TY  - JOUR
PB  - SciPost Foundation
DO  - 10.21468/SciPostPhysCore.8.2.037
TI  - Optimal, fast, and robust inference of reionization-era cosmology with the 21cmPIE-INN
PY  - 2025/04/10
UR  - https://scipost.org/SciPostPhysCore.8.2.037
JF  - SciPost Physics Core
JA  - SciPost Phys. Core
VL  - 8
IS  - 2
SP  - 037
A1  - Schosser, Benedikt
AU  - Heneka, Caroline
AU  - Plehn, Tilman
AB  - 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.
ER  -

@Article{10.21468/SciPostPhysCore.8.2.037,
	title={{Optimal, fast, and robust inference of reionization-era cosmology with the 21cmPIE-INN}},
	author={Benedikt Schosser and Caroline Heneka and Tilman Plehn},
	journal={SciPost Phys. Core},
	volume={8},
	pages={037},
	year={2025},
	publisher={SciPost},
	doi={10.21468/SciPostPhysCore.8.2.037},
	url={https://scipost.org/10.21468/SciPostPhysCore.8.2.037},
}