Recovering CMB polarization maps with neural networks: Performance in realistic simulations

Recovering the polarized cosmic microwave background (CMB) is essential for shedding light on the exponential expansion of the very early Universe, known as cosmic inflation. Achieving this goal requires not only improved instrumental sensitivity but also the development of robust and diverse data analysis techniques. In this work, we explore a novel component separation approach based on neural networks to reconstruct the Stokes $Q$ and $U$ polarization maps. To validate the method, we first test the network on realistic \textit{Planck} sky simulations, finding a mean absolute error of $0.1 \pm 0.3~μK^{2}$ for the $E$-mode and $-0.1 \pm 0.3~μK^{2}$ for the $B$-mode. We then apply the trained network to public \textit{Planck} observations, with results consistent with those obtained using the Commander method. Based on these findings, we conclude that neural network-based methods show potential as component separation techniques in polarization CMB experiments. However, substantial improvements and more comprehensive analyses are necessary before these methods can provide reliable high-precision cosmological estimates.