Each winter, Australia faces overlapping epidemics of influenza, COVID-19 and RSV that strain hospitals, especially when their peaks coincide. Knowing when and how high these epidemics will peak would let health authorities plan ahead and move resources before surges hit, but current methods forecast poorly beyond a certain time frame.

This project will develop machine-learning methods that make richer, more realistic epidemic models fast enough to use in real time. The student will train neural networks that stand in for the most demanding computations these models require, then use them to forecast respiratory disease across Australian states. The student will join the Australia-Aotearoa Consortium for Epidemic Forecasting & Analytics and the forecasts will feed into a national program reporting weekly to every state and territory health authority. The project builds transferable skills in machine learning, statistics and epidemic modelling, at the interface of mathematics and public health with direct real-world impact.

This project would suit applicants in Mathematics, Statistics or Physics. Knowledge of machine learning would be useful, but is not essential. The scholarship is open to both domestic and international students.