Research map / Learning theory and algorithms / Matrix and tensor methods
Fast and Sample Efficient Inductive Matrix Completion via Multi-Phase Procrustes Flow
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 Multi-Phase Procrustes Flow instead of Maxide | Inductive matrix completion · NUS-WIDE-OBJECT, p = 10% | averaged AP: 0.5349 (0.0034) -> 0.5434 (0.0040) | Helped |
| Use Multi-Phase Procrustes Flow instead of Maxide | Inductive matrix completion · NUS-WIDE-OBJECT, p = 25% | averaged AP: 0.5562 (0.0021) -> 0.5677 (0.0027) | Helped |
| Use Multi-Phase Procrustes Flow instead of Maxide | Inductive matrix completion · NUS-WIDE-OBJECT, p = 50% | averaged AP: 0.5629 (0.0023) -> 0.5718 (0.0023) | Helped |
| Use a new projected-gradient phase with fresh samples at each update, followed by vanilla gradient descent | Procrustes Flow for inductive matrix completion · Inductive matrix completion | sample complexity: O(r3n2 log n log(1/ϵ)) -> O(r2n log n log d) | Helped |
| Add side information and use the inductive formulation rather than standard matrix completion | Matrix completion · Synthetic datasets | relative reconstruction error | Helped |
| Increase the assumed rank | Multi-Phase Procrustes Flow · OMIM gene-disease prediction | percentage of hidden genes found | Helped |
Closest papers by meaning
- Fine-grained Generalization Analysis of Inductive Matrix Completion (NeurIPS 2021 · similarity 0.92)
- Inductive Matrix Completion: No Bad Local Minima and a Fast Algorithm (ICML 2022 · similarity 0.92)
- Generalization Bounds for Inductive Matrix Completion in Low-Noise Settings (AAAI 2023 · similarity 0.91)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.90)
- Generalization Analysis of Deep Non-linear Matrix Completion (ICML 2024 · similarity 0.88)
- Semi-Random Matrix Completion via Flow-Based Adaptive Reweighting (NeurIPS 2024 · 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.