Research map / Learning theory and algorithms / Kernels and regression theory
Optimal Minimax Variable Selection for Large-Scale Matrix Linear Regression Model
Meiling Hao, Lianqiang Qu, Dehan Kong, Liuquan Sun, Hongtu Zhu
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 SRLS joint group-sparse selection instead of marginal SIS* screening | SIS* · Example 1, n=100, (50,50), corr 0.9 | correct fitting rate: 0.00 -> 0.85 | Improved |
| use SRLS instead of ISIS* | ISIS* · Example 1, n=100, (50,50), corr 0.9 | estimation error: 6.01 -> 0.38 | Improved |
| use SRLS instead of KAZZ | KAZZ · Example 1, n=100, (50,50), corr 0.9 | estimation error: 38.9 -> 0.38 | Improved |
| use SRLS instead of SIS* | SIS* · Example 2, n=100, p=q=50 | correct fitting rate: 0.00 -> 1.00 | Improved |
| use SRLS instead of ISIS* | ISIS* · Example 2, n=100, p=q=50 | false positives: 3.90 -> 0.00 | Improved |
| use SRLS instead of ISIS* | ISIS* · Example 3, n=100 | correct fitting rate: 0.00 -> 0.99 | Improved |
| use SRLS instead of KAZZ | KAZZ · Example 3, n=200 | true positives: 7.18 -> 13.6 | Improved |
| increase predictor correlation from 0.1 to 0.9 | SRLS · Example 1, n=100, (50,50) | estimation error: 0.05 -> 0.38 | Hurt |
| increase predictor correlation from 0.1 to 0.9 | ISIS* · Example 1, n=100, (50,50) | correct fitting rate: 0.92 -> 0.08 | Hurt |
| increase predictor correlation from 0.1 to 0.9 | SIS* · Example 1, n=100, (50,50) | estimation error: 0.64 -> 38.9 | Hurt |
| use a sparse-structure SRLS (S-SRLS) for matrix response instead of N-SRLS | SRLS · Support recovery, n=100 | support recovery: 0.76 -> 0.98 | Improved |
| use S-SRLS instead of N-SRLS | SRLS · Support recovery, n=200 | support recovery: 0.94 -> 0.99 | Improved |
Closest papers by meaning
- A convex optimization formulation for multivariate regression (NeurIPS 2020 · similarity 0.86)
- Regularized Modal Regression with Applications in Cognitive Impairment Prediction (NeurIPS 2017 · similarity 0.86)
- A Regularization-Based Adaptive Test for High-Dimensional GLMs (JMLR 2020 · similarity 0.86)
- A Comparison of Hamming Errors of Representative Variable Selection Methods (ICLR 2022 · similarity 0.85)
- A Well-Conditioned and Sparse Estimation of Covariance and Inverse Covariance Matrices Using a Joint Penalty (JMLR 2016 · similarity 0.85)
- Scalable high-dimensional Bayesian varying coefficient models with unknown within-subject covariance (JMLR 2023 · similarity 0.85)
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.
