Foreground model recognition through Neural Networks for CMB B-mode observations

In this work we present a Neural Network (NN) algorithm for the identification of the appropriate parametrization of diffuse polarized Galactic emissions in the context of Cosmic Microwave Background (CMB) $B$-mode multi-frequency observations. In particular, we focus our analysis on the low frequency polarized foregrounds represented by Galactic Synchrotron and Anomalous Microwave Emission (AME). We have implemented and tested our approach on a set of simulated maps corresponding to the frequency coverage and sensitivity represented by future satellite and low frequency ground based probes. The NN efficiency in recognizing the underlying foreground model in different sky regions reaches an accuracy above $90\%$, while the same information using a standard ${\chi }^2$ approach following parametric component separation corresponds to about 70%. Our results indicate a significant improvement when NN-based algorithms are applied to foreground model recognition in CMB $B$-mode observations, and stimulate the design and exploitation of dedicated procedures to this purpose.