Hierarchical Bayesian CMB Component Separation with the No-U-Turn Sampler

Abstract

In this paper we present a novel implementation of Bayesian CMB component separation. We sample from the full posterior distribution using the No-U-Turn Sampler (NUTS), a gradient-based sampling algorithm. Alongside this, we introduce new foreground modelling approaches. We use the mean-shift algorithm to define regions on the sky, clustering according to naively estimated foreground spectral parameters. Over these regions we adopt a complete pooling model, where we assume constant spectral parameters, and a hierarchical model, where we model individual pixel spectral parameters as being drawn from underlying hyper-distributions. We validate the algorithm against simulations of the \textit{LiteBIRD} and C-BASS experiments, with an input tensor-to-scalar ratio of $r=5\times 10^{-3}$. Considering multipoles $30\le \ell <180$, we are able to recover estimates for $r$. With \textit{LiteBIRD}-only observations, and using the complete pooling model, we recover $r=(12.9\pm 1.4)\times 10^{-3}$. For C-BASS and \textit{LiteBIRD} observations we find $r=(9.0\pm 1.1)\times 10^{-3}$ using the complete pooling model, and $r=(5.2\pm 1.0)\times 10^{-3}$ using the hierarchical model. Unlike the complete pooling model, the hierarchical model captures pixel-scale spatial variations in the foreground spectral parameters, and therefore produces cosmological parameter estimates with reduced bias, without inflating the their uncertainties. Measured by the rate of effective sample generation, NUTS offers performance improvements of $\sim 10^3$ over using Metropolis-Hastings to fit the complete pooling model. The efficiency of NUTS allows us to fit the more sophisticated hierarchical foreground model, that would likely be intractable with non-gradient based sampling algorithms.

19 pages, 8 figures, matches version accepted in MNRAS

References in corpus (68)

