Research map / Time series / Causal and generative time series
Clinical Risk Prediction with Temporal Probabilistic Asymmetric Multi-Task Learning
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 |
|---|---|---|---|
| replace independent task models with joint MTL | MNIST-Variation classification · MNIST-Variation | AUROC (average): 0.6432$\pm$0.01 -> 0.6565$\pm$0.01 | Helped |
| use uncertainty-based asymmetric transfer instead of loss-based transfer | P-AMTL · MNIST-Variation | AUROC (average): 0.6345$\pm$0.01 -> 0.6778$\pm$0.01 | Helped |
| use temporal probabilistic asymmetric transfer | clinical risk prediction · MIMIC-III Infection | AUROC (average): 0.6798 $\pm$ 0.01 -> 0.7102 $\pm$ 0.01 | Helped |
| use temporal probabilistic asymmetric transfer | clinical risk prediction · PhysioNet | AUROC (average): 0.8626 $\pm$ 0.01 -> 0.8743 $\pm$ 0.01 | Helped |
| use temporal probabilistic asymmetric transfer | clinical risk prediction · MIMIC-III Heart Failure | AUROC (average): 0.7440 $\pm$ 0.01 -> 0.7345 $\pm$ 0.01 | Hurt |
| use temporal probabilistic asymmetric transfer | clinical risk prediction · MIMIC-III Respiratory Failure (4,000 instances) | AUROC (average): 0.8062 $\pm$ 0.02 -> 0.8316 $\pm$ 0.01 | Helped |
| use temporal probabilistic asymmetric transfer | clinical risk prediction · MIMIC-III Respiratory Failure (37,818 instances) | AUROC (average): 0.8726 $\pm$ 0.05 -> 0.8739 $\pm$ 0.01 | Helped |
| add inter-task and inter-timestep knowledge transfer rather than no transfer (ablation) | TP-AMTL · MIMIC-III Infection | AUROC (average): 0.6968 $\pm$ 0.01 -> 0.7119 $\pm$ 0.00 | Helped |
| add inter-task and inter-timestep knowledge transfer rather than no transfer (ablation) | TP-AMTL · PhysioNet | AUROC (average): 0.8653 $\pm$ 0.01 -> 0.8743 $\pm$ 0.01 | Helped |
| add inter-task transfer instead of restricting transfer within-task (ablation) | TP-AMTL · PhysioNet | AUROC (average): 0.8625 $\pm$ 0.01 -> 0.8743 $\pm$ 0.01 | Helped |
| add inter-timestep transfer instead of restricting transfer to the same timestep (ablation) | TP-AMTL · PhysioNet | AUROC (average): 0.8557 $\pm$ 0.01 -> 0.8743 $\pm$ 0.01 | Helped |
| add feature-level uncertainty to deterministic transfer (ablation) | TP-AMTL · PhysioNet | AUROC (average): 0.8133 $\pm$ 0.02 -> 0.8743 $\pm$ 0.01 | Helped |
Closest papers by meaning
- Learning Implicit Tasks for Patient-Specific Risk Modeling in ICU (AAAI 2017 · similarity 0.87)
- Multi-task learning with summary statistics (NeurIPS 2023 · similarity 0.86)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · similarity 0.83)
- XTSFormer: Cross-Temporal-Scale Transformer for Irregular-Time Event Prediction in Clinical Applications (AAAI 2025 · similarity 0.83)
- Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning (AAAI 2025 · similarity 0.83)
- Automated Multi-Task Learning for Joint Disease Prediction on Electronic Health Records (NeurIPS 2024 · similarity 0.83)
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.