Frequentist uncertainties on neural density ratios with w i f i ensembles
Year: 2025
Type: article
Abstract: We introduce w i f i ensembles as a novel framework to obtain asymptotic frequentist uncertainties on density ratios, with a particular focus on neural ratio estimation in the context of high-energy physics. When the density ratio of interest is a likelihood ratio conditioned on parameters, w i f i ensembles can be used to perform simulation-based inference on those parameters. After training the basis functions f i ( x ) , uncertainties on the weights w i can be straightforwardly propagated to the estimated parameters without requiring extraneous bootstraps. To demonstrate this approach, we present an application in quantum chromodynamics at the Large Hadron Collider, using w i f i ensembles to estimate the likelihood ratio between generated quark and gluon jets. We use this learned likelihood ratio to estimate the quark fraction in a synthetic mixed quark/gluon sample, showing that the resultant uncertainties empirically satisfy the desired coverage properties.
DOI: 10.1103/w28w-x5wh
Authors: Sean Benevedes, Jesse Thaler
Language: English
Subfield: Nuclear and High Energy Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4414345732
