Research map / Time series / Causal and generative time series
Probabilistic Imputation for Time-series Classification with Missing Data
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 |
|---|---|---|---|
| add obsdropout (dropping observed values during classifier training) (ablation) | supnotMIWAE classifier · MIMIC-III | AUROC: 0.860 vs 0.849 | Helped |
| add obsdropout (ablation) | supnotMIWAE classifier · PhysioNet 2012 | AUROC: 0.874 vs 0.866 | Helped |
| add obsdropout (ablation) | supnotMIWAE classifier · Human Activity Recognition | Accuracy: 0.883 vs 0.867 | Helped |
| add MNAR assumption (not-MIWAE style missing model) (ablation) | supnotMIWAE · MIMIC-III | AUROC: 0.860 vs 0.851 | Helped |
| add MNAR assumption (ablation) | supnotMIWAE · PhysioNet 2012 | AUROC: 0.874 vs 0.869 (MNAR helps slightly) | Hurt |
| add MNAR assumption (ablation) | supnotMIWAE · Human Activity Recognition | Accuracy: 0.883 vs 0.882 | No clear effect |
| add obsdropout and supervision in imputation setting (ablation) | supnotMIWAE imputation · PhysioNet 2012 | MAE: 0.367 vs 0.377 (w/o obsdropout) | Helped |
| remove supervision term from training (generative only) (ablation) | supnotMIWAE imputation · MIMIC-III | MAE: 0.149 vs 0.148 | No clear effect |
| proposed model (supnotMIWAE + obsdropout) vs GRU-based imputation baselines | classification · MIMIC-III | AUROC: 0.860 vs 0.856 (best GRU baseline, GRU-Forward) | Helped |
| proposed model vs GRU-based imputation baselines | classification · Human Activity Recognition | Accuracy: 0.883 vs 0.861 (GRU-Simple) | Helped |
| proposed model vs SAITS and GP-VAE imputation | imputation · PhysioNet 2012 | MAE: 0.367 vs 0.653 (SAITS) | Helped |
Closest papers by meaning
- Frequency-aware Generative Models for Multivariate Time Series Imputation (NeurIPS 2024 · similarity 0.91)
- Diffusion Transformers for Imputation: Statistical Efficiency and Uncertainty Quantification (NeurIPS 2025 · similarity 0.87)
- BayOTIDE: Bayesian Online Multivariate Time Series Imputation with Functional Decomposition (ICML 2024 · similarity 0.87)
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training (NeurIPS 2026 · similarity 0.87)
- Self-attention-based Diffusion Model for Time-series Imputation in Partial Blackout Scenarios (AAAI 2025 · similarity 0.86)
- S4M: S4 for multivariate time series forecasting with Missing values (ICLR 2025 · 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.