Sprecher
Beschreibung
Following the completion of its neutrino mass measurement program at the end of 2025, the KATRIN experiment aims to probe keV-scale sterile neutrinos by analyzing the full tritium beta decay spectrum with a novel detector system, TRISTAN. Leveraging KATRIN's high source activity, this search is sensitive to mixing amplitudes at the parts-per-million level. However, extracting a potential sterile neutrino signature is challenging, as it relies on detailed modeling of the observed tritium spectrum and requires computationally intensive Monte Carlo simulations. To address this challenge, we implement neural simulation-based inference using normalizing flows to approximate the underlying probability density of the physics simulation. We demonstrate that continuous normalizing flows trained via conditional flow-matching enable modeling of experimental spectra and discuss the challenges involved. Furthermore, we cast the signal extraction as a neural ratio estimation problem to infer the sterile-neutrino mixing amplitude directly from the spectrum without an explicit likelihood, complemented by a model-independent cross-check of any candidate signature.