Speaker
Description
Data-driven modelling and optimisation play a fundamental role in science and engineering. In particular, hybrid models, i.e., those that combine physical models (white-box modelling) with data-driven machine learning approaches (black-box modelling), have great potential for simulating and controlling vector-borne diseases, including dengue virus, yellow fever virus, Chikungunya virus, and Zika virus. Ordinary differential equation (ODE)-based models are standard for simulating vector-borne diseases, while spatio-temporal partial differential equation (PDE)-based models and stochastic differential equation (SDE)-based models allow for more accurate simulations. In this work, we propose a new stochastic partial differential equation (SPDE)-based model and use it as the white-box component in our machine learning approach for simulation and control of vector-borne diseases. Furthermore, we introduce a novel feedback control approach within this framework.
