Research map / Learning theory and algorithms / Matrix and tensor methods
MARS: A Second-Order Reduction Algorithm for High-Dimensional Sparse Precision Matrices Estimation
Qian Li, Binyan Jiang, Defeng Sun
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 adaptive sieving reduction strategy to SSNAL (i.e. MARS) (ablation) | SSNAL · Simulated Model 1 (p=2000, n=50) | relative KKT residual: 6.87E-05 -> 3.56E-05 | Improved |
| add adaptive sieving reduction strategy to SSNAL (i.e. MARS) (ablation) | SSNAL · Simulated Model 2 (p=2000, n=50) | relative KKT residual: 5.74E-05 -> 3.85E-05 | Improved |
| add adaptive sieving reduction strategy to SSNAL (i.e. MARS) (ablation) | SSNAL · Simulated Model 1 solution path (p=2000, n=50) | relative KKT residual: 2.60E-05 -> 9.75E-06 | Improved |
| tighten MARS stopping tolerance from 1e-4 to 1e-8 (ablation) | MARS · Simulated Model 1 solution path (p=2000, n=50) | relative KKT residual: 9.75E-06 -> 1.59E-09 | Improved |
| loosen EQUAL stopping tolerance from 1e-6 to 1e-5 (ablation) | EQUAL · prostate data, control group, lambda=0.99 | relative KKT residual: 5.05e-02 -> 5.16e-01 | Hurt |
| loosen EQUAL stopping tolerance from 1e-6 to 1e-5 (ablation) | EQUAL · prostate data, control group, lambda=0.99 | objective value: 924.68 -> 276912.70 | Hurt |
| use MARS instead of glasso | precision matrix estimation · Simulated Model 4 (lambda=0.35) | Frobenius norm error: 14.4898 -> 12.8997 | Improved |
| use MARS instead of glasso | precision matrix estimation · Simulated Model 4 (lambda=0.26) | Frobenius norm error: 12.8643 -> 19.5567 | Hurt |
| use MARS instead of glasso | precision matrix estimation · Simulated Model 1 (lambda=0.39) | true positive rate: 0.2735 -> 0.2078 | Mixed |
Closest papers by meaning
- Covariate Adjusted Precision Matrix Estimation via Nonconvex Optimization (ICML 2018 · similarity 0.88)
- A Completely Tuning-Free and Robust Approach to Sparse Precision Matrix Estimation (ICML 2022 · similarity 0.88)
- Meta Learning for Support Recovery in High-dimensional Precision Matrix Estimation (ICML 2021 · similarity 0.87)
- Guarantees of a Preconditioned Subgradient Algorithm for Overparameterized Asymmetric Low-rank Matrix Recovery (ICML 2025 · similarity 0.87)
- Non-Asymptotic Analysis Of Data Augmentation For Precision Matrix Estimation (NeurIPS 2025 · similarity 0.86)
- Sparse Polyak: an adaptive step size rule for high-dimensional M-estimation (NeurIPS 2025 · 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.
