XIIIth School of Cosmology
November 12 - 18, 2017IESC, Cargèse
The CMB from A to Z
promises and challenges of the CMB as a cosmological probe

CosmoML: A Machine Learning method to measure the cosmological parameters

Martın de los Rios
Universidad Nacional de Cordoba

Short Talk

Abstract:In this talk we will present CosmoML, a machine learning algorithm that estimate the cosmological parameters using the cosmic microwave background (CMB).
This algorithm was trained using 5000 simulated power spectra of different cosmological models.It has the advantage of been faster than traditional Monte Carlo methods without loosing precision.
As a first application we also present a measurement of the angular distribution of the cosmological parameters performed with the latest CMB data from Planck.