Capse.jl: efficient and auto-differentiable CMB power spectra emulation

Abstract

We present Capse.jl, a novel neural network-based emulator designed for rapid and accurate prediction of Cosmic Microwave Background (CMB) temperature, polarization, and lensing angular power spectra. The emulator computes predictions in just a few microseconds with emulation errors below $0.1\sigma$ for all the scales relevant for the upcoming CMB-S4 survey. Capse.jl can also be trained in an hour's time on a 8-cores CPU. We test Capse.jl on Planck 2018, ACT DR4, and 2018 SPT-3G data and demonstrate its capability to derive cosmological constraints comparable to those obtained by traditional methods, but with a computational efficiency that is three to six orders of magnitude higher. We take advantage of the differentiability of our emulators to use gradient-based methods, such as Pathfinder and Hamiltonian Monte Carlo (HMC), which speed up the convergence and increase sampling efficiency. Together, these features make Capse.jl a powerful tool for studying the CMB and its implications for cosmology. When using the fastest combination of our likelihoods, emulators, and analysis algorithm, we are able to perform a Plancky TT + TE + EE analysis in less than a second. To ensure full reproducibility, we provide open access to the codes and data required to reproduce all the results of this work.

16 pages, 6 figures. Accepted for publication in The Open Journal of Astrophysics. Capse.jl is available $\text{href}$

References in corpus (26)

• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Anisotropic galaxy clustering in Fourier-space
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• Fast cosmological parameter estimation using neural networks
• Massive data compression for parameter-dependent covariance matrices
• SPHEREx: NASA's Near-Infrared Spectrophotmetric All-Sky Survey
• JAX-COSMO: An End-to-End Differentiable and GPU Accelerated Cosmology Library
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Accelerating Large-Scale-Structure data analyses by emulating Boltzmann solvers and Lagrangian Perturbation Theory
• CosmicNet II: Emulating extended cosmologies with efficient and accurate neural networks
• Kernel-Based Emulator for the 3D Matter Power Spectrum from CLASS
• Bispectrum-window convolution via Hankel transform
• $\text{matryoshka}$: Halo Model Emulator for the Galaxy Power Spectrum
• Parameter Inference for Weak Lensing using Gaussian Processes and MOPED
• COMET: Clustering Observables Modelled by Emulated perturbation Theory
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• LINNA: Likelihood Inference Neural Network Accelerator
• $\text{matryoshka}$ II: Accelerating Effective Field Theory Analyses of the Galaxy Power Spectrum
• Pathfinder: Parallel quasi-Newton variational inference
• Microcanonical Hamiltonian Monte Carlo
• Uncertainty-aware and Data-efficient Cosmological Emulation using Gaussian Processes and PCA
• Window function convolution with deep neural network models
• Field-Level Inference with Microcanonical Langevin Monte Carlo
• Fast emulation of two-point angular statistics for photometric galaxy surveys
• Fluctuation without dissipation: Microcanonical Langevin Monte Carlo
• High-accuracy emulators for observables in $\Lambda$CDM, $N_{\mathrm{eff}}$, $\Sigma m_{\nu }$, and $w$ cosmologies

Cited by in corpus (15)

• The CosmoVerse White Paper: Addressing observational tensions in cosmology with systematics and fundamental physics
• candl: Cosmic Microwave Background Analysis with a Differentiable Likelihood
• DISCO-DJ I: a differentiable Einstein-Boltzmann solver for cosmology
• Fast and effortless computation of profile likelihoods using CONNECT
• LINX: A Fast, Differentiable, and Extensible Big Bang Nucleosynthesis Package
• Selecting samples of galaxies with fewer Fingers-of-God
• Reducing nuisance prior sensitivity via non-linear reparameterization, with application to EFT analyses of large-scale structure
• Calculating Bayesian evidence for inflationary models using CONNECT
• DeepSSM: an emulator of gravitational wave spectra from sound waves during cosmological first-order phase transitions
• Constraining the primordial power spectrum using a differentiable likelihood
• Cosmological parameters from the joint analysis of Density Split and Second Order Statistics: an Emulator based on the Halo Occupation Distribution
• Design and optimization of neural networks for multifidelity cosmological emulation
• Constraining cosmological parameters using density split lensing and the conditional stellar mass function
• On the accuracy of posterior recovery with neural network emulators
• ABCMB: A Python+JAX Package for the Cosmic Microwave Background Power Spectrum