Reconstructing Probability Distributions with Gaussian Processes

Abstract

Modern cosmological analyses constrain physical parameters using Markov Chain Monte Carlo (MCMC) or similar sampling techniques. Oftentimes, these techniques are computationally expensive to run and require up to thousands of CPU hours to complete. Here we present a method for reconstructing the log-probability distributions of completed experiments from an existing MCMC chain (or any set of posterior samples). The reconstruction is performed using Gaussian process regression for interpolating the log-probability. This allows for easy resampling, importance sampling, marginalization, testing different samplers, investigating chain convergence, and other operations. As an example use-case, we reconstruct the posterior distribution of the most recent Planck 2018 analysis. We then resample the posterior, and generate a new MCMC chain with forty times as many points in only thirty minutes. Our likelihood reconstruction tool can be found online at https://github.com/tmcclintock/AReconstructionTool.

arXiv comment7 pages, 4 figures, repository at https://github.com/tmcclintock/AReconstructionTool

References in corpus (21)

• Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant
• Measurements of Omega and Lambda from 42 High-Redshift Supernovae
• emcee: The MCMC Hammer
• The NumPy array: a structure for efficient numerical computation
• Cosmological parameters from CMB and other data: a Monte-Carlo approach
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample
• The 6dF Galaxy Survey: Baryon Acoustic Oscillations and the Local Hubble Constant
• Dark Energy Survey Year 1 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• Conservative Constraints on Early Cosmology: an illustration of the Monte Python cosmological parameter inference code
• Cosmology from cosmic shear power spectra with Subaru Hyper Suprime-Cam first-year data
• CosmoSIS: modular cosmological parameter estimation
• Dark Energy Survey Year 1 Results: A Precise H0 Measurement from DES Y1, BAO, and D/H Data
• The Coyote Universe II: Cosmological Models and Precision Emulation of the Nonlinear Matter Power Spectrum
• Fast likelihood-free cosmology with neural density estimators and active learning
• Dark Quest. I. Fast and Accurate Emulation of Halo Clustering Statistics and Its Application to Galaxy Clustering
• The Mira-Titan Universe II: Matter Power Spectrum Emulation
• The Aemulus Project II: Emulating the Halo Mass Function
• Nuisance hardened data compression for fast likelihood-free inference
• Dark Energy Survey Year 1 Results: Methodology and Projections for Joint Analysis of Galaxy Clustering, Galaxy Lensing, and CMB Lensing Two-point Functions
• Hunting high and low: Disentangling primordial and late-time non-Gaussianity with cosmic densities in spheres
• Posterior inference unchained with EL_2O

Cited by in corpus (11)

• Simultaneous modelling of matter power spectrum and bispectrum in the presence of baryons
• Accelerating Large-Scale-Structure data analyses by emulating Boltzmann solvers and Lagrangian Perturbation Theory
• On The XUV Luminosity Evolution of TRAPPIST-1
• Cosmological parameter estimation via iterative emulation of likelihoods
• The bias of dark matter tracers: assessing the accuracy of mapping techniques
• Density estimation with Gaussian processes for gravitational-wave posteriors
• CombineHarvesterFlow: Joint Probe Analysis Made Easy with Normalizing Flows
• The BACCO Simulation Project: A baryonification emulator with Neural Networks
• PoPE: A population-based approach to model spatial structure of astronomical systems
• Boosting MonteCarlo sampling with a non-Gaussian fit
• How to embed any likelihood into SBI: Application to Planck + Stage IV galaxy surveys and Dynamical Dark Energy