Deep learning for intensity mapping observations: Component extraction

Abstract

Line intensity mapping (LIM) is an emerging observational method to study the large-scale structure of the Universe and its evolution. LIM does not resolve individual sources but probes the fluctuations of integrated line emissions. A serious limitation with LIM is that contributions of different emission lines from sources at different redshifts are all confused at an observed wavelength. We propose a deep learning application to solve this problem. We use conditional generative adversarial networks to extract designated information from LIM. We consider a simple case with two populations of emission line galaxies; H$\alpha$ emitting galaxies at $z=1.3$ are confused with [OIII] emitters at $z=2.0$ in a single observed waveband at 1.5 $\mu$m. Our networks trained with 30,000 mock observation maps are able to extract the total intensity and the spatial distribution of H$\alpha$ emitting galaxies at $z=1.3$. The intensity peaks are successfully located with 74% precision. The precision increases to 91% when we combine the results of 5 networks. The mean intensity and the power spectrum are reconstructed with an accuracy of $\sim$10%. The extracted galaxy distributions at a wider range of redshift can be used for studies on cosmology and on galaxy formation and evolution.

arXiv comment6 pages, 3 figures, accepted for publication in MNRAS Letter

References in corpus (22)

• Adam: A Method for Stochastic Optimization
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• 21-cm cosmology
• A large Hα survey at z=2.23, 1.47, 0.84 & 0.40: the 11 Gyr evolution of star-forming galaxies from HiZELS
• Line-Intensity Mapping: 2017 Status Report
• Measuring the 3D Clustering of Undetected Galaxies Through Cross Correlation of their Cumulative Flux Fluctuations from Multiple Spectral Lines
• An accurate tool for the fast generation of dark matter halo catalogs
• Evolution of the H$\beta$+[OIII] and [OII] luminosity functions and the [OII] star-formation history of the Universe up to $z$ ~ 5 from HiZELS
• Deep learning from 21-cm tomography of the Cosmic Dawn and Reionization
• Constraining the astrophysics and cosmology from 21cm tomography using deep learning with the SKA
• Spectral Line De-confusion in an Intensity Mapping Survey
• Cosmology with intensity mapping techniques using atomic and molecular lines
• Foreground contamination in Ly-alpha intensity mapping during the epoch of reionization
• Identifying Reionization Sources from 21cm Maps using Convolutional Neural Networks
• The distribution and physical properties of high-redshift [OIII] emitters in a cosmological hydrodynamics simulation
• Science Impacts of the SPHEREx All-Sky Optical to Near-Infrared Spectral Survey: Report of a Community Workshop Examining Extragalactic, Galactic, Stellar and Planetary Science
• Denoising Weak Lensing Mass Maps with Deep Learning
• HIGAN: Cosmic Neutral Hydrogen with Generative Adversarial Networks
• Tomographic Intensity Mapping versus Galaxy Surveys: Observing the Universe in H-alpha emission with new generation instruments
• CDIM: Cosmic Dawn Intensity Mapper Final Report
• Generative Single Image Reflection Separation
• Deconfusing intensity maps with neural networks

Cited by in corpus (16)

• CCAT-prime Collaboration: Science Goals and Forecasts with Prime-Cam on the Fred Young Submillimeter Telescope
• Probing Cosmic Reionization and Molecular Gas Growth with TIME
• Phase-Space Spectral Line De-confusion in Intensity Mapping
• Neutrino Properties with Ground-Based Millimeter-Wavelength Line Intensity Mapping
• Machine Learning for Observational Cosmology
• [CII] line intensity mapping the epoch of reionization with the Prime-Cam on FYST
• Deep Learning for Line Intensity Mapping Observations: Information Extraction from Noisy Maps
• Forecasts and Statistical Insights for Line Intensity Mapping Cross-Correlations: A Case Study with 21cm x [CII]
• A multi-tracer empirically-driven approach to line-intensity mapping lightcones
• FERIA: Flat Envelope Model with Rotation and Infall under Angular Momentum Conservation
• Deep learning reconstruction of three-dimensional galaxy distributions with intensity mapping observations
• Predicting 21cm-line map from Lyman $\alpha$ emitter distribution with Generative Adversarial Networks
• Bayesian Multi-line Intensity Mapping
• Leveraging cross-correlations and linear covariance-based filtering for line-intensity map reconstructions at linear scales
• Removal of interloper contamination to line-intensity maps using correlations with ancillary tracers of the large-scale structure
• CosmoGLINT: Cosmological Generative Model for Line Intensity Mapping with Transformer