Research map / Probabilistic and scientific modeling / Molecules and proteins
Error-Based Knockoffs Inference for Controlled Feature Selection
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 error-difference knockoff statistics with k-FWER control rather than FDR control | E-Knockoff · Simulated data, p=2000 | FDP_max: 0.52 -> 0.07 | Helped |
| Use error-difference statistics with FDP control rather than FDR control | E-Knockoff · Simulated data, p=2000 | FDP_max: 0.52 -> 0.13 | Helped |
| Use FDP or k-FWER control rather than FDR control | E-Knockoff · Simulated data, p=2000 | Power: 0.92 -> 0.77; 0.92 -> 0.57 | Hurt |
| Replace MX-Knockoff coefficient-difference statistics with E-Knockoff error-difference statistics | Knockoff inference · Simulated data, p=2000 | FDP_max: 0.41 -> 0.13 | Helped |
| Increase the second-part sample size n2 (ablation) | E-Knockoff (k-FWER) · Power analysis, DNN predictor | Power: 0.10 -> 1.00 | Helped |
| Use DNN instead of KRR as the predictor (ablation) | E-Knockoff (k-FWER) · Power analysis, n2=2000 | Power: 0.26 -> 1.00 | Helped |
| Use DNN instead of Lasso as the predictor (ablation) | E-Knockoff (k-FWER) · Power analysis, n2=2000 | Power: 0.94 -> 1.00 | Helped |
| Increase the training-part sample size n1 under an unknown Gaussian graphical model (ablation) | E-Knockoff · Robustness simulation | Power: 0.18 -> 1.00 | Helped |
Closest papers by meaning
- DeepDRK: Deep Dependency Regularized Knockoff for Feature Selection (NeurIPS 2024 · similarity 0.87)
- Knockoffs Inference for Partially Linear Models with Automatic Structure Discovery (AAAI 2025 · similarity 0.86)
- Statistical Test for Feature Selection Pipelines by Selective Inference (ICML 2025 · similarity 0.85)
- Multivariate Conformal Selection (ICML 2025 · similarity 0.82)
- Conformalized Multiple Testing after Data-dependent Selection (NeurIPS 2024 · similarity 0.82)
- Most Influential Subset Selection: Challenges, Promises, and Beyond (NeurIPS 2024 · similarity 0.82)
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.