Research map / Learning theory and algorithms / Matrix and tensor methods
A Note on Entrywise Consistency for Mixed-data Matrix Completion
Yunxiao Chen, Xiaoou Li
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 |
|---|---|---|---|
| apply one-step refinement (Algorithm 1) to NBE | NBE · MovieLens (rank 3) | test-set log-likelihood: -56091 -> -49284 | Improved |
| apply one-step refinement (Algorithm 1) to NBE | NBE · MovieLens (rank 1) | test-set log-likelihood: -48928 -> -49247 | Hurt |
| apply refinement Algorithm 1 to NBE | NBE · PISA 2018 (rank 3) | test-set log-likelihood: -75816 -> -70092 | Improved |
| apply refinement Algorithm 1 to NBE | NBE · PISA 2018 (rank 1) | test-set log-likelihood: -67205 -> -67938 | Hurt |
| apply data-splitting refinement Algorithm 3 to CJMLE | CJMLE · MovieLens (rank 3) | test-set log-likelihood: -49022 -> -48207 | Improved |
| apply data-splitting refinement Algorithm 3 to CJMLE | CJMLE · MovieLens (rank 4) | test-set log-likelihood: -51192 -> -48271 | Improved |
| apply refinement Algorithm 1 to CJMLE | CJMLE · PISA 2018 (rank 2) | test-set log-likelihood: -63250 -> -63313 | No clear effect |
| apply data-splitting refinement Algorithm 3 to CJMLE | CJMLE · PISA 2018 (rank 2) | test-set log-likelihood: -63250 -> -64842 | Hurt |
| apply data-splitting refinement Algorithm 3 to CJMLE | CJMLE · PISA 2018 (rank 4) | test-set log-likelihood: -72320 -> -75989 | Hurt |
| raise rank from 3 to 2 for CJMLE | CJMLE · PISA 2018 | test-set log-likelihood: -65476 -> -63250 | Improved |
| use Algorithm 1 instead of Algorithm 2 refinement on NBE | NBE refinement · Simulation settings 1-6 (scaled Frobenius and max norm) | scaled Frobenius norm and max norm error | Improved |
| aggregate five data splits (Algorithm 3 instead of Algorithm 2) | NBE/CJMLE refinement · Simulation settings 1-6 | scaled Frobenius norm and max norm error | Improved |
Closest papers by meaning
- Missing Not at Random in Matrix Completion: The Effectiveness of Estimating Missingness Probabilities Under a Low Nuclear Norm Assumption (NeurIPS 2019 · similarity 0.89)
- A Pairwise Pseudo-likelihood Approach for Matrix Completion with Informative Missingness (NeurIPS 2024 · similarity 0.88)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.86)
- Fast exact recovery of noisy matrix from few entries: the infinity norm approach (NeurIPS 2025 · similarity 0.86)
- Semi-Random Matrix Completion via Flow-Based Adaptive Reweighting (NeurIPS 2024 · similarity 0.85)
- Generalization Analysis of Deep Non-linear Matrix Completion (ICML 2024 · similarity 0.85)
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.
