Cosmological super-resolution of the 21-cm signal

TL;DR

This paper introduces a score-based diffusion model that super-resolves large-scale cosmological 21-cm simulations, enabling detailed analysis of early Universe signals with high accuracy and potential for future astrophysical insights.

Contribution

The study demonstrates that a single simulation suffices for training a super-resolution model, achieving high accuracy across multiple resolutions and enabling utilization of SKA1-Low spatial scales.

Findings

Achieved pixelwise RMSE of ~0.57 mK in super-resolution.

Residuals in power spectrum range from 10^{-2} to 10^{-1} mK^2.

Model performs well across different simulation volumes at redshift 10.

Abstract

In this study, we train score-based diffusion models to super-resolve gigaparsec-scale cosmological simulations of the 21-cm signal. We examine the impact of network and training dataset size on model performance, demonstrating that a single simulation is sufficient for a model to learn the super-resolution task regardless of the initial conditions. Our best-performing model achieves pixelwise $\mathrm{RMSE}\sim0.57\ \mathrm{mK}$ and dimensionless power spectrum residuals ranging from $10^{-2}-10^{-1}\ \mathrm{mK^2}$ for $128^3$, $256^3$ and $512^3$ voxel simulation volumes at redshift $10$. The super-resolution network ultimately allows us to utilize all spatial scales covered by the SKA1-Low instrument, and could in future be employed to help constrain the astrophysics of the early Universe.