Research map / Learning theory and algorithms / Matrix and tensor methods
Polynomial Matrix Completion for Missing Data Imputation and Transductive Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Mice protein, θ=10% | classification error (%): 0.8 -> 0.41 (PMC-S); 0.8 -> 0.39 (PMC-W) | Helped |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Mice protein, θ=50% | classification error (%): 5.26 -> 0.71 (PMC-S); 5.26 -> 0.63 (PMC-W) | Helped |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Shuttle, θ=10% | classification error (%): 2.48 -> 2.66 (PMC-S); 2.48 -> 3.86 (PMC-W) | Hurt |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Shuttle, θ=50% | classification error (%): 9.58 -> 8.02 (PMC-S); 9.58 -> 9.16 (PMC-W) | Helped |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Dermatology, θ=10% | classification error (%): 3.28 -> 2.84 (PMC-S); 3.28 -> 2.84 (PMC-W) | Helped |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Dermatology, θ=50% | classification error (%): 8.31 -> 8.16 (PMC-S); 8.31 -> 7.98 (PMC-W) | Helped |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Satimage, θ=10% | classification error (%): 14.38 -> 13.06 (PMC-S); 14.38 -> 14.24 (PMC-W) | Mixed |
| Minimize feature-space rank using PMC-S or PMC-W instead of the strongest listed baseline | Transductive classification · Satimage, θ=50% | classification error (%): 16.96 -> 14.82 (PMC-S); 16.96 -> 15.18 (PMC-W) | Helped |
| Penalize only the smallest feature-space singular values (PMC-S) rather than all singular values (PMC-R1/NLMC) (ablation) | PMC · Synthetic data; subspace clustering; motion capture recovery | recovery or clustering error | Helped |
| Use increasing weights on feature-space singular values (PMC-W) rather than the unweighted small-singular-value penalty (PMC-S) (ablation) | PMC · Synthetic multiple-manifold data | RSE | Helped |
| Use the simple increasing weight heuristic w₂ instead of weights inversely proportional to initial singular values w₁ (ablation) | PMC-W · Matrix completion experiments | recovery performance | No clear effect |
Closest papers by meaning
- Online High Rank Matrix Completion (CVPR 2019 · similarity 0.91)
- High-Rank Matrix Completion and Clustering under Self-Expressive Models (NeurIPS 2016 · similarity 0.89)
- A Pairwise Pseudo-likelihood Approach for Matrix Completion with Informative Missingness (NeurIPS 2024 · similarity 0.87)
- Fast exact recovery of noisy matrix from few entries: the infinity norm approach (NeurIPS 2025 · similarity 0.87)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.86)
- RGNMR: A Gauss-Newton method for robust matrix completion with theoretical guarantees (NeurIPS 2025 · similarity 0.86)
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.