Research map / Learning theory and algorithms / Matrix and tensor methods
Efficient Nonconvex Regularized Tensor Completion with Structure-aware Proximal Iterations
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 |
|---|---|---|---|
| adaptive momentum added to the sparse-plus-low-rank proximal average solver (NORT vs sNORT) (ablation) | NORT · synthetic small I3 and large I3 | CPU time: 27.9 s to 5.9 s (small I3); 575.9 s to 89.4 s (large I3) | Helped |
| sparse-plus-low-rank structure exploited in proximal steps (sNORT vs GDPAN) (ablation) | proximal average solver · synthetic small I3 and large I3 | CPU time and space: 665.4 s to 27.9 s CPU (small I3); 14.0 MB vs 171.5 MB space | Helped |
| use D=3 modes instead of D=2 on tensors with large I3 (ablation) | NORT · synthetic large I3 | RMSE: 0.0009 (D=2) vs 0.0006 (D=3) for capped-l1 | Helped |
| use D=3 modes instead of D=2 on tensors with small I3 (ablation) | NORT · synthetic small I3 | CPU time: 5.9 s (D=2) vs 918.7 s (D=3) for capped-l1 | Hurt |
| nonconvex regularizer (NORT) vs convex overlapped nuclear norm (PA-APG) | tensor completion · synthetic small I3 and large I3 | RMSE: 0.0149 to 0.0103 (small I3); 0.0098 to 0.0006 (large I3) | Helped |
| nonconvex regularizer with NORT vs convex and factorization methods on color images | tensor completion · rice, tree, windows color images | RMSE: rice 0.468 vs 0.576 (FaLRTC); windows 0.297 vs 0.388 (RP) | Helped |
| NORT vs other methods on remote sensing hyperspectral data | tensor completion · Cabbage, Scene, Female | RMSE: Cabbage 3.76 vs 4.91 (RP); Female 5.92 vs 6.47 (RP) | Helped |
| NORT vs baselines on YouTube social network data | tensor completion · YouTube subset and full set | RMSE: subset 0.482 vs 0.522 (RP); full 0.370 vs 0.410 (RP) | Helped |
Closest papers by meaning
- Tensor Completion Made Practical (NeurIPS 2020 · similarity 0.93)
- Fast Tensor Completion via Approximate Richardson Iteration (ICML 2025 · similarity 0.92)
- Efficient Sparse Low-Rank Tensor Completion Using the Frank-Wolfe Algorithm (AAAI 2017 · similarity 0.91)
- Nonconvex Low-Rank Tensor Completion from Noisy Data (NeurIPS 2019 · similarity 0.91)
- A Dual Framework for Low-rank Tensor Completion (NeurIPS 2018 · similarity 0.91)
- Tensor Ring Decomposition with Rank Minimization on Latent Space: An Efficient Approach for Tensor Completion (AAAI 2019 · similarity 0.90)
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.