Research map / Time series / Causal and generative time series
Interpolation-Prediction Networks for Irregularly Sampled Time Series
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 |
|---|---|---|---|
| Use semi-parametric interpolation with three channels (smooth, transient, intensity) before prediction | GRU prediction network · MIMIC-III mortality | AUPRC | Helped |
| Use semi-parametric interpolation with three channels before prediction | LSTM prediction network · MIMIC-III mortality benchmark | AUPRC: 0.4933 -> 0.5370 | Helped |
| Add smooth interpolation and intensity to transients (ablation) | Interpolation-Prediction Network · MIMIC-III mortality | AUC: 0.705 \pm 0.009 -> 0.853 \pm 0.007 | Helped |
| Add smooth interpolation and intensity to transients (ablation) | Interpolation-Prediction Network · MIMIC-III length of stay | MedAE: 2.995\pm0.130 -> 2.862\pm0.166 | Helped |
| Add transients to smooth interpolation and intensity (ablation) | Interpolation-Prediction Network · MIMIC-III mortality | AUC: 0.852 \pm 0.005 -> 0.853 \pm 0.007 | No clear effect |
| Add transients to smooth interpolation and intensity (ablation) | Interpolation-Prediction Network · MIMIC-III mortality | cross-entropy loss: 0.210 \pm 0.004 -> 0.210 \pm 0.004 | No clear effect |
| Combine smooth interpolation and intensity rather than using either alone (ablation) | Interpolation-Prediction Network · MIMIC-III mortality | AUC: 0.816 \pm 0.009 / 0.786\pm 0.010 -> 0.852 \pm 0.005 | Helped |
| Use intensity alone instead of all three channels (ablation) | Interpolation-Prediction Network · MIMIC-III length of stay | MedAE: 2.862\pm0.166 -> 2.697\pm0.072 | Helped |
| Use intensity and transients without smooth interpolation (ablation) | Interpolation-Prediction Network · MIMIC-III length of stay | EV score: 0.245\pm0.019 -> 0.290\pm0.010 | Helped |
Closest papers by meaning
- Hi-Patch: Hierarchical Patch GNN for Irregular Multivariate Time Series (ICML 2025 · similarity 0.85)
- Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach (ICML 2024 · similarity 0.84)
- GraFITi: Graphs for Forecasting Irregularly Sampled Time Series (AAAI 2024 · similarity 0.84)
- TimeDART: A Diffusion Autoregressive Transformer for Self-Supervised Time Series Representation (ICML 2025 · similarity 0.84)
- STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction (ICLR 2024 · similarity 0.84)
- MPTSNet: Integrating Multiscale Periodic Local Patterns and Global Dependencies for Multivariate Time Series Classification (AAAI 2025 · 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.