Research map / Probabilistic and scientific modeling / Variational and generative sampling
Sobolev Independence Criterion
Youssef Mroueh, Tom Sercu, Mattia Rigotti, Inkit Padhi, Cicero Nogueira dos Santos
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 |
|---|---|---|---|
| Select features by SIC importance scores rather than Elastic Net, then train an NN regressor | CCLE PLX4720 feature selection · CCLE PLX4720 | heldout MSE: Elastic-Net2 top-10: 0.663 $\pm$ 0.161; SIC top-10: 0.706 $\pm$ 0.158 | Mixed |
| Select top 10 features by SIC importance rather than Elastic-Net2 top-10 | CCLE PLX4720 feature selection with RF regressor · CCLE PLX4720 | heldout MSE: 0.830 $\pm$ 0.190 -> 0.817 $\pm$ 0.173 | Improved |
| Use Boosted SIC importance scores instead of GLM importance scores with knockoffs | HIV-1 Protease Inhibitors · HIV-1 PIs | FDP: ATV: 0.26 -> 0.05; IDV: 0.38 -> 0.16; LPV: 0.05 -> 0.12; SQV: 0.19 -> 0.36 | Mixed |
| Use Boosted SIC importance scores instead of GLM importance scores with knockoffs | HIV-1 NRTIs · HIV-1 NRTIs | FDP: ABC: 0.09 -> 0.08; AZT: 0.2 -> 0.29; D4T: 0.14 -> 0 | Mixed |
| Use Boosted SIC importance scores instead of GLM importance scores with knockoffs | HIV-1 NNRTIs · HIV-1 NNRTIs | FDP: DLV: 0.56 -> 0.55; EFV: 0.5 -> 0.47; NVP: 0.58 -> 0.611 | Mixed |
| Use SIC importance scores to select top 7 features rather than Elastic-Net1 top-7 | CCLE PLX4720 feature selection with NN regressor · CCLE PLX4720 | heldout MSE: 0.864 $\pm$ 0.432 -> 0.728 $\pm$ 0.166 | Improved |
Closest papers by meaning
- Differentially Private High-dimensional Variable Selection via Integer Programming (NeurIPS 2025 · similarity 0.83)
- Scalable Sobolev IPM for Probability Measures on a Graph (ICML 2025 · similarity 0.83)
- Optimality and Adaptivity of Deep Neural Features for Instrumental Variable Regression (ICLR 2025 · similarity 0.82)
- The optimality of kernel classifiers in Sobolev space (ICLR 2024 · similarity 0.82)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.82)
- Efficiently Learning Significant Fourier Feature Pairs for Statistical Independence Testing (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.
