ABC-RF-Rejection: A two-stage machine-learning-enhanced framework for efficient likelihood-free inference

Accurate parameter estimation is fundamental to quantitative epidemiology as it provides thefoundation for robust modelling and evidence-based decision-making. We present a novel twostageframework, that is designed to enhance computational efficiency for parameter estimation ofcomplex, stochastic, or spatially explicit models. Traditional Approximate Bayesian Computation(ABC) methods often face prohibitive costs when likelihoods are analytically intractable andacceptance rates are low. Our hybrid approach, ABC-RF-rejection, integrates ABC rejectionsampling with Random Forest (RF) classification to selectively identify parameter sets likely tosatisfy observed data constraints. In the first stage, a small-scale ABC rejection step generatesa labelled training dataset of accepted and rejected particles. In the second stage, a trainedRF model decouples posterior exploration from expensive forward simulations by predictingacceptance probabilities for a substantially larger candidate set. This allows the algorithm tofocus computational resources on high-probability regions of the parameter space. We apply theABC-RF-rejection approach to three distinct epidemiological contexts: stochastic simulations ofonchocerciasis vector control, spatially explicit modelling of cassava brown streak virus spreadin Uganda, and the 2014–2015 West Africa Ebola outbreak in heterogeneous populations. Thisframework provides an adaptable and robust solution for rapid, evidence-based decision-makingin settings characterized by spatial heterogeneity, stochasticity, and limited surveillance data.