arXiv · 2502.09494
Communicating Likelihoods with Normalising Flows
Abstract
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 of the joint distribution, 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.
Explore related subjects
Keep this discovery
Jack Y. Araz, Anja Beck, Méril Reboud, Michael Spannowsky, Danny van Dyk. 2025-02-13. Communicating Likelihoods with Normalising Flows. https://doi.org/10.1140/epjc%2Fs10052-026-16045-9
Cite the original work for its findings. Save a collection to share your selection of sources.