Research map / Time series / Causal and generative time series
Scalable Joint Models for Reliable Uncertainty-Aware Event Prediction
Hossein Soleimani, James J. Hensman, Suchi Saria
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 |
|---|---|---|---|
| train sparse multi-output GP joint model end-to-end so longitudinal uncertainty reaches the event model, instead of two-stage MoGP (ablation) | J-LTM · MIMIC-II septic shock | AUC: 0.79 -> 0.84 | Improved |
| use J-LTM instead of B-spline two-stage joint model JM | event predictor · MIMIC-II septic shock | AUC: 0.78 -> 0.84 | Improved |
| use J-LTM instead of logistic regression on binned features | event predictor · MIMIC-II septic shock | AUC: 0.80 -> 0.84 | Improved |
| use J-LTM instead of RNN on binned series | event predictor · MIMIC-II septic shock | TPR at FPR=0.2: 0.59 -> 0.75 | Improved |
| use J-LTM instead of MoGP at FPR 0.2 | event predictor · MIMIC-II septic shock | TPR at FPR=0.2: 0.62 -> 0.75 | Improved |
| use J-LTM instead of LR at FPR 0.2 | event predictor · MIMIC-II septic shock | TPR at FPR=0.2: 0.57 -> 0.75 | Improved |
| use robust uncertainty-aware policy with abstention instead of point-estimate policy | J-LTM · MIMIC-II septic shock | PPV at TPR 0.4-0.6: 13%-23% PPV gain over MoGP | Improved |
| use J-LTM instead of JM, PPV | event predictor · MIMIC-II septic shock | PPV at TPR 0.4-0.6: 18%-26% PPV gain over JM | Improved |
| use J-LTM instead of LR, PPV | event predictor · MIMIC-II septic shock | PPV at TPR 0.4-0.6: 31%-36% PPV gain over LR | Improved |
Closest papers by meaning
- BayOTIDE: Bayesian Online Multivariate Time Series Imputation with Functional Decomposition (ICML 2024 · similarity 0.87)
- Bayesian Temporal Factorization for Multidimensional Time Series Prediction (TPAMI 2021 · similarity 0.86)
- Diffusion Transformers for Imputation: Statistical Efficiency and Uncertainty Quantification (NeurIPS 2025 · similarity 0.84)
- S4M: S4 for multivariate time series forecasting with Missing values (ICLR 2025 · similarity 0.84)
- Frequency-aware Generative Models for Multivariate Time Series Imputation (NeurIPS 2024 · similarity 0.84)
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training (NeurIPS 2026 · similarity 0.84)
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.
