AI and the World's Hard Problems · Medicine & drug discovery · arXiv · 2026-10-22
Learning Epidemic Threshold Equations from Simulation: REEBLRA, an Interpretable Bayesian Classification Framework for Reproduction Numbers
Analytic expressions for the basic reproduction number $R_0$ are hard to derive for high-dimensional, nonlinear, or structurally complex epidemic models. REEBLRA uses Bayesian logistic regression to learn explicit surrogate threshold equations from parameter-response data, providing a binary equation for classifying epidemic invasion ($R_0>1$) and ordered threshold equations that bracket $R_0$. Across seven ODE-model benchmarks, including a 15-compartment five-group SIR system, five-fold binary AUC was 0.999-1.000 and ordered-bracket accuracy was 0.951-0.998.