Research map / Learning theory and algorithms / Matrix and tensor methods
Large-Scale Low-Rank Matrix Learning with Nonconvex Regularizers
Quanming Yao, James Tin-Yau Kwok, Taifeng Wang, Tie‐Yan Liu
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 |
|---|---|---|---|
| replace nuclear norm (AIS-Impute) with nonconvex LSP regularizer | FaNCL-acc · Synthetic matrix completion (m=2000) | NMSE (x1e-2): 3.50 -> 1.77 | Improved |
| replace fixed-rank LMaFit with nonconvex LSP regularizer | FaNCL-acc · Synthetic matrix completion (m=2000) | NMSE (x1e-2): 2.84 -> 1.77 | Improved |
| use FaNCL solver instead of GPG | LSP regularizer · Synthetic matrix completion (m=2000) | training CPU time (s): 2908.9 -> 5.5 | Improved |
| add acceleration (FaNCL-acc over FaNCL) | FaNCL with LSP · Synthetic matrix completion (m=2000) | training CPU time (s): 5.5 -> 2.4 | Improved |
| replace nuclear norm (AIS-Impute) with nonconvex regularizer | FaNCL-acc · MovieLens-1M | test RMSE: 0.819 -> 0.787 | Improved |
| replace fixed-rank LMaFit with nonconvex regularizer | FaNCL-acc · Netflix | test RMSE: 0.811 -> 0.792 | Improved |
| replace fixed-rank LMaFit with nonconvex regularizer | FaNCL-acc · Yahoo | test RMSE: 0.666 -> 0.650 | Improved |
| use FaNCL-acc instead of FaNCL on MovieLens-100K (TNN) | FaNCL with TNN · MovieLens-100K | test RMSE: 0.865 -> 0.861 | No clear effect |
| use FaNCL solver instead of GPG for LSP image inpainting | LSP · Grayscale image inpainting (tree) | CPU time (s): 195.9 -> 10.6 | Improved |
| replace nuclear norm (AIS-Impute) with LSP regularization | FaNCL-acc · Grayscale image inpainting (tree) | test RMSE: 0.432 -> 0.414 | Improved |
Closest papers by meaning
- Model-free Nonconvex Matrix Completion: Local Minima Analysis and Applications in Memory-efficient Kernel PCA (JMLR 2019 · similarity 0.90)
- Low-rank Tensor Learning with Nonconvex Overlapped Nuclear Norm Regularization (JMLR 2022 · similarity 0.90)
- Efficient Nonconvex Regularized Tensor Completion with Structure-aware Proximal Iterations (ICML 2019 · similarity 0.90)
- Structured Low-Rank Matrix Factorization: Global Optimality, Algorithms, and Applications (TPAMI 2019 · similarity 0.90)
- Iteratively Capped Reweighting Norm Minimization With Global Convergence Guarantee for Low-Rank Matrix Learning (TPAMI 2024 · similarity 0.89)
- Efficient Alternating Minimization with Applications to Weighted Low Rank Approximation (ICLR 2025 · similarity 0.87)
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.
