Missing Data Imputation under Manifold Hypothesis
About
The manifold hypothesis posits that high-dimensional data are concentrated near a low-dimensional embedded manifold. Recent advances in mixture variational autoencoders (VAEs) provide a powerful tool for extracting such underlying structure in a faithful manner. The resulting geometric structure naturally introduces local and global relationships among variables, thereby providing a systematic way of imputing missing data. We propose a model-based imputation method that enables sampling from \( p(\bm{x}_{\mathrm{mis}} \mid \bm{x}_{\mathrm{obs}}) \) via a sampling-importance-resampling (SIR) procedure, which can be further augmented with a joint diffusion model in the latent space. Our method imputes missing data while respecting the underlying geometry, achieves competitive performance compared to state-of-the-art procedures, quantifies uncertainty in the imputations, and is model-based, thereby enabling on-the-fly imputation without rerunning the entire procedure.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Regression | Superconduct | RMSE0.2661 | 228 | |
| Data Imputation | Superconductivity | Wasserstein Distance0.6334 | 216 | |
| Data Imputation | WINE (test) | RMSE1.0436 | 205 | |
| Imputation | facialexpression | RMSE0.0667 | 195 | |
| Imputation | Superconductivity (MCAR) | RMSE0.2153 | 72 | |
| Imputation | Superconductivity (MAR) | RMSE0.2154 | 72 | |
| Imputation | Superconductivity MNAR | RMSE0.2724 | 70 | |
| Imputation | powerplant MAR | RMSE0.8511 | 67 | |
| Data Imputation | powerplant MAR (test) | Wasserstein Distance0.3143 | 66 | |
| Imputation | Facial expression dataset (MNAR) | Wasserstein Distance0.5063 | 66 |