Communicating likelihoods with normalising flows

doi:10.1140/epjc/s10052-026-16045-9

Communicating likelihoods with normalising flows

2026-07-13

0

OA

AI

DOI：10.1140/epjc/s10052-026-16045-9

Original

Original request for help

Save

Abstract

En 中文

We present a machine-learning-based workflow to model an unbinned likelihood from its samples. A key advancement over existing approaches is the validation of the learned likelihood using rigorous statistical tests, such as the Kolmogorov–Smirnov test of the joint distribution. Our method enables the reliable communication of experimental and phenomenological likelihoods for subsequent analyses. We demonstrate its effectiveness through three case studies in high-energy physics. To support broader adoption, we provide an open-source reference implementation, nabu.

Communicating likelihoods with normalising flows

Abstract

Journal

Organization