) -> 1.19(0.061) | No clear effect |
| Use GEPNN instead of Spinn | penalized neural network · CCLE | MSPE: 1.24(0.17) -> 1.19(0.061) | Helped |
| Use GEPNN instead of BART-50 | tree ensemble baseline · Simulation case 1 | number of selected variables: 9.6(3.06) -> 5.2(0.4) | Helped |
| Use GEPNN instead of SIS-SCAD linear selector | linear baseline · Simulation case 1 | MSPE: 9.87(1.80) -> 2.34(0.25) | Helped |
| Use GEPNN instead of LR-OGL overlapping group lasso | linear group baseline · Simulation case 2 | MSPE: 8.16(3.73) -> 4.20(0.80) | Helped |
| Rank variables by Drop-Out-One loss instead of squared l2 norm of Spinn weights (ablation) | variable importance criterion · Simulation case 1, p=100,200,500 | gap between relevant and irrelevant variables | Helped |
| Vary lambda1 (ablation) | GEPNN versus Spinn · Variable selection | sensitivity to lambda1 | Helped |
Closest papers by meaning
- Out-of-Variable Generalisation for Discriminative Models (ICLR 2024 · similarity 0.83)
- Differentially Private High-dimensional Variable Selection via Integer Programming (NeurIPS 2025 · similarity 0.82)
- High-Dimensional Analysis for Generalized Nonlinear Regression: From Asymptotics to Algorithm (AAAI 2024 · similarity 0.82)
- Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large $p$ (ICML 2025 · similarity 0.82)
- Regression Learning with Limited Observations of Multivariate Outcomes and Features (ICML 2024 · similarity 0.82)
- Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond (ICML 2024 · similarity 0.81)
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.