Deep21: a deep learning method for 21 cm foreground removal. Issue 4 (30th April 2021)

• Record Type: Journal Article
• Authors: Makinen, T. Lucas; Lancaster, Lachlan; Villaescusa-Navarro, Francisco; Melchior, Peter; Ho, Shirley; Perreault-Levasseur, Laurence; Spergel, David N.
• Is Part Of: Journal of cosmology and astroparticle physics. Volume 2021:Issue 4(2021)
• Publication Date: 2021-04-30
• Subjects: cosmological simulations -- redshift surveys -- power spectrum -- reionization; Cosmology -- Periodicals; Astrophysics -- Periodicals; 523.0105
• DOI: 10.1088/1475-7516/2021/04/081
• Languages: English
• ISSNs: 1475-7516

Abstract

Abstract: We seek to remove foreground contaminants from 21 cm intensity mapping observations. We demonstrate that a deep convolutional neural network (CNN) with a UNet architecture and three-dimensional convolutions, trained on simulated observations, can effectively separate frequency and spatial patterns of the cosmic neutral hydrogen (HI) signal from foregrounds in the presence of noise. Cleaned maps recover cosmological clustering amplitude and phase within 20% at all relevant angular scales and frequencies. This amounts to a reduction in prediction variance of over an order of magnitude across angular scales, and improved accuracy for intermediate radial scales (0.025 < k∥ < 0.075 h Mpc -1 ) compared to standard Principal Component Analysis (PCA) methods. We estimate epistemic confidence intervals for the network's prediction by training an ensemble of UNets. Our approach demonstrates the feasibility of analyzing 21 cm intensity maps, as opposed to derived summary statistics, for upcoming radio experiments, as long as the simulated foreground model is sufficiently realistic. We provide the code used for this analysis on GitHub, as well as a browser-based tutorial for the experiment and UNet model via the accompanying Colab notebook .