Removing Astrophysics in 21 cm maps with Neural Networks

Abstract

Measuring temperature fluctuations in the 21 cm signal from the Epoch of Reionization and the Cosmic Dawn is one of the most promising ways to study the Universe at high redshifts. Unfortunately, the 21 cm signal is affected by both cosmology and astrophysics processes in a non-trivial manner. We run a suite of 1,000 numerical simulations with different values of the main astrophysical parameters. From these simulations we produce tens of thousands of 21 cm maps at redshifts $10\le z\le 20$. We train a convolutional neural network to remove the effects of astrophysics from the 21 cm maps, and output maps of the underlying matter field. We show that our model is able to generate 2D matter fields that not only resemble the true ones visually, but whose statistical properties agree with the true ones within a few percent down to pretty small scales. We demonstrate that our neural network retains astrophysical information, that can be used to constrain the value of the astrophysical parameters. Finally, we use saliency maps to try to understand which features of the 21 cm maps is the network using in order to determine the value of the astrophysical parameters.

17 pages, 10 figures

References in corpus (9)

• Cosmology at Low Frequencies: The 21 cm Transition and the High-Redshift Universe
• How accurately can 21 cm tomography constrain cosmology?
• 21CMMC: an MCMC analysis tool enabling astrophysical parameter studies of the cosmic 21 cm signal
• Signature of Excess Radio Background in the 21-cm Global Signal and Power Spectrum
• Emulation of reionization simulations for Bayesian inference of astrophysics parameters using neural networks
• Nonlinear 3D Cosmic Web Simulation with Heavy-Tailed Generative Adversarial Networks
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs
• Deep-Learning Study of the 21cm Differential Brightness Temperature During the Epoch of Reionization
• 21cm Global Signal Extraction: Extracting the 21cm Global Signal using Artificial Neural Networks

Cited by in corpus (29)

• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• The CAMELS project: public data release
• deep21: a Deep Learning Method for 21cm Foreground Removal
• Learning cosmology and clustering with cosmic graphs
• Recovering the Wedge Modes Lost to 21-cm Foregrounds
• Deep learning approach for identification of HII regions during reionization in 21-cm observations
• Machine Learning for Observational Cosmology
• Eliminating Primary Beam Effect in Foreground Subtraction of Neutral Hydrogen Intensity Mapping Survey with Deep Learning
• Explaining deep learning of galaxy morphology with saliency mapping
• HIFlow: Generating Diverse HI Maps and Inferring Cosmology while Marginalizing over Astrophysics using Normalizing Flows
• HInet: Generating neutral hydrogen from dark matter with neural networks
• Inferring Astrophysics and Dark Matter Properties from 21cm Tomography using Deep Learning
• Interpreting automatic AGN classifiers with saliency maps
• Eliminating polarization leakage effect for neutral hydrogen intensity mapping with deep learning
• Cornering Extended Starobinsky Inflation with CMB and SKA
• Mitigating Foreground Bias to the CMB Lensing Power Spectrum for a CMB-HD Survey
• The Importance of Being Interpretable: Toward An Understandable Machine Learning Encoder for Galaxy Cluster Cosmology
• Understanding the Impact of Semi-numeric Reionization Models when using CNNs
• Probing light relics through cosmic dawn
• Deep learning approach for identification of HII regions during reionization in 21-cm observations -- III. image recovery
• Probing the Epoch of Reionization using synergies of line intensity mapping
• Predictive uncertainty on improved astrophysics recovery from multifield cosmology
• Sensitivity toward dark matter annihilation imprints on 21-cm signal with SKA-Low: A convolutional neural network approach
• Inferring halo masses with Graph Neural Networks
• Impact of astrophysical effects on the dark matter mass constraint with 21cm intensity mapping
• Interpretability of deep-learning methods applied to large-scale structure surveys
• Probing the Inflaton Potential with SKA