Research map / Time series / Causal and generative time series
Learning from Irregularly-Sampled Time Series: A Missing Data Perspective
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 continuous P-BiGAN instead of Latent ODE | MIMIC-III mortality prediction · MIMIC-III | AUC (%): 85.71 ± 0.38 -> 86.05 ± 0.36 | No clear effect |
| use continuous P-BiGAN instead of Latent ODE | MIMIC-III mortality prediction · MIMIC-III | training time per epoch (hours): 2.62 -> 0.22 | Helped |
| use continuous P-VAE instead of Latent ODE | MIMIC-III mortality prediction · MIMIC-III | training time per epoch (hours): 2.62 -> 0.05 | Helped |
| add continuous convolutional features without generative modeling (ablation) | MIMIC-III mortality prediction · MIMIC-III | AUC (%): 83.88 ± 0.65 -> 84.87 ± 0.18 | Helped |
| add autoencoding regularization (λ=1) (ablation) | Cont P-BiGAN · MIMIC-III | AUC (%): 83.56 ± 0.49 -> 86.05 ± 0.36 | Helped |
| replace MisGAN with P-VAE | image imputation · MNIST and CelebA | FID | Mixed |
| replace MisGAN with P-BiGAN | image imputation · MNIST and CelebA | FID | Mixed |
| add autoencoding regularization to P-BiGAN (ablation) | image imputation · MNIST and CelebA | FID | Helped |
| use a continuous convolutional encoder instead of the recurrent Latent ODE architecture | irregularly-sampled time series classification · MIMIC-III | AUC (%): 85.71 ± 0.38 -> 86.05 ± 0.36 | No clear effect |
Closest papers by meaning
- Interpolation-Prediction Networks for Irregularly Sampled Time Series (ICLR 2019 · similarity 0.89)
- A Diffusion Model for Regular Time Series Generation from Irregular Data with Completion and Masking (NeurIPS 2025 · similarity 0.86)
- Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning (AAAI 2025 · similarity 0.85)
- Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs (ICLR 2024 · similarity 0.85)
- GraFITi: Graphs for Forecasting Irregularly Sampled Time Series (AAAI 2024 · similarity 0.84)
- TimeCHEAT: A Channel Harmony Strategy for Irregularly Sampled Multivariate Time Series Analysis (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.