WSINDy Learns Structured Population Model Components
Lyons, Dukic, and Bortz (published December 8, 2025 in PLoS Computational Biology) extend the WSINDy framework to learn structured population model ingredients from noisy time-series histogram data. The method recovers state-dependent growth, death, and birth terms, learns heterogeneous dynamics and boundary processes, and uses cross-validation for hyperparameter tuning, with tests on synthetic and real datasets showing robustness to heavy noise.
Key Points
- 1Extends WSINDy to select state-dependent growth, death, and birth functions from histogram time-series.
- 2Demonstrates robustness to heavy noise and learns heterogeneous dynamics and boundary processes.
- 3Enables practitioners to infer structured-population PDE components without repeated forward simulations.
Scoring Rationale
High novelty and peer-reviewed validation enable practical adoption, limited by niche domain scope and term distinguishability.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
