Math Modelling Versus Machine Learning for COVID-19

Which models drive decision-making and policy?

Lawrence Fishburne and Bryan Cranston in Contagion, a film describing the outbreak of a virus MEV-1. Image source: Vanity Fair (Fair Use)

When COVID-19 swept the world in early 2020, researchers swarmed in with their modelling expertise to forecast epidemic spread and derive optimum interventions. Here’s a high-level view of the whole party.

The majority of mathematical models are derived from the SIR and SEIR compartment models. The primary use cases are population-level forecasting (e.g. predict timing of epidemic peak and hospitalisation numbers) and informing interventions strategies (e.g. lockdowns, quarantine, social distancing and wearing masks).

The majority of Machine Learning (ML) models are Convolutional Neural Networks (CNN) addressing a variety of challenges, from diagnosing patients through CT images and tracking epidemic spread through mobile phones, to designing molecules in vaccines and building AI-robots that disinfect hospitals.

For reference, here’s a classification of model approaches from the prequel article here.

New to AI and ML? Check out my explainer article.