Research map / Time series / Causal and generative time series
When in Doubt: Neural Non-Parametric Uncertainty Quantification for Epidemic Forecasting
Harshavardhan Kamarthi, Lingkai Kong, Alexander Rodriguez, Chao Zhang, B. Aditya Prakash
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 |
|---|---|---|---|
| Use EpiFNP instead of Delta Density | US National wILI forecasting · US National wILI, 2-week ahead | RMSE: 0.6 -> 0.48 | Improved |
| Use EpiFNP instead of Gaussian Process | US National wILI forecasting · US National wILI, 2-week ahead | Log Score: 2.02 -> 0.56 | Improved |
| Use EpiFNP instead of Empirical Bayes | US National wILI forecasting · US National wILI, 4-week ahead | Calibration Score: 0.085 -> 0.035 | Improved |
| Use EpiFNP rather than Gaussian Process for autoregressive inference | US National wILI forecasting · US National wILI autoregressive inference, 2-week ahead | RMSE: 1.24 -> 0.6 | Improved |
| Add a local latent variable (ablation) | EpiFNP · US National wILI, 2-week ahead | RMSE: 0.99 -> 0.48 | Improved |
| Add a local latent variable (ablation) | EpiFNP · US National wILI, 2-week ahead | Log Score: 3.51 -> 0.51 | Improved |
| Add a global latent variable (ablation) | EpiFNP · US National wILI, 2-week ahead | RMSE: 1.76 -> 0.48 | Improved |
| Use stochastic rather than deterministic sequence embeddings (ablation) | EpiFNP · US National wILI, 2-week ahead | RMSE: 0.87 -> 0.48 | Improved |
| Add both stochastic encoder and local latent variable (ablation) | EpiFNP · US National wILI, 2-week ahead | RMSE: 1.18 -> 0.48 | Improved |
| Add both stochastic encoder and global latent variable (ablation) | EpiFNP · US National wILI, 2-week ahead | RMSE: 0.67 -> 0.48 | Improved |
| Apply post-hoc isotonic calibration | EpiFNP · US National wILI, 2-week ahead | Calibration Score: 0.068 -> 0.08 | Hurt |
Closest papers by meaning
- EARTH: Epidemiology-Aware Neural ODE with Continuous Disease Transmission Graph (ICML 2025 · similarity 0.85)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.85)
- Adaptive Sampling to Reduce Epistemic Uncertainty Using Prediction Interval-Generation Neural Networks (AAAI 2025 · similarity 0.83)
- Estimating Epistemic and Aleatoric Uncertainty with a Single Model (NeurIPS 2024 · similarity 0.83)
- Non-stationary Diffusion For Probabilistic Time Series Forecasting (ICML 2025 · similarity 0.83)
- Learning Diffusions under Uncertainty (AAAI 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.
