Three-dimensional Conditional Diffusion Models for Cosmological 21 cm Lightcone Emulation
Abstract
We investigate conditional diffusion modeling for three-dimensional 21 cm lightcone emulation, focusing on cubes with a sky-plane size of 64×64 and a line-of-sight depth up to 1024 cells. Relative to earlier 2D studies, the 3D setting is substantially harder because memory limits enforce very small micro-batches while the underlying voxel distribution is highly skewed and long tailed. We perform controlled comparisons across preprocessing choices, dynamic-range compression settings, architecture depth, and training duration using 25,600 training lightcones and validation ensembles at fixed parameter points. For validation, each reference parameter point contains 800 21cmFAST realizations with independent initial conditions, and we use 800 samples per model and per reference set for the reported ensemble comparisons. We evaluate generated lightcones with complementary diagnostics in both image and summary-statistic spaces: brightness-temperature slices, the global signal, the power spectrum, and reduced scattering coefficients. Across the tested configurations, preprocessing is the dominant factor governing stable training and the resulting physical fidelity. Among the configurations explored here, Yeo-Johnson preprocessing combined with moderate amplitude compression gives the most consistently favorable trade-off, with the strongest quantitative support coming from rankings based on the standard-deviation-normalized mean absolute error (MAEstd) of the global signal and qualitatively compatible behavior in the complementary diagnostics. At the same time, visually plausible 3D samples still retain measurable biases in two-point and higher-order statistics. We therefore view the present work as a simulation-level baseline for three-dimensional 21 cm emulation and for future studies that incorporate more realistic observational effects.
Versions
Related
- Extra-dimensional Origins of Chemical Potentials at the Cosmological Collider
- A Two-Dimensional Test of Cosmological Parity Violation with the 3-Point Correlation Function
- BINDing the lightcone: A suite of astrophysical ray-traced weak lensing and SZ maps
- Introducing the GION Lightcone Simulation Suite for Multi-Wavelength Large-Scale Structure Cosmology
- Evolving only the observed spacetime region: perfectly parallel lightcone simulations with sCOLA
BibTeX
@misc{xia2026threedimensional,
title = {{Three-dimensional Conditional Diffusion Models for Cosmological 21 cm Lightcone Emulation}},
author = {Bin Xia and John H. Wise},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2605.29016},
eprint = {2605.29016},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.29016},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Three-dimensional Conditional Diffusion Models for Cosmological 21 cm Lightcone Emulation
AU - Bin Xia
AU - John H. Wise
PY - 2026
DA - 2026/08/31
PB - arXiv
DO - 10.48550/arxiv.2605.29016
UR - https://arxiv.org/abs/2605.29016
AB - We investigate conditional diffusion modeling for three-dimensional 21 cm lightcone emulation, focusing on cubes with a sky-plane size of 64×64 and a line-of-sight depth up to 1024 cells. Relative to earlier 2D studies, the 3D setting is substantially harder because memory limits enforce very small micro-batches while the underlying voxel distribution is highly skewed and long tailed. We perform controlled comparisons across preprocessing choices, dynamic-range compression settings, architecture depth, and training duration using 25,600 training lightcones and validation ensembles at fixed parameter points. For validation, each reference parameter point contains 800 21cmFAST realizations with independent initial conditions, and we use 800 samples per model and per reference set for the reported ensemble comparisons. We evaluate generated lightcones with complementary diagnostics in both image and summary-statistic spaces: brightness-temperature slices, the global signal, the power spectrum, and reduced scattering coefficients. Across the tested configurations, preprocessing is the dominant factor governing stable training and the resulting physical fidelity. Among the configurations explored here, Yeo-Johnson preprocessing combined with moderate amplitude compression gives the most consistently favorable trade-off, with the strongest quantitative support coming from rankings based on the standard-deviation-normalized mean absolute error (MAE_std) of the global signal and qualitatively compatible behavior in the complementary diagnostics. At the same time, visually plausible 3D samples still retain measurable biases in two-point and higher-order statistics. We therefore view the present work as a simulation-level baseline for three-dimensional 21 cm emulation and for future studies that incorporate more realistic observational effects.
ER -