• Scikit-learn: Machine Learning in Python
• Astropy: A Community Python Package for Astronomy
• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• Efficient Computation of CMB anisotropies in closed FRW models
• MCMC using Hamiltonian dynamics
• Nine-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Final Maps and Results
• A weakly informative default prior distribution for logistic and other regression models
• The Simons Observatory: Science goals and forecasts
• The Temperature of the Cosmic Microwave Background
• A Joint Analysis of BICEP2/Keck Array and Planck Data
• MASTER of the CMB Anisotropy Power Spectrum: A Fast Method for Statistical Analysis of Large and Complex CMB Data Sets
• The Primordial Inflation Explorer (PIXIE): A Nulling Polarimeter for Cosmic Microwave Background Observations
• A Conceptual Introduction to Hamiltonian Monte Carlo
• BICEP2 / Keck Array x: Constraints on Primordial Gravitational Waves using Planck, WMAP, and New BICEP2/Keck Observations through the 2015 Season
• The prior can generally only be understood in the context of the likelihood
• Planck intermediate results. XIX. An overview of the polarized thermal emission from Galactic dust
• A unified pseudo-$C_{\ell }$ framework
• The Quest for B Modes from Inflationary Gravitational Waves
• CMB-S4 Science Book, First Edition
• Joint Bayesian component separation and CMB power spectrum estimation
• Planck 2013 results. XII. Component separation
• Planck 2015 results. X. Diffuse component separation: Foreground maps
• Planck 2015 results. IX. Diffuse component separation: CMB maps
• Component separation methods for the Planck mission
• The Python Sky Model: software for simulating the Galactic microwave sky
• Global, Exact Cosmic Microwave Background Data Analysis Using Gibbs Sampling
• Separation of anomalous and synchrotron emissions using WMAP polarization data
• Power spectrum estimation from high-resolution maps by Gibbs sampling
• Multi-Detector Multi-Component spectral matching and applications for CMB data analysis
• A determination of the Spectra of Galactic components observed by WMAP
• Foreground component separation with generalised ILC
• Robust forecasts on fundamental physics from the foreground-obscured, gravitationally-lensed CMB polarization
• All-sky Galactic radiation at 45 MHz and spectral index between 45 and 408 MHz
• Rethinking CMB foregrounds: systematic extension of foreground parameterizations
• Maximum Likelihood algorithm for parametric component separation in CMB experiments
• The State-of-Play of Anomalous Microwave Emission (AME) Research
• Planck 2015 results. XXV. Diffuse low-frequency Galactic foregrounds
• The S-PASS view of polarized Galactic Synchrotron at 2.3 GHz as a contaminant to CMB observations
• Polarised radio filaments outside the Galactic plane
• Large Angular Scale Polarization of the Cosmic Microwave Background and the Feasibility of its Detection
• Sensitivity and foreground modelling for large-scale CMB B-mode polarization satellite missions
• CMBPol Mission Concept Study: Prospects for polarized foreground removal
• Full Sky Study of Diffuse Galactic Emission at Decimeter Wavelengths
• Exploring Cosmic Origins with CORE: B-mode Component Separation
• Fast optimal CMB power spectrum estimation with Hamiltonian sampling
• Exploring cosmic origins with CORE: gravitational lensing of the CMB
• The C-Band All-Sky Survey (C-BASS): Design and capabilities
• FR II radio galaxies at low frequencies I: morphology, magnetic field strength and energetics
• Measuring the tensor to scalar ratio from CMB B-modes in presence of foregrounds
• Forecasting performance of CMB experiments in the presence of complex foreground contaminations
• Diagnosing Suboptimal Cotangent Disintegrations in Hamiltonian Monte Carlo
• Galactic foreground contributions to the WMAP5 maps
• CMB power spectrum estimation and map reconstruction with the Expectation - Maximization algorithm
• Weak Informativity and the Information in One Prior Relative to Another
• Mitigating Complex Dust Foregrounds in Future CMB Polarization Experiments
• Unbiased pseudo-Cl power spectrum estimation with mode projection
• Quantum Suppression of Alignment in Ultrasmall Grains: Microwave Emission from Spinning Dust will be Negligibly Polarized
• Maximum likelihood, parametric component separation and CMB B-mode detection in suborbital experiments
• Simulated forecasts for primordial B-mode searches in ground-based experiments
• Framework for performance forecasting and optimization of CMB B-mode observations in presence of astrophysical foregrounds
• Multi-resolution Bayesian CMB component separation through Wiener-filtering with a pseudo-inverse preconditioner
• Determining thermal dust emission from Planck HFI data using a sparse, parametric technique
• The spectral index of polarized diffuse Galactic emission between 30 and 44 GHz
• A Novel CMB Component Separation Method: Hierarchical Generalized Morphological Component Analysis
• Probing the origin of our Universe through cosmic microwave background constraints on gravitational waves
• Data driven foreground clustering approach to component separation in multifrequency CMB experiments: A new Planck CMB map
• Removal of Galactic foregrounds for the Simons Observatory primordial gravitational wave search
• The C-Band All-Sky Survey (C-BASS): Simulated parametric fitting in single pixels in total intensity and polarization

Cited by in corpus (8)

• A minimal power-spectrum-based moment expansion for CMB B-mode searches
• The SAMI galaxy survey: galaxy size can explain the offset between star-forming and passive galaxies in the mass-metallicity relationship
• QUIJOTE scientific results -- VIII. Diffuse polarized foregrounds from component separation with QUIJOTE-MFI
• Improved Galactic Foreground Removal for B-Modes Detection with Clustering Methods
• K-CLASH: Strangulation and Ram Pressure Stripping in Galaxy Cluster Members at 0.3 < z < 0.6
• A hybrid map-$C_{\ell }$ component separation method for primordial CMB $B$-mode searches
• Self-supervised component separation for the extragalactic submillimeter sky
• Bayesian inference methodology to characterize the dust emissivity at far-infrared and submillimeter frequencies