The goal of this project is to develop applications of machine learning methods to the measurement and interpretation of the 21-cm signal from the Epoch of Reionization. Possible avenues of research are (i) the application of Machine Learning to forward or backward modeling of the signal to provide alternative parameter constraints methods, (ii) the enhancement of the modeling itself (numerical simulations) with Machine Learning, (iii) exploring the use of Machine Learning to speed up the calibration of observations, (iv) the use of unsupervised learning for signal characterization

Simulations de l’émission HI-21cm pendant l’Epoque de Réionisation (EoR). On voit ici une portion d’univers avec en rouge les zones encore neutres (HI) et en bleu les zones ionisées