| KEYWORD |
Epidemic inference in metapopulations
keywords COMPLEX NETWORKS, COMPLEX SYSTEMS, EPIDEMIC MODELING, STATISTICAL PHYSICS
Reference persons LUCA DALL'ASTA
Research Groups AA - Statistical Physics and Interdisciplinary Applications
Description The recent COVID19 pandemic has shown the importance of an effective epidemic plan, developed through prevention, surveillance, and containment methods of both pharmacological and non-pharmacological nature. Computational epidemiology combines the mathematical modeling of epidemic diffusion with the growing availability of data on human mobility in order to develop agent-based models that provide real-time predictive scenarios, useful for epidemic risk assessment on different spatial and temporal scales. In this context, the use of Bayesian inference techniques is essential for the correct integration of (real-time) epidemic data and computational models.
The thesis is focused on the development of novel algorithmic methods for epidemic inference in metapopulations (i.e. networks whose nodes represent populations of individuals), generalizing message-passing techniques recently introduced for contact networks and based on mean-field, variational methods, and neural networks.
Required skills Good preparation in mathematics and physics (in particular statistical physics and stochastic processes), and good programming skills (Julia and/or Python languages).
Deadline 02/02/2024
PROPONI LA TUA CANDIDATURA