Filling in Cosmic Microwave Background map missing regions via Generative Adversarial Networks

Abstract

In this work, we propose a new method to fill in (or in-paint) the CMB signal in regions masked out following a point source extraction process. We adopt a modified Generative Adversarial Network (GAN) and compare different combinations of internal (hyper-)parameters and training strategies. We study the performance using a suitable 𝒞r variable in order to estimate the performance regarding the CMB power spectrum recovery. We consider a test set where one point source is masked out in each sky patch with a 1.83 × 1.83 squared degree extension, which, in our gridding, corresponds to 64 × 64 pixels. The GAN is optimized for estimating performance on Planck 2018 total intensity simulations. The training makes the GAN effective in reconstructing a masking corresponding to about 1500 pixels with 1% error down to angular scales corresponding to about 5 arcminutes. Finally, we show that GAN is able to mimic PDF and number density of peaks for both Gaussian and non-Gaussian data with less than 0.5% relative error.