Uncertainty-Aware Neural Networks for Fuzzy Dark Matter Model Selection from \texorpdfstring{$x_{\rm HI}$}{x_HI} Measurements

Abstract

The nature of dark matter remains a central question in cosmology, with fuzzy dark matter (FDM) models offering a compelling alternative to the cold dark matter (CDM) paradigm. We explore FDM scenarios by performing 21-cm simulations across a parameter space with \texorpdfstring{$f_{\rm FDM} \in [0.02, 0.10]$}{f_FDM in [0.02, 0.10]} and \texorpdfstring{$m_{\rm FDM} \in [10^{-24}, 10^{-21}]\,\mathrm{eV}$}{m_FDM in [10^-24, 10^-21] eV}, obtaining global neutral hydrogen fractions (\texorpdfstring{$x_{\rm HI}$}{x_HI}) for each model. Observational \texorpdfstring{$x_{\rm HI}$}{x_HI} data and associated uncertainties from JWST are incorporated by estimating full probability density functions (PDFs) for both \texorpdfstring{$x_{\rm HI}$}{x_HI} and redshift \texorpdfstring{$z$}{z} using Bayesian inference with the No-U-Turn Sampler (NUTS), yielding non-Gaussian multivariate uncertainty distributions. A hybrid machine learning framework is then trained on these observational PDFs to learn both central values and correlated uncertainties in \texorpdfstring{$x_{\rm HI}$}{x_HI} and \texorpdfstring{$z$}{z}, iteratively refining its parameters in each training epoch through direct incorporation of the multivariate PDFs derived from observational constraints. We then compare the simulation outputs to the machine-learned observational trends to identify the most consistent models. Our results indicate that FDM models with \texorpdfstring{$m_{\rm FDM} \simeq 10^{-22}\,\mathrm{eV}$}{m_FDM approx 10^-22 eV} and \texorpdfstring{$f_{\rm FDM} \simeq 0.04$}{f_FDM approx 0.04} best match current data, while lighter masses are strongly constrained. By integrating simulations and machine learning in an uncertainty-aware framework, this work explores the physics of the early Universe and guides future studies of 21-cm cosmology and reionization.