Research map / Learning theory and algorithms / Matrix and tensor methods
Column Thresholding for Sparse Spiked Wigner Models: Improved Signal Strength Requirements
Jian-Feng Cai, Zhuozhi Xian, Jiaxi Ying
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 support from the largest-diagonal column instead of diagonal entries (column vs diagonal thresholding) (ablation) | sparse spiked Wigner recovery · Three-peak benchmark (p=5000, beta=100) | estimation error: 0.68 -> 0.14 | Improved |
| refine column thresholding output with truncated power method (ablation) | column thresholding · Three-peak benchmark (p=5000, beta=100) | estimation error: 0.14 -> 0.10 | Improved |
| use TPM instead of diagonal thresholding at weak signal strength | TPM · d=2000, s=10, varying beta | estimation error and support recovery | Improved |
| normalization variant instead of eigenvector reconstruction in column thresholding (ablation) | column thresholding · theory | required signal strength | No clear effect |
| sparsity scaling s = d^0.4 (impossible/hard boundary) vs d^0.6 (ablation) | column thresholding · support recovery, psi=0.2 | success rate: near 0 at s=d^0.4 -> 1 at s=d^0.6 | Mixed |
Closest papers by meaning
- The Price of Sparsity: Sufficient Conditions for Sparse Recovery using Sparse and Sparsified Measurements (NeurIPS 2025 · similarity 0.86)
- Spectral Phase Transition and Optimal PCA in Block-Structured Spiked Models (ICML 2024 · similarity 0.86)
- Fast and Provable Algorithms for Sparse PCA with Improved Sample Complexity (ICML 2025 · similarity 0.86)
- Straight-Through Meets Sparse Recovery: the Support Exploration Algorithm (ICML 2024 · similarity 0.85)
- Gradient Compressed Sensing: A Query-Efficient Gradient Estimator for High-Dimensional Zeroth-Order Optimization (ICML 2024 · similarity 0.84)
- Nonlinear Laplacians: Tunable principal component analysis under directional prior information (NeurIPS 2025 · similarity 0.84)
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.
