Cosmic Microwave Background Recovery: A Graph-Based Bayesian Convolutional Network Approach
This addresses the challenge of improving CMB data quality for cosmological parameter estimation in astrophysics, representing an incremental advance by applying a novel deep learning method to a known bottleneck.
The paper tackled the problem of cleaning cosmic microwave background (CMB) data from foreground contamination using a graph-based Bayesian convolutional neural network, demonstrating accurate recovery of cleaned CMB maps and angular power spectra on simulated Planck data with pixel-wise uncertainty estimates.
The cosmic microwave background (CMB) is a significant source of knowledge about the origin and evolution of our universe. However, observations of the CMB are contaminated by foreground emissions, obscuring the CMB signal and reducing its efficacy in constraining cosmological parameters. We employ deep learning as a data-driven approach to CMB cleaning from multi-frequency full-sky maps. In particular, we develop a graph-based Bayesian convolutional neural network based on the U-Net architecture that predicts cleaned CMB with pixel-wise uncertainty estimates. We demonstrate the potential of this technique on realistic simulated data based on the Planck mission. We show that our model accurately recovers the cleaned CMB sky map and resulting angular power spectrum while identifying regions of uncertainty. Finally, we discuss the current challenges and the path forward for deploying our model for CMB recovery on real observations.