ForSE: A GAN-based Algorithm for Extending CMB Foreground Models to Subdegree Angular Scales
Nicoletta Krachmalnicoff, Giuseppe Puglisi
Abstract
Abstract We present F or SE (Foreground Scale Extender), a novel Python package that aims to overcome the current limitations in the simulation of diffuse Galactic radiation, in the context of cosmic microwave background (CMB) experiments. F or SE exploits the ability of generative adversarial neural networks (GANs) to learn and reproduce complex features present in a set of images, with the goal of simulating realistic and non-Gaussian foreground radiation at subdegree angular scales. This is of great importance in order to estimate the foreground contamination to lensing reconstruction, delensing, and primordial B -modes for future CMB experiments. We applied this algorithm to Galactic thermal dust emission in both total intensity and polarization. Our results show how F or SE is able to generate small-scale features (at 12′) having as input the large-scale ones (80′). The injected structures have statistical properties, evaluated by means of the Minkowski functionals, in good agreement with those of the real sky and which show the correct amplitude scaling as a function of the angular dimension. The obtained thermal dust Stokes Q and U full-sky maps as well as the F or SE package are publicly available for download.