2021/11/01 by Niall Jeffrey, François Boulanger, Benjamin D Wandelt +5
Physics and Astronomy · #Convolutional neural network #Cosmic microwave background #Cosmology and Gravitation Theories #Galaxies: Formation, Evolution, Phenomena #Inference #Pattern recognition (psychology) #Polarization (electrochemistry) #Radio Astronomy Observations and Technology #Robustness (evolution) #Sky #Wavelet #astro-ph.CO #astro-ph.IM
paper · pdf · doi:10.1093/mnrasl/slab120
Accepted by Monthly Notices of the Royal Astronomical Society Letters. 5 pages with 3 figures (plus 1 page of Supporting Materials with 2 figures)
arxiv created 2021/11/01 · openalex publication_date 2021/11/05 · openalex created_date 2021/11/08 · arxiv updated 2021/11/17 · openalex updated_date 2026/08/05
ABSTRACT With a single training image and using wavelet phase harmonic augmentation, we present polarized Cosmic Microwave Background (CMB) foreground marginalization in a high-dimensional likelihood-free (Bayesian) framework. We demonstrate robust foreground removal using only a single frequency of simulated data for a BICEP-like sky patch. Using Moment Networks, we estimate the pixel-level posterior probability for the underlying E, B signal and validate the statistical model with a quantile-type test using the estimated marginal posterior moments. The Moment Networks use a hierarchy of U-Net convolutional neural networks. This work validates such an approach in the most difficult limiting case: pixel-level, noise-free, highly non-Gaussian dust foregrounds with a single training image at a single frequency. For a real CMB experiment, a small number of representative sky patches would provide the training data required for full cosmological inference. These results enable robust likelihood-free, simulation-based parameter, and model inference for primordial B-mode detection using observed CMB polarization data.