Saved in:
Bibliographic Details
Main Authors: Kang, Yuqi, Bin, Hu, Li, Dongxing, Hamann, Jan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.18165
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917352413593600
author Kang, Yuqi
Bin, Hu
Li, Dongxing
Hamann, Jan
author_facet Kang, Yuqi
Bin, Hu
Li, Dongxing
Hamann, Jan
contents In this work, we introduce TUNeS (Temporal UNet emulator for Structure formation), a neural network framework for accelerating N-body simulations by predicting the nonlinear evolution of the matter density field from an initial particle distribution. TUNeS employs a two-stage modeling strategy, combining particle-based inference with a density-field refinement on a regular grid, enabling accurate reconstruction of both large- and small-scale structures. The model is designed to operate across redshift, taking particle snapshots at arbitrary input redshifts and predicting density fields at arbitrary target redshifts. In this work, we evaluate its performance using simulations initialized at $z=100$, with predictions generated at multiple lower redshifts. Trained on only eight N-body simulations, TUNeS reproduces reference results with good agreement in both Gaussian and non-Gaussian statistics, including two-point correlations, one-point distributions, peak counts, and three-dimensional Minkowski functionals. In particular, at $k \simeq 1\,h\,\mathrm{Mpc}^{-1}$, the power spectrum error remains at the few-percent level. End-to-end inference from $256^3$ particles to a $256^3$ density grid can be completed in $\sim25\,\mathrm{second}$ on a single GPU. Thanks to its architectural design, the model naturally scales to larger particle numbers and larger volumes through particle batching and window-based refinement.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18165
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TUNeS: Neural Emulation of Large-Scale Structure Across Redshifts
Kang, Yuqi
Bin, Hu
Li, Dongxing
Hamann, Jan
Cosmology and Nongalactic Astrophysics
Astrophysics of Galaxies
In this work, we introduce TUNeS (Temporal UNet emulator for Structure formation), a neural network framework for accelerating N-body simulations by predicting the nonlinear evolution of the matter density field from an initial particle distribution. TUNeS employs a two-stage modeling strategy, combining particle-based inference with a density-field refinement on a regular grid, enabling accurate reconstruction of both large- and small-scale structures. The model is designed to operate across redshift, taking particle snapshots at arbitrary input redshifts and predicting density fields at arbitrary target redshifts. In this work, we evaluate its performance using simulations initialized at $z=100$, with predictions generated at multiple lower redshifts. Trained on only eight N-body simulations, TUNeS reproduces reference results with good agreement in both Gaussian and non-Gaussian statistics, including two-point correlations, one-point distributions, peak counts, and three-dimensional Minkowski functionals. In particular, at $k \simeq 1\,h\,\mathrm{Mpc}^{-1}$, the power spectrum error remains at the few-percent level. End-to-end inference from $256^3$ particles to a $256^3$ density grid can be completed in $\sim25\,\mathrm{second}$ on a single GPU. Thanks to its architectural design, the model naturally scales to larger particle numbers and larger volumes through particle batching and window-based refinement.
title TUNeS: Neural Emulation of Large-Scale Structure Across Redshifts
topic Cosmology and Nongalactic Astrophysics
Astrophysics of Galaxies
url https://arxiv.org/abs/2603.18165