Research map / Learning theory and algorithms / Matrix and tensor methods
Scaling and Scalability: Provable Nonconvex Low-Rank Tensor Estimation from Incomplete Measurements
Tian Tong, Cong Ma, Ashley Prater-Bennette, Erin Tripp, Yuejie Chi
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 |
|---|---|---|---|
| scale the gradient with the factor-wise preconditioner (ScaledGD) instead of plain GD | low-rank tensor completion · Tensor completion, n=100, r=5, p=0.1, kappa=1,2,5,10 | iterations to convergence: ScaledGD takes 17 iterations over the entire range of kappa, while GD takes thousands of iterations | Improved |
| use ScaledGD instead of GD with the step size tuned for GD | tensor completion · n=100, r=5, p=0.1, step size 0.1 to 0.9 | relative error after 80 iterations | Improved |
| random initialization instead of spectral initialization (ablation) | ScaledGD · n=100, r=5, p=0.1, kappa=1,2,5,10 | convergence behavior | No clear effect |
| random initialization instead of spectral initialization (ablation) | GD · n=100, r=5, p=0.1, kappa=1,2,5,10 | convergence behavior | Hurt |
Closest papers by meaning
- Guaranteed Nonconvex Low-Rank Tensor Estimation via Scaled Gradient Descent (JMLR 2026 · similarity 0.95)
- Nonconvex Low-Rank Tensor Completion from Noisy Data (NeurIPS 2019 · similarity 0.94)
- Uncertainty quantification for nonconvex tensor completion: Confidence intervals, heteroscedasticity and optimality (ICML 2020 · similarity 0.91)
- Score-Based Model for Low-Rank Tensor Recovery (AAAI 2026 · similarity 0.91)
- Non-Convex Tensor Recovery from Local Measurements (AAAI 2025 · similarity 0.91)
- Fast Tensor Completion via Approximate Richardson Iteration (ICML 2025 · 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.
