Research map / Probabilistic and scientific modeling / Dynamical systems
EINNs: Epidemiologically-Informed Neural Networks
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 |
|---|---|---|---|
| Transfer ODE dynamics to the feature RNN using gradient matching (ablation) | EINNs · COVID-19 US National + 47 states, 1–8 weeks | Pearson correlation (PC): 0.03 -> 0.46 | Helped |
| Transfer ODE dynamics to the feature RNN using gradient matching (ablation) | EINNs · Influenza, 10 HHS regions, 1–8 weeks | Pearson correlation (PC): 0.50 -> 0.68 | Helped |
| Train the time-module PINN without the feature module (ablation) | EINNs · COVID-19 US National + 47 states, 1–8 weeks | Pearson correlation (PC): 0.53 -> 0.24 | Hurt |
| Train the time-module PINN without the feature module (ablation) | EINNs · Influenza, 10 HHS regions, 1–8 weeks | Pearson correlation (PC): 0.68 -> -0.47 | Hurt |
| Use EINNs rather than Generation | COVID-19 forecasting · US National + 47 states, 1–4 weeks | NR1: 0.79 -> 0.54 | Helped |
| Use EINNs rather than Generation | COVID-19 forecasting · US National + 47 states, 5–8 weeks | NR1: 0.93 -> 0.85 | Helped |
| Use EINNs rather than RNN (GRU+attention) | Influenza forecasting · 10 HHS regions, 1–4 weeks | NR1: 0.72 -> 0.53 | Helped |
| Use EINNs rather than Ensembling | Influenza forecasting · 10 HHS regions, 5–8 weeks | NR1: 0.83 -> 1.01 | Hurt |
| Increase ODE-loss weight from 1 to 10 (ablation) | EINNs · COVID-19, five-region hyperparameter sensitivity evaluation | Pearson correlation (PC): 0.68 -> 0.72 | Helped |
| Increase ODE-loss weight from 10 to 100 (ablation) | EINNs · COVID-19, five-region hyperparameter sensitivity evaluation | Pearson correlation (PC): 0.72 -> 0.55 | Hurt |
| Increase transfer-loss weight from 1 to 100 (ablation) | EINNs · COVID-19, five-region hyperparameter sensitivity evaluation | Pearson correlation (PC): 0.72 -> 0.76 | Helped |
Closest papers by meaning
- EARTH: Epidemiology-Aware Neural ODE with Continuous Disease Transmission Graph (ICML 2025 · similarity 0.88)
- Dynamics-Based Feature Augmentation of Graph Neural Networks for Variant Emergence Prediction (AAAI 2025 · similarity 0.85)
- Learning Diffusions under Uncertainty (AAAI 2024 · similarity 0.82)
- Observable Neural ODEs for Identifiable Causal Forecasting in Continuous Time (NeurIPS 2026 · similarity 0.82)
- SNN-PDE: Learning Dynamic PDEs from Data with Simplicial Neural Networks (AAAI 2024 · similarity 0.82)
- Zero-shot forecasting of chaotic systems (ICLR 2025 · similarity 0.82)
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.