COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys

TL;DR

CosmoPower introduces neural emulators for cosmological power spectra, significantly accelerating parameter estimation in large-scale structure and CMB analyses without sacrificing accuracy, enabling rapid inference for upcoming surveys.

Contribution

It provides a suite of neural emulators that replace Boltzmann code computations, offering high accuracy and speed for cosmological parameter inference across various datasets.

Findings

Emulation error less than 0.4% for matter power spectra.

Achieves up to 10,000x speed-up in inference pipelines.

Successfully recovers fiducial cosmological constraints in tests.

Abstract

We present $\it{CosmoPower}$, a suite of neural cosmological power spectrum emulators providing orders-of-magnitude acceleration for parameter estimation from two-point statistics analyses of Large-Scale Structure (LSS) and Cosmic Microwave Background (CMB) surveys. The emulators replace the computation of matter and CMB power spectra from Boltzmann codes; thus, they do not need to be re-trained for different choices of astrophysical nuisance parameters or redshift distributions. The matter power spectrum emulation error is less than $0.4\%$ in the wavenumber range $k \in [10^{-5}, 10] \, \mathrm{Mpc}^{-1}$, for redshift $z \in [0, 5]$. $\it{CosmoPower}$ emulates CMB temperature, polarisation and lensing potential power spectra in the $5\sigma$ region of parameter space around the $\it{Planck}$ best fit values with an error $\lesssim 10\%$ of the expected shot noise for the forthcoming Simons Observatory. $\it{CosmoPower}$ is showcased on a joint cosmic shear and galaxy clustering analysis from the Kilo-Degree Survey, as well as on a Stage IV $\it{Euclid}$-like simulated cosmic shear analysis. For the CMB case, $\it{CosmoPower}$ is tested on a $\it{Planck}$ 2018 CMB temperature and polarisation analysis. The emulators always recover the fiducial cosmological constraints with differences in the posteriors smaller than sampling noise, while providing a speed-up factor up to $O(10^4)$ to the complete inference pipeline. This acceleration allows posterior distributions to be recovered in just a few seconds, as we demonstrate in the $\it{Planck}$ likelihood case. $\it{CosmoPower}$ is written entirely in Python, can be interfaced with all commonly used cosmological samplers and is publicly available at https://github.com/alessiospuriomancini/cosmopower .