AI-assisted super-resolution cosmological simulations V: Cosmology-aware super-resolution
Abstract
We present a cosmology-aware generative model for super-resolution (SR) of cosmological N-body simulations in particle phase space. The model jointly generates displacements and velocities, conditioning a generative adversarial network (GAN) based generator and discriminator on evolved low-resolution (LR) fields and five ΛCDM parameters (Ωm,Ωb,h,ns,σ8). We train a single model on 100 matched low-high resolution pairs from the Quijote Latin-hypercube suite at z = 0, mapping 5123 to 10243 particles in 1 h−1 Gpc boxes. On ten cosmologies excluded from SR training, the model reproduces the broad matter distribution and the cosmological variation of clustering and halo abundances. Over 0.1<k<3 h Mpc−1, the mean of the per-cosmology maximum absolute power-spectrum errors is 3.6%, while FoF halo abundances agree with HR to within 15% in the analyzed mass bins below 1014.8 h−1M⊙. Satellite abundances and occupations retain larger, mass-dependent discrepancies. Corrections measured at one near-fiducial cosmology reduce a shared component of these systematic errors when applied to the other nine test cosmologies. A power spectrum based inference test also shows that calibration reduces the mean shifts in all parameters. These results extend our SR model across cosmologies and provide 6D phase space realizations for subsequent studies of non-linear structure and field-level cosmological inference.
Related
- Cosmology with a Non-minimally Coupled Dark Matter Fluid II. Cosmological Perturbations
- Rates of tidal disruption events from constrained cosmological simulations of the local Universe: population properties and implications for transient surveys
- Coupling between stellar and HI lopsidedness in Milky Way--type galaxies from the Auriga Superstars cosmological simulations
- Four-Sector Cosmology (FSC): A Complex Dark Energy Field, Spontaneous Symmetry Breaking, and the Origin of the Cosmological Constant from Observed Matter Asymmetry δ = 0.2%
- Bridging perturbation theory and simulations: initial conditions and fast integrators for cosmological simulations
BibTeX
@misc{zhang2026aiassisted,
title = {{AI-assisted super-resolution cosmological simulations V: Cosmology-aware super-resolution}},
author = {Xiaowen Zhang and Patrick Lachance and Rupert A. C. Croft and Tiziana Di Matteo and Francisco Villaescusa-Navarro},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.06710},
eprint = {2610.06710},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.06710},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - AI-assisted super-resolution cosmological simulations V: Cosmology-aware super-resolution
AU - Xiaowen Zhang
AU - Patrick Lachance
AU - Rupert A. C. Croft
AU - Tiziana Di Matteo
AU - Francisco Villaescusa-Navarro
PY - 2026
DA - 2026/10/05
PB - arXiv
DO - 10.48550/arxiv.2610.06710
UR - https://arxiv.org/abs/2610.06710
AB - We present a cosmology-aware generative model for super-resolution (SR) of cosmological N-body simulations in particle phase space. The model jointly generates displacements and velocities, conditioning a generative adversarial network (GAN) based generator and discriminator on evolved low-resolution (LR) fields and five ΛCDM parameters (Ω_m,Ω_b,h,n_s,σ_8). We train a single model on 100 matched low-high resolution pairs from the Quijote Latin-hypercube suite at z = 0, mapping 512^3 to 1024^3 particles in 1 h^−1 Gpc boxes. On ten cosmologies excluded from SR training, the model reproduces the broad matter distribution and the cosmological variation of clustering and halo abundances. Over 0.1<k<3 h Mpc^−1, the mean of the per-cosmology maximum absolute power-spectrum errors is 3.6%, while FoF halo abundances agree with HR to within 15% in the analyzed mass bins below 10^14.8 h^−1M_⊙. Satellite abundances and occupations retain larger, mass-dependent discrepancies. Corrections measured at one near-fiducial cosmology reduce a shared component of these systematic errors when applied to the other nine test cosmologies. A power spectrum based inference test also shows that calibration reduces the mean shifts in all parameters. These results extend our SR model across cosmologies and provide 6D phase space realizations for subsequent studies of non-linear structure and field-level cosmological inference.
ER -