Understanding the Impact of Semi-numeric Reionization Models when using CNNs

Abstract

Interpreting 21cm measurements from current and upcoming experiments like HERA and the SKA will provide new scientific insights and exciting implications for astrophysics and cosmology regarding the Epoch of Reionization (EoR). Several recent works have proposed using machine learning methods, such as convolutions neural networks (CNNs), to analyze images of reionization generated by these experiments since they could take full advantage of information contained in the image. Generally, these studies have used only a single semi-numeric method to generate the input 21cm data. In this work, we investigate the extent to which training CNNs for reionization applications depends on the underlying semi-numeric models. Working in the context of predicting CMB optical depth from 21cm images, we compare networks trained on similar datasets from 21cmfast and zreion, two widely used semi-numeric reionization methods. We show that neural networks trained on input data from only one model produce poor predictions on data from the other model. Satisfactory results are only achieved when both models are included in the training data. This finding has important implications for future analyses on observation data, and encourages the use of multiple models to produce images that capture the full complexity of the EoR.

arXiv comment33 pages, 18 figures

References in corpus (49)

• Planck 2015 results. XIII. Cosmological parameters
• Large scale bias and the peak background split
• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• LOFAR: The LOw-Frequency ARray
• In the Beginning: The First Sources of Light and the Reionization of the Universe
• The Murchison Widefield Array: the Square Kilometre Array Precursor at low radio frequencies
• An absorption profile centred at 78 megahertz in the sky-averaged spectrum
• Observational constraints on Cosmic Reionization
• 21cmFAST: A Fast, Semi-Numerical Simulation of the High-Redshift 21-cm Signal
• Hydrogen Epoch of Reionization Array (HERA)
• Visualizing Deep Convolutional Neural Networks Using Natural Pre-Images
• Reionization and Cosmology with 21 cm Fluctuations
• Efficient Simulations of Early Structure Formation and Reionization
• 21 Centimeter Fluctuations from Cosmic Gas at High Redshifts
• The morphology of HII regions during reionization
• A Per-Baseline, Delay-Spectrum Technique for Accessing the 21cm Cosmic Reionization Signature
• Quantitative Constraints on the Reionization History from the IGM Damping Wing Signature in Two Quasars at z > 7
• Line Emitting Galaxies Beyond a Redshift of 7: An Improved Method for Estimating the Evolving Neutrality of the Intergalactic Medium
• Inferences on the Timeline of Reionization at z~8 From the KMOS Lens-Amplified Spectroscopic Survey
• First Results from HERA Phase I: Upper Limits on the Epoch of Reionization 21 cm Power Spectrum
• Opening the 21cm EoR Window: Measurements of Foreground Isolation with PAPER
• The Cosmic Dawn and Epoch of Reionization with the Square Kilometre Array
• 21cmFAST v3: A Python-integrated C code forgenerating 3D realizations of the cosmic 21cm signal
• Eliminating the optical depth nuisance from the CMB with 21 cm cosmology
• Astraeus I: The interplay between galaxy formation and reionization
• Quantifying the non-Gaussianity in the EoR 21-cm signal through bispectrum
• Deep learning from 21-cm tomography of the Cosmic Dawn and Reionization
• 21CMMC with a 3D light-cone: the impact of the co-evolution approximation on the astrophysics of reionisation and cosmic dawn
• Confirmation of Wide-Field Signatures in Redshifted 21 cm Power Spectra
• A tale of two sites -- I: Inferring the properties of minihalo-hosted galaxies from current observations
• Reionization on Large Scales III: Predictions for Low-ell Cosmic Microwave Background Polarization and High-ell Kinetic Sunyaev-Zel'dovich Observables
• Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images
• Constraining the redshifted 21-cm signal with the unresolved soft X-ray background
• 21cm-line bispectrum as method to probe Cosmic Dawn and Epoch of Reionization
• The 21cm bispectrum during reionization: a tracer of the ionization topology
• The Formation of Very Massive Stars in Early Galaxies and Implications for Intermediate Mass Black Holes
• Recovering the Wedge Modes Lost to 21-cm Foregrounds
• Reionization on Large Scales IV: Predictions for the 21 cm signal incorporating the light cone effect
• Signatures of metal-free star formation in Planck 2015 Polarization Data
• Removing Astrophysics in 21 cm maps with Neural Networks
• Machine Learning Applied to the Reionization History of the Universe in the 21 cm Signal
• HIGAN: Cosmic Neutral Hydrogen with Generative Adversarial Networks
• The 21 cm-kSZ-kSZ Bispectrum during the Epoch of Reionization
• Reionization on Large Scales II: Detecting Patchy Reionization through Cross Correlation of the Cosmic Microwave Background
• The Effect of Pair Cascades on the High-Energy Spectral Cutoff in Gamma-Ray Bursts
• Constraining the Reionization History using Bayesian Normalizing Flows
• Single Shot MC Dropout Approximation
• Deep-Learning Study of the 21cm Differential Brightness Temperature During the Epoch of Reionization
• Contribution of Radio Halos to the Foreground for SKA EoR Experiments

Cited by in corpus (7)

• Implicit Likelihood Inference of Reionization Parameters from the 21 cm Power Spectrum
• Probing the Epoch of Reionization using synergies of line intensity mapping
• Comparison of Bayesian inference methods using the Loreli II database of hydro-radiative simulations of the 21-cm signal
• An Alcock-Paczynski Test on Reionization Bubbles for Cosmology
• Machine Learning-Driven Analysis of kSZ Maps to Predict CMB Optical Depth $\tau$
• Exploring the Model Dependence of MCMC-Based 21 cm Power Spectrum Parameter Constraints
• Beyond the Power Spectrum: A New Framework for Non-Stationary Fields with Applications to Light-Cone Effects in Line Intensity Mapping