Sommaire:

Future large galaxy surveys (such as Euclid, LSST, DESI) will probe the growth of structures and the expansion history with an unprecedented precision, offering new insights on the nature of dark matter, dark energy and gravity. However, the increased quality of these data also means a significant increase in the number of nuisance parameters, making the cosmological inference a very challenging task. Indeed, conventional likelihood-based methods like Markov-Chain Monte Carlo (MCMC) become extremely time-consuming when the dimensionality of the parameter space is very high. In this talk, I discuss the first application of Marginal Neural Ration Estimation (MNRE) (a recent approach in so-called simulation-based inference) to Euclid primary observables, like cosmic shear and galaxy-clustering spectra. Using expected Euclid experimental noise, I show how it’s possible to recover the posterior distribution for the cosmological parameters using an order of magnitude fewer simulations than MCMC. If time permits, I will end by briefly discussing another powerful application of MNRE: the reconstruction of the 3D primordial matter fluctuations from which present non-linear observations formed.

Pour plus d'informations, merci de contacter Poulin V.