PINNferring the Hubble Function with Uncertainties

Autor:	Röver, Lennart, Schäfer, Björn Malte, Plehn, Tilman
Rok vydání:	2024
Předmět:	Astrophysics - Cosmology and Nongalactic Astrophysics Astrophysics - Instrumentation and Methods for Astrophysics High Energy Physics - Phenomenology
Druh dokumentu:	Working Paper
Popis:	The Hubble function characterizes a given Friedmann-Robertson-Walker spacetime and can be related to the densities of the cosmological fluids and their equations of state. We show how physics-informed neural networks (PINNs) emulate this dynamical system and provide fast predictions of the luminosity distance for a given choice of densities and equations of state, as needed for the analysis of supernova data. We use this emulator to perform a model-independent and parameter-free reconstruction of the Hubble function on the basis of supernova data. As part of this study, we develop and validate an uncertainty treatment for PINNs using a heteroscedastic loss and repulsive ensembles. Comment: 21 pages, 12 figures
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2403.13899 Zobrazit plný text záznamu View this record from Arxiv