Design and optimization of neural networks for multifidelity cosmological emulation

Abstract

Accurate and efficient simulation-based emulators are essential for interpreting cosmological survey data down to nonlinear scales. Multifidelity emulation techniques reduce simulation costs by combining high- and low-fidelity data, but traditional regression methods such as Gaussian processes struggle with scalability in sample size and dimensionality. In this work, we present T2N-MusE, a neural network framework characterized by (i) a novel 2-step multifidelity architecture, (ii) a 2-stage Bayesian hyperparameter optimization, (iii) a 2-phase $k$-fold training strategy, and (iv) a per-$z$ principal component analysis strategy. We apply T2N-MusE to selected data from the Goku simulation suite, covering a 10-dimensional cosmological parameter space, and build emulators for the matter power spectrum over a range of redshifts with different configurations. We find the emulators outperform our earlier Gaussian process models significantly and demonstrate that each of these techniques is efficient in training neural networks or/and effective in improving generalization accuracy. We observe a reduction in the mean error by more than a factor of five and in the worst-case error by approximately a factor of eight in leave-one-out cross-validation, relative to previous work. This framework has been used to build the most powerful emulator for the matter power spectrum, GokuNEmu, and will also be used to construct emulators for other statistics in future.

16 pages, 13 figures, 5 tables. Updated to match the version published in Phys. Rev. D

References in corpus (31)

• Extragalactic Science, Cosmology and Galactic Archaeology with the Subaru Prime Focus Spectrograph (PFS)
• The Quijote simulations
• The Coyote Universe I: Precision Determination of the Nonlinear Matter Power Spectrum
• Cosmology from the Chinese Space Station Optical Survey (CSS-OS)
• The Coyote Universe II: Cosmological Models and Precision Emulation of the Nonlinear Matter Power Spectrum
• The Coyote Universe Extended: Precision Emulation of the Matter Power Spectrum
• Dark Quest. I. Fast and Accurate Emulation of Halo Clustering Statistics and Its Application to Galaxy Clustering
• Euclid preparation: II. The EuclidEmulator -- A tool to compute the cosmology dependence of the nonlinear matter power spectrum
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• Euclid preparation: IX. EuclidEmulator2 -- Power spectrum emulation with massive neutrinos and self-consistent dark energy perturbations
• The Aemulus Project I: Numerical Simulations for Precision Cosmology
• The Mira-Titan Universe II: Matter Power Spectrum Emulation
• The Mira-Titan Universe: Precision Predictions for Dark Energy Surveys
• The Aemulus Project III: Emulation of the Galaxy Correlation Function
• Multi-fidelity regression using artificial neural networks: efficient approximation of parameter-dependent output quantities
• The Aemulus Project II: Emulating the Halo Mass Function
• Fast cosmological parameter estimation using neural networks
• {\sc CosmoNet}: fast cosmological parameter estimation in non-flat models using neural networks
• The Mira-Titan Universe IV. High Precision Power Spectrum Emulation
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Precision modelling of the matter power spectrum in a Planck-like Universe
• CosmicNet II: Emulating extended cosmologies with efficient and accurate neural networks
• Capse.jl: efficient and auto-differentiable CMB power spectra emulation
• Multi-Fidelity Emulation for the Matter Power Spectrum using Gaussian Processes
• PRIYA: A New Suite of Lyman-alpha Forest Simulations for Cosmology
• CSST Cosmological Emulator I: Matter Power Spectrum Emulation with one percent accuracy to k = 10 h/Mpc
• A neural network emulator for the Lyman-$\alpha$ 1D flux power spectrum
• BE-HaPPY: Bias Emulator for Halo Power Spectrum including massive neutrinos
• Galaxy Clustering in the Mira-Titan Universe I: Emulators for the redshift space galaxy correlation function and galaxy-galaxy lensing
• MF-Box: Multi-fidelity and multi-scale emulation for the matter power spectrum
• SageNet: Fast Neural Network Emulation of the Stiff-amplified Gravitational Waves from Inflation

Cited by in corpus (1)

• Ten-dimensional neural network emulator for the nonlinear matter power spectrum