Research map / Learning theory and algorithms / Matrix and tensor methods
Local Search Algorithms for Rank-Constrained Convex Optimization
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 |
|---|---|---|---|
| Use Fast Local Search instead of Fast Greedy | Rank-constrained matrix completion · random_error_13 | normalized MSE: $0.0673/30$ -> $0.0613/14$ | Helped |
| Use Fast Local Search instead of Fast Greedy | Rank-constrained matrix completion · random_error_1 | normalized MSE: $0.1948/13$ -> $0.1952/15$ | Hurt |
| Use Fast Local Search instead of Fast Greedy | Rank-constrained matrix completion · random_error_2 | normalized MSE: $0.1826/21$ -> $0.1811/15$ | Helped |
| Use Fast Greedy instead of SoftImpute | Rank-constrained matrix completion · random_error_13 | normalized MSE: $0.1759/10$ -> $0.0673/30$ | Helped |
| Use Fast Greedy instead of SoftImpute | Rank-constrained matrix completion · random_error_1 | normalized MSE: $0.2465/28$ -> $0.1948/13$ | Helped |
| Use Fast Greedy instead of SoftImpute | Rank-constrained matrix completion · random_error_2 | normalized MSE: $0.2284/30$ -> $0.1826/21$ | Helped |
| Use Fast Greedy instead of the strongest baseline | Movie recommendation · MovieLens 100K | test RMSE: $0.9355$ -> $0.9451$ | Hurt |
| Use Fast Greedy instead of the strongest baseline | Movie recommendation · MovieLens 1M | test RMSE: $0.8732$ -> $0.8714$ | Helped |
| Use Fast Greedy instead of the strongest baseline | Movie recommendation · MovieLens 10M | test RMSE: $0.8315$ -> $0.8330$ | Hurt |
| Use Fast Greedy instead of basic Greedy | Rank-30 matrix factorization · MovieLens 100K | runtime (seconds): $418.4$ -> $43.4$ | Helped |
| Use Fast Greedy instead of basic Greedy | Rank-30 matrix factorization · MovieLens 1M | runtime (seconds): $4087.3$ -> $244.2$ | Helped |
| Run only a few inner-optimization iterations (ablation) | Fast Greedy and Fast Local Search in machine-learning applications · Matrix completion generalization | test error | Mixed |
Closest papers by meaning
- On Approximation Guarantees for Greedy Low Rank Optimization (ICML 2017 · similarity 0.89)
- No Spurious Local Minima in Nonconvex Low Rank Problems: A Unified Geometric Analysis (ICML 2017 · similarity 0.89)
- General Low-rank Matrix Optimization: Geometric Analysis and Sharper Bounds (NeurIPS 2021 · similarity 0.89)
- Faster Accelerated First-order Methods for Convex Optimization with Strongly Convex Function Constraints (NeurIPS 2024 · similarity 0.88)
- Single-Loop Stochastic Algorithms for Difference of Max-Structured Weakly Convex Functions (NeurIPS 2024 · similarity 0.87)
- From Linear to Linearizable Optimization: A Novel Framework with Applications to Stationary and Non-stationary DR-submodular Optimization (NeurIPS 2024 · 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.