Using artificial neural networks to extract the 21-cm global signal from the EDGES data

Choudhury, Madhurima; Chatterjee, Atrideb; Datta, Abhirup; Choudhury, Tirthankar Roy

doi:10.1093/mnras/stab180

Cited by 15 publications

(8 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kern et al (2017) have used Gaussian Process (GP) regression based signal power spectrum emulators to constrain the astrophysics and cosmology of the EoR. In a different approach, Choudhury et al (2020Choudhury et al ( , 2021 have used ANNs to directly predict the EoR parameters from the simulated signal statistics.…”

Section: Introductionmentioning

confidence: 99%

Improving constraints on the reionization parameters using 21-cm bispectrum

Tiwari,

Shaw,

Majumdar

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Radio interferometric experiments aim to constrain the reionization model parameters by measuring the 21-cm signal statistics, primarily the power spectrum. However the Epoch of Reionization (EoR) 21-cm signal is highly non-Gaussian, and this non-Gaussianity encodes important information about this era. The bispectrum is the lowest order statistic able to capture this inherent non-Gaussianity. Here we are the first to demonstrate that bispectra for large and intermediate length scales and for all unique 𝑘-triangle shapes provide tighter constraints on the EoR parameters compared to the power spectrum or the bispectra for a limited number of shapes of 𝑘-triangles. We use the Bayesian inference technique to constrain EoR parameters. We have also developed an Artificial Neural Network (ANN) based emulator for the EoR 21-cm power spectrum and bispectrum which we use to remarkably speed up our parameter inference pipeline. Here we have considered the sample variance and the system noise uncertainties corresponding to 1000 hrs of SKA-Low observations for estimating errors in the signal statistics. We find that using all unique 𝑘-triangle bispectra improves the constraints on parameters by a factor of 2 − 4 (depending on the stage of reionization) over the constraints that are obtained using power spectrum alone.

show abstract

Section: Introductionmentioning

confidence: 99%

Improving constraints on the reionization parameters using 21-cm bispectrum

Tiwari,

Shaw,

Majumdar

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Because the spatial and frequency distributions of the 21-cm emission are highly non-linear and thus difficult to interpret, ML could be the most effective way to extract cosmological information. Various ML methods have been proposed to infer cosmological and astrophysical parameters from the intensity power spectrum [187][188][189], other summary statistics [190,191], and directly from tomographic maps [192][193][194] of the 21-cm line intensity. Ref.…”

Section: Line Intensity Mappingmentioning

confidence: 99%

Machine Learning for Observational Cosmology

Moriwaki¹,

Nishimichi²,

Yoshida³

2023

Preprint

View full text Add to dashboard Cite

An array of large observational programs using ground-based and spaceborne telescopes is planned in the next decade. The forthcoming wide-field sky surveys are expected to deliver a sheer volume of data exceeding an exabyte. Processing the large amount of multiplex astronomical data is technically challenging, and fully automated technologies based on machine learning and artificial intelligence are urgently needed. Maximizing scientific returns from the big data requires communitywide efforts. We summarize recent progress in machine learning applications in observational cosmology. We also address crucial issues in high-performance computing that are needed for the data processing and statistical analysis.

show abstract

“…Their parameter estimation method gave 92% accuracy, even for corrupted data sets. In Choudhury et al [2021a], they have developed ANN models to directly extract astrophysical characteristics from 21-cm Global signal observations, using physically justified 21-cm signal and foreground models. They have further developed their network in Choudhury et al [2021b] to extract the 21-cm PS and the corresponding EoR parameters from synthetic observations.…”

Section: Parameter Estimationmentioning

confidence: 99%

Synthetic Observations with the Square Kilometre Array (SKA) -- development towards an end-to-end pipeline

Mazumder¹,

Datta²,

Rao³

et al. 2022

Preprint

View full text Add to dashboard Cite

Detection of the redshifted 21-cm signal of neutral hydrogen from the Cosmic Dawn and the Epoch of Reionization is one of the final frontiers of modern observational cosmology. The inherently faint signal makes it susceptible to contamination by several sources like astrophysical foregrounds and instrumental systematics. Nevertheless, developments achieved in the recent times will combine to make signal detection possible with the upcoming Square Kilometer Array (SKA), both statistically and via tomography. This review describes an indigenously developed end-to-end pipeline that simulates sensitive interferometric observations. It mainly focuses on the requirements for Hi detection in interferometers. In its present form, it can mimic the effects of realistic point source foregrounds and systematics-calibration error and position error on 21-cm observations. The performance of the pipeline is demonstrated for test cases with 0.01% calibration error and position error. Its performance is consistent across telescope, foreground, and signal models. The focus of the simulation pipeline during the initial stages was for EoR science. But since this is a general interferometric simulation pipeline, it will be helpful to the entire SKA user community, irrespective of the science goals.

show abstract

Using artificial neural networks to extract the 21-cm global signal from the EDGES data

Cited by 15 publications

References 47 publications

Improving constraints on the reionization parameters using 21-cm bispectrum

Improving constraints on the reionization parameters using 21-cm bispectrum

Machine Learning for Observational Cosmology

Synthetic Observations with the Square Kilometre Array (SKA) -- development towards an end-to-end pipeline

Contact Info

Product

Resources

About