Neural Network Acceleration of Large-scale Structure Theory Calculations

Abstract

We make use of neural networks to accelerate the calculation of power spectra required for the analysis of galaxy clustering and weak gravitational lensing data. For modern perturbation theory codes, evaluation time for a single cosmology and redshift can take on the order of two seconds. In combination with the comparable time required to compute linear predictions using a Boltzmann solver, these calculations are the bottleneck for many contemporary large-scale structure analyses. In this work, we construct neural network-based surrogate models for Lagrangian perturbation theory (LPT) predictions of matter power spectra, real and redshift space galaxy power spectra, and galaxy--matter cross power spectra that attain $\sim 0.1\%$ (at one sigma) accuracy over a broad range of scales in a $w$CDM parameter space. The neural network surrogates can be evaluated in approximately one millisecond, a factor of 1000 times faster than the full Boltzmann code and LPT computations. In a simulated full-shape redshift space galaxy power spectrum analysis, we demonstrate that the posteriors obtained using our surrogates are accurate compared to those obtained using the full LPT model. We make our surrogate models public at https://github.com/sfschen/EmulateLSS, so that others may take advantage of the speed gains they provide to enable rapid iteration on analysis settings, something that is essential in complex contemporary large-scale structure analyses.

arXiv comment12 pages, 4 figures, models available at https://github.com/sfschen/EmulateLSS

References in corpus (34)

• Efficient Computation of CMB anisotropies in closed FRW models
• The Cosmic Linear Anisotropy Solving System (CLASS) II: Approximation schemes
• Stable clustering, the halo model and nonlinear cosmological power spectra
• The Baryon Oscillation Spectroscopic Survey of SDSS-III
• Revising the Halofit Model for the Nonlinear Matter Power Spectrum
• The SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Overview and Early Data
• Planck 2018 results. VIII. Gravitational lensing
• Cobaya: Code for Bayesian Analysis of hierarchical physical models
• Cosmological Non-Linearities as an Effective Fluid
• Clustering of dark matter tracers: generalizing bias for the coming era of precision LSS
• Non-linear perturbation theory extension of the Boltzmann code CLASS
• The BACCO Simulation Project: Exploiting the full power of large-scale structure for cosmology
• DEMNUni: The clustering of large-scale structures in the presence of massive neutrinos
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• A new analysis of galaxy 2-point functions in the BOSS survey, including full-shape information and post-reconstruction BAO
• Blinded challenge for precision cosmology with large-scale structure: results from effective field theory for the redshift-space galaxy power spectrum
• A Lagrangian effective field theory
• Efficient Cosmological Analysis of the SDSS/BOSS data from the Effective Field Theory of Large-Scale Structure
• Redshift-Space Distortions in Lagrangian Perturbation Theory
• The Aemulus Project I: Numerical Simulations for Precision Cosmology
• Consistent Modeling of Velocity Statistics and Redshift-Space Distortions in One-Loop Perturbation Theory
• Cosmological constraints from the tomographic cross-correlation of DESI Luminous Red Galaxies and Planck CMB lensing
• The Gaussian streaming model and Lagrangian effective field theory
• On the EFT of Large Scale Structures in Redshift Space
• Cosmic Calibration
• The cosmology dependence of galaxy clustering and lensing from a hybrid $N$-body-perturbation theory model
• Simulations and symmetries
• SPECULATOR: Emulating stellar population synthesis for fast and accurate galaxy spectra and photometry
• {\sc CosmoNet}: fast cosmological parameter estimation in non-flat models using neural networks
• PkANN - II. A non-linear matter power spectrum interpolator developed using artificial neural networks
• Modeling CMB Lensing Cross Correlations with {\sc CLEFT}
• Efficient exploration of cosmology dependence in the EFT of LSS
• Euclid-era cosmology for everyone: Neural net assisted MCMC sampling for the joint 3x2 likelihood
• CosmicNet I: Physics-driven implementation of neural networks within Boltzmann-Einstein solvers

Cited by in corpus (6)

• Cosmological Analysis of Three-Dimensional BOSS Galaxy Clustering and Planck CMB Lensing Cross Correlations via Lagrangian Perturbation Theory
• The Bacco Simulation Project: Bacco Hybrid Lagrangian Bias Expansion Model in Redshift Space
• COMET: Clustering Observables Modelled by Emulated perturbation Theory
• The Aemulus Project VI: Emulation of beyond-standard galaxy clustering statistics to improve cosmological constraints
• LINNA: Likelihood Inference Neural Network Accelerator
• Window function convolution with deep neural network models