Papers

Optimal, fast, and robust inference of reionization-era cosmology with the 21cmPIE-INN

B Schosser, C Heneka, T Plehn - arXiv preprint arXiv:2401.04174, 2024 - arxiv.org
Astrophysics paper astro-ph.CO Suggest

… Our goal is to use simulation-based inference to extract as much information as possible from complex 21cm data and avoid the bottlenecks of classic analysis methods. …

Link to paper

BibTeX

@article{2401.04174v3,
Author        = {Benedikt Schosser and Caroline Heneka and Tilman Plehn},
Title         = {Optimal, fast, and robust inference of reionization-era cosmology with
  the 21cmPIE-INN},
Eprint        = {2401.04174v3},
DOI           = {10.21468/SciPostPhysCore.8.2.037},
ArchivePrefix = {arXiv},
PrimaryClass  = {astro-ph.CO},
Abstract      = {Modern machine learning will allow for simulation-based inference from
reionization-era 21cm observations at the Square Kilometre Array. Our framework
combines a convolutional summary network and a conditional invertible network
through a physics-inspired latent representation. It allows for an efficient
and extremely fast determination of the posteriors of astrophysical and
cosmological parameters, jointly with well-calibrated and on average unbiased
summaries. The sensitivity to non-Gaussian information makes our method a
promising alternative to the established power spectra.},
Year          = {2024},
Month         = {Jan},
Note          = {SciPost Phys. Core 8, 037 (2025)},
Url           = {http://arxiv.org/abs/2401.04174v3},
File          = {2401.04174v3.pdf}
}

Share