Research map / Learning theory and algorithms / Kernels and regression theory
ENNS: Variable Selection, Regression, Classification, and Deep Neural Network for High-Dimensional Data
Kaixu Yang, Arkaprabha Ganguli, Tapabrata Maiti
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 |
|---|---|---|---|
| use ENNS ensemble selection instead of pure stage-wise DNP | DNP · Simulation regression, s=10 | false positive rate: 41.4% -> 12.8% | Improved |
| use ENNS ensemble selection instead of pure stage-wise DNP | DNP · Simulation classification, s=10 | false positive rate: 40.5% -> 9.8% | Improved |
| use ENNS ensemble selection instead of pure stage-wise DNP | DNP · Simulation regression, s=2 | false positive rate: 22.5% -> 10.4% | Improved |
| use ENNS instead of DNP for selection (low signal) | DNP · Simulation classification, s=10 | avg correct variables selected: 5.95 -> 8.04 | Improved |
| use ENNS instead of LASSO for selection (low signal) | LASSO · Simulation regression, s=10 | avg correct variables selected: 9.62 -> 9.25 | No clear effect |
| use ENNS instead of HSIC-LASSO for selection (low signal) | HSIC-LASSO · Simulation classification, s=10 | avg correct variables selected: 3.61 -> 8.04 | Improved |
| raise signal strength from normal to high | ENNS · Simulation classification, s=10 | avg correct variables selected: 8.38 -> 9.90 | Improved |
| raise signal strength from normal to high | HSIC-LASSO · Simulation classification, s=10 | avg correct variables selected: 4.95 -> 7.76 | Improved |
| use ENNS + l1 neural network instead of LASSO linear model | LASSO · Riboflavin | test MSE: 0.286 -> 0.268 | Improved |
| use ENNS + l1 neural network instead of regularized neural network | Regularized neural network · Prostate cancer | test accuracy: 0.955 -> 0.956 | No clear effect |
| use ENNS + l1 neural network instead of GAM with group lasso | GAM with group lasso · Prostate cancer | test accuracy: 0.918 -> 0.956 | Improved |
| add more correct variables already in the stage-wise model (0 to 4) | stage-wise neural network selection · Linear regression simulation | probability next variable correct: 0.995 -> 0.430 | Hurt |
Closest papers by meaning
- Variable Selection via Penalized Neural Network: a Drop-Out-One Loss Approach (ICML 2018 · similarity 0.89)
- LassoNet: A Neural Network with Feature Sparsity (JMLR 2021 · similarity 0.85)
- Model-free Variable Selection in Reproducing Kernel Hilbert Space (JMLR 2016 · similarity 0.84)
- SAND: One-Shot Feature Selection with Additive Noise Distortion (ICML 2025 · similarity 0.83)
- Repeated Random Sampling for Minimizing the Time-to-Accuracy of Learning (ICLR 2024 · similarity 0.83)
- Neural Networks for Learnable and Scalable Influence Estimation of Instruction Fine-Tuning Data (NeurIPS 2025 · 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.
