Research map / Learning theory and algorithms / Prediction and decision losses
Multilinear Regression for Embedded Feature Selection with Application to fMRI Analysis
Xiaonan Song, Haiping Lu
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 |
|---|---|---|---|
| add l1-norm sparsity to nuclear-norm-only multilinear regression (ablation) | Remurs · CMU2008 fMRI average accuracy | accuracy: 78.15 -> 78.15 | No clear effect |
| add l1-norm sparsity (feature selection) to nuclear norm (ablation) | Remurs · CMU2008 fMRI | number of features: 20,640 -> 4,549 features | Improved |
| add l1-norm to nuclear norm (ablation) | Remurs · Synthetic data | prediction RMSE | Improved |
| add tensor nuclear norm (multilinear structure) to l1 regression | Lasso/ENet · CMU2008 fMRI average accuracy | accuracy: 75.46 -> 78.15 | Improved |
| add tensor nuclear norm to l1 regression | Lasso/ENet · Synthetic data, severe small sample size | prediction RMSE | Improved |
| use l1 + nuclear norm mixture as feature size grows | Remurs · Synthetic data, feature size | prediction RMSE stability | Improved |
| Remurs vs Remurs gamma=0 on individual subjects (ablation) | Remurs · CMU2008 fMRI per subject | accuracy: 79.17 -> 72.50 (subject 6); 73.33 vs 76.67 (subject 9) | Mixed |
| use Remurs instead of ENet for selected voxels spatial coherence | Remurs · CMU2008 fMRI subject 1 | connected components (fewer is more coherent): 276 -> 208 | Improved |
| fixed-rank CP multilinear regression vs nuclear norm | MMR/orTRR · CMU2008 fMRI average accuracy | accuracy: 78.15 -> 56.11 (MMR), 68.33 (orTRR) | Hurt |
Closest papers by meaning
- Adaptive Feature Selection With Augmented Attributes (TPAMI 2023 · similarity 0.85)
- Latent Sparse Modeling of Longitudinal Multi-Dimensional Data (AAAI 2018 · similarity 0.85)
- E2E-FS: An End-to-End Feature Selection Method for Neural Networks (TPAMI 2022 · similarity 0.84)
- A Probabilistic Derivation of LASSO and L12-Norm Feature Selections (AAAI 2019 · similarity 0.84)
- Optimal Minimax Variable Selection for Large-Scale Matrix Linear Regression Model (JMLR 2021 · similarity 0.83)
- Functional Linear Regression with Mixed Predictors (JMLR 2022 · 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.
