Inpainting CMB maps using Partial Convolutional Neural Networks

TL;DR

This paper introduces a novel partial convolutional neural network approach for inpainting masked cosmic microwave background maps, achieving high accuracy in reconstructing maps and power spectra, especially useful for irregular masks from astrophysical sources.

Contribution

The paper demonstrates that PCNNs can effectively inpaint CMB maps with various mask shapes, maintaining high fidelity in both maps and power spectra, advancing cosmological data analysis techniques.

Findings

Reconstructed maps and spectra are indistinguishable from original at 99.9% confidence.

Effective inpainting for masks covering up to 10% of the map area.

PCNNs perform equally well on regular and irregular masks.

Abstract

We present a novel application of partial convolutional neural networks (PCNN) that can inpaint masked images of the cosmic microwave background. The network can reconstruct both the maps and the power spectra to a few percent for circular and irregularly shaped masks covering up to ~10% of the image area. By performing a Kolmogorov-Smirnov test we show that the reconstructed maps and power spectra are indistinguishable from the input maps and power spectra at the 99.9% level. Moreover, we show that PCNNs can inpaint maps with regular and irregular masks to the same accuracy. This should be particularly beneficial to inpaint irregular masks for the CMB that come from astrophysical sources such as galactic foregrounds. The proof of concept application shown in this paper shows that PCNNs can be an important tool in data analysis pipelines in cosmology.