Aedes-AI: Neural Network Models of Mosquito Abundance

TL;DR

This paper explores neural network models as flexible, data-driven alternatives to mechanistic models for predicting mosquito abundance, aiming to improve vector control and disease risk estimation.

Contribution

It introduces neural network architectures for modeling mosquito populations and evaluates their ability to replicate mechanistic model predictions.

Findings

Neural networks can effectively mimic mechanistic mosquito models.

Augmenting training data improves model performance.

Equation-free models offer potential for scalable vector control.

Abstract

We present artificial neural networks as a feasible replacement for a mechanistic model of mosquito abundance. We develop a feed-forward neural network, a long short-term memory recurrent neural network, and a gated recurrent unit network. We evaluate the networks in their ability to replicate the spatiotemporal features of mosquito populations predicted by the mechanistic model, and discuss how augmenting the training data with time series that emphasize specific dynamical behaviors affects model performance. We conclude with an outlook on how such equation-free models may facilitate vector control or the estimation of disease risk at arbitrary spatial scales.