Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

An implementation of neural simulation-based inference for parameter estimation in ATLAS

Aad, G. ; Åkesson, T.P.A. LU orcid ; Astrand, K.S.V. LU ; Doglioni, C. LU ; Ekman, P.A. LU orcid ; Hedberg, V. LU ; Herde, H. LU orcid ; Konya, B. LU ; Lytken, E. LU orcid and Poettgen, R. LU orcid , et al. (2025) In Reports on Progress in Physics 88(6).
Abstract
Neural simulation-based inference (NSBI) is a powerful class of machine-learning-based methods for statistical inference that naturally handles high-dimensional parameter estimation without the need to bin data into low-dimensional summary histograms. Such methods are promising for a range of measurements, including at the Large Hadron Collider, where no single observable may be optimal to scan over the entire theoretical phase space under consideration, or where binning data into histograms could result in a loss of sensitivity. This work develops a NSBI framework for statistical inference, using neural networks to estimate probability density ratios, which enables the application to a full-scale analysis. It incorporates a large number... (More)
Neural simulation-based inference (NSBI) is a powerful class of machine-learning-based methods for statistical inference that naturally handles high-dimensional parameter estimation without the need to bin data into low-dimensional summary histograms. Such methods are promising for a range of measurements, including at the Large Hadron Collider, where no single observable may be optimal to scan over the entire theoretical phase space under consideration, or where binning data into histograms could result in a loss of sensitivity. This work develops a NSBI framework for statistical inference, using neural networks to estimate probability density ratios, which enables the application to a full-scale analysis. It incorporates a large number of systematic uncertainties, quantifies the uncertainty due to the finite number of events in training samples, develops a method to construct confidence intervals, and demonstrates a series of intermediate diagnostic checks that can be performed to validate the robustness of the method. As an example, the power and feasibility of the method are assessed on simulated data for a simplified version of an off-shell Higgs boson couplings measurement in the four-lepton final states. This approach represents an extension to the standard statistical methodology used by the experiments at the Large Hadron Collider, and can benefit many physics analyses. © 2025 The Author(s). Published by IOP Publishing Ltd. (Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; ; and (Less)
author collaboration
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
frequentist statistics, likelihood-free inference, machine learning, neural simulation-based inference, parameter inference
in
Reports on Progress in Physics
volume
88
issue
6
article number
067801
publisher
IOP Publishing
external identifiers
  • scopus:105007089483
  • pmid:40315869
ISSN
0034-4885
DOI
10.1088/1361-6633/add370
language
English
LU publication?
yes
id
b608c76c-aee7-46de-8dcc-20c200434e07
date added to LUP
2026-03-09 16:37:30
date last changed
2026-03-10 03:49:37
@article{b608c76c-aee7-46de-8dcc-20c200434e07,
  abstract     = {{Neural simulation-based inference (NSBI) is a powerful class of machine-learning-based methods for statistical inference that naturally handles high-dimensional parameter estimation without the need to bin data into low-dimensional summary histograms. Such methods are promising for a range of measurements, including at the Large Hadron Collider, where no single observable may be optimal to scan over the entire theoretical phase space under consideration, or where binning data into histograms could result in a loss of sensitivity. This work develops a NSBI framework for statistical inference, using neural networks to estimate probability density ratios, which enables the application to a full-scale analysis. It incorporates a large number of systematic uncertainties, quantifies the uncertainty due to the finite number of events in training samples, develops a method to construct confidence intervals, and demonstrates a series of intermediate diagnostic checks that can be performed to validate the robustness of the method. As an example, the power and feasibility of the method are assessed on simulated data for a simplified version of an off-shell Higgs boson couplings measurement in the four-lepton final states. This approach represents an extension to the standard statistical methodology used by the experiments at the Large Hadron Collider, and can benefit many physics analyses. © 2025 The Author(s). Published by IOP Publishing Ltd.}},
  author       = {{Aad, G. and Åkesson, T.P.A. and Astrand, K.S.V. and Doglioni, C