Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Impute Missing Entries with Uncertainty
Jaesung Lim, Seunghwan An, Jong-June Jeon
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 |
|---|---|---|---|
| use U-VAE instead of MICE for imputation | MICE · MNARQ 0.3 missingness | Bias: .011(3.6) -> .005(1.4) | Improved |
| use U-VAE instead of GAIN for multiple imputation | GAIN · MNARL 0.3 missingness | Bias: up to 13.8% vs 2.9% | Improved |
| use U-VAE instead of MICE | MICE · MNARL 0.3 missingness | Coverage: .595 -> .839 | Improved |
| use U-VAE instead of MIWAE | MIWAE · MCAR 0.3 missingness | Coverage: .769 -> .867 | Improved |
| use U-VAE instead of VAEAC | VAEAC · MCAR 0.3 missingness | SMAPE: .216 -> .209 | Improved |
| use U-VAE instead of VAEAC | VAEAC · MAR 0.3 missingness | downstream accuracy: .765 -> .761 | Hurt |
| use U-VAE instead of MICE | MICE · MAR 0.3 missingness | feature selection consistency: .756 -> .747 | Hurt |
| use U-VAE instead of missForest | missForest · Congeniality anuran | L1 coefficient deviation: .357 -> .338 | Improved |
| use U-VAE instead of Mean imputation | Mean · Congeniality default | L2 coefficient deviation: .166 -> .158 | Improved |
| raise missingness rate from 0.1 to 0.9 | U-VAE vs not-MIWAE, GAIN, MICE · MCAR/MAR/MNAR | Bias | Improved |
Closest papers by meaning
- Learning What Not to Impute: An Uncertainty-Aware Diffusion Framework for Meaningful Missingness (NeurIPS 2026 · similarity 0.91)
- Distribution Shift in Missing Data Imputation: A Risk-Based Perspective and Importance-Weighted Correction under MAR (NeurIPS 2026 · similarity 0.90)
- Rethinking the Diffusion Models for Missing Data Imputation: A Gradient Flow Perspective (NeurIPS 2024 · similarity 0.88)
- DiffPuter: Empowering Diffusion Models for Missing Data Imputation (ICLR 2025 · similarity 0.87)
- Diffusion Transformers for Imputation: Statistical Efficiency and Uncertainty Quantification (NeurIPS 2025 · similarity 0.87)
- Missing Data Imputation by Reducing Mutual Information with Rectified Flows (NeurIPS 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.
