Benchmarking AI-evolved cosmological structure formation

Abstract The potential of deep learning-based image-to-image translations has attracted significant attention as a fast, approximate alternative to cosmological simulations, which would be particularly useful in various contexts, including covariance studies, investigations of systematics, and cosmological parameter inference. We model the evolution of structure formation using U-Net, a widely employed convolutional image translation framework with previous applications in cosmology. To study aspects of learning-based cosmological mappings in a controlled, illustrative setting, we use two approaches for generating suitable cosmological matter fields as datasets: a simple analytical prescription provided by the Zel’dovich approximation, and a standard N-body method using Particle-Mesh simulations. We apply established metrics such as matter power spectra and bispectra, topological indicators, and statistical independence of density fields and study their reproducibility and convergence properties. The U-Net approach performs well only for some of these physical metrics, and accuracy is worse at more nonlinear scales, where the dynamic range in density is larger. Motivated by field-level emulator studies utilizing Eulerian-aware loss functions and beyond pixel-wise mean squared errors, we examine a density-weighted objective. In the PM tests ending at z = 4, where the evolution is only mildly nonlinear, using the Eulerian matter density evaluated at each particle position as a direct weight on that particle’s Lagrangian displacement error improves the U-Net results at smaller scales; this result does not establish comparable accuracy in the strongly nonlinear low-redshift regime.

Authors

Institutions

Publication Details

Journal
Monthly Notices of the Royal Astronomical Society
Published
2026-11-14
DOI
https://doi.org/10.1093/mnras/stag2032
Primary Topic
Cosmology and Gravitation Theories
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Benchmarking AI-evolved cosmological structure formation

Katrin Heitmann, Nesar Ramachandra, Salman Habib, Xiaofeng Dong
Monthly Notices of the Royal Astronomical Society
Cosmology and Gravitation Theories
article

Benchmarking AI-evolved cosmological structure formation

Katrin Heitmann, Nesar Ramachandra, Salman Habib, Xiaofeng Dong
article en

Abstract

Abstract The potential of deep learning-based image-to-image translations has attracted significant attention as a fast, approximate alternative to cosmological simulations, which would be particularly useful in various contexts, including covariance studies, investigations of systematics, and cosmological parameter inference. We model the evolution of structure formation using U-Net, a widely employed convolutional image translation framework with previous applications in cosmology. To study aspects of learning-based cosmological mappings in a controlled, illustrative setting, we use two approaches for generating suitable cosmological matter fields as datasets: a simple analytical prescription provided by the Zel’dovich approximation, and a standard N-body method using Particle-Mesh simulations. We apply established metrics such as matter power spectra and bispectra, topological indicators, and statistical independence of density fields and study their reproducibility and convergence properties. The U-Net approach performs well only for some of these physical metrics, and accuracy is worse at more nonlinear scales, where the dynamic range in density is larger. Motivated by field-level emulator studies utilizing Eulerian-aware loss functions and beyond pixel-wise mean squared errors, we examine a density-weighted objective. In the PM tests ending at z = 4, where the evolution is only mildly nonlinear, using the Eulerian matter density evaluated at each particle position as a direct weight on that particle’s Lagrangian displacement error improves the U-Net results at smaller scales; this result does not establish comparable accuracy in the strongly nonlinear low-redshift regime.

Monthly Notices of the Royal Astronomical Society
Argonne National Laboratory (US), University of Chicago (US)
U.S. Department of Energy, University of Chicago, Office of Science, Advanced Scientific Computing Research, High Energy Physics, Argonne National Laboratory
Cosmology and Gravitation Theories
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon