Reconstruction of full sky CMB and modes spectra removing -to- leakage from partial sky using deep learning
arXiv:2211.09112 · doi:10.1007/s12036-023-09974-4
Abstract
Incomplete sky analysis of cosmic microwave background (CMB) polarization spectra poses a major problem of leakage between - and -modes. We present a machine learning approach to remove this -to- leakage using a convolutional neural network (CNN) in presence of detector noise. The CNN predicts the full sky - and -modes spectra for multipoles from the partial sky spectra for . We use tensor-to-scalar ratio to simulate the CMB polarization maps. We train our CNN using full sky target spectra and an equal number of noise contaminated partial sky spectra obtained from the simulated maps. The CNN works well for two masks covering the sky area of and respectively after training separately for each mask. For the assumed theoretical - and -modes spectra, predicted full sky - and -modes spectra agree well with the corresponding target spectra and their means agree with theoretical spectra. The CNN preserves the cosmic variances at each multipole, effectively removes correlations of the partial sky - and -modes spectra, and retains the entire statistical properties of the targets avoiding the problem of so-called -to- leakage for the chosen theoretical model.
arXiv comment24 pages, 13 figures
References in corpus (36)
- HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
- Nine-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Parameter Results
- The Cosmic Microwave Background Spectrum from the Full COBE/FIRAS Data Set
- 4-Year COBE DMR Cosmic Microwave Background Observations: Maps and Basic Results
- An All-Sky Analysis of Polarization in the Microwave Background
- Statistics of Cosmic Microwave Background Polarization
- Detection of Polarization in the Cosmic Microwave Background using DASI
- Calibrator Design for the COBE Far Infrared Absolute Spectrophotometer (FIRAS)
- CMB Anisotropies: Total Angular Momentum Method
- A unified pseudo- framework
- COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
- Analysis of CMB polarization on an incomplete sky
- E/B decomposition of finite pixelized CMB maps
- Planck constraints on the tensor-to-scalar ratio
- Measuring Polarization with DASI
- Neural networks and separation of Cosmic Microwave Background and astrophysical signals in sky maps
- A deep learning approach to cosmological dark energy models
- Reconstructing Functions and Estimating Parameters with Artificial Neural Networks: A Test with the Hubble Parameter and SNe Ia
- A general solution to the E-B mixing problem
- Neural Network Reconstruction of Late-Time Cosmology and Null Tests
- Harmonic E/B decomposition for CMB polarization maps
- Full-sky Cosmic Microwave Background Foreground Cleaning Using Machine Learning
- Separating E and B types of polarization on an incomplete sky
- Accelerated Bayesian inference using deep learning
- Neural network reconstructions for the Hubble parameter, growth rate and distance modulus
- Reconstructing Cosmological Initial Conditions from Late-Time Structure with Convolutional Neural Networks
- Methods for pixel domain correction of EB leakage
- Observational cosmology with Artificial Neural Networks
- Efficiency of pseudo-spectrum methods for estimation of Cosmic Microwave Background B-mode power spectrum
- Wiener filtering and pure E/B decomposition of CMB maps with anisotropic correlated noise
- Pure E and B polarization maps via Wiener filtering
- Blind correction of the EB-leakage in the pixel domain
- Detection of Dipole Modulation in CMB Temperature Anisotropy Maps from WMAP and Planck using Artificial Intelligence
- Optimal and fast E/B separation with a dual messenger field
- Estimation of Full Sky Power Spectrum between Intermediate to Large Angular Scales from Partial Sky CMB Anisotropies using Artificial Neural Network
- An Unbiased Estimator of the Full-sky CMB Angular Power Spectrum at Large Scales using Neural Networks