In collaboration with Boris Hejblum, Sébastien Rouillon, Rodolphe Thiébaut, Pierre-Yves Oudeyer, Clément Moulin-Frier & Mélanie Prague.

Epidemiologists model the dynamics of epidemics in order to propose mitigation strategies based on pharmaceutical and non-pharmaceutical interventions (contact limitation, lock down, vaccination, etc.). Hand-designing such strategies is not trivial because of the number of possible interventions and the difficulty to predict their long-term effects. This task can be seen as an optimization problem where state-of-the-art machine learning algorithms might bring significant value.

This website presents an interactive demo of a set of machine learning methods applied to the automatic design of a lock-down intervention strategy in the context of the COVID19 epidemic in the French Ile-de-France region. Technical details can be found in the associated paper: EpidemiOptim: A Toolbox for the Optimization of Control Policies in Epidemiological Models. The full code of this toolbox is open-source and available on github here.

Cite this blog post

@article{colas2020epidemioptim,
    title={ {E}pidemiOptim: {A} {T}oolbox for the {O}ptimization of {C}ontrol {P}olicies in {E}pidemiological Models},
    author={Colas, C{\'e}dric and Hejblum, Boris and Rouillon, S{\'e}bastien and Thi{\'e}baut, Rodolphe and Oudeyer, Pierre-Yves and Moulin-Frier, Cl{\'e}ment and Prague, M{\'e}lanie},
    journal={Journal of Artificial Intelligence Research},
    year={2020}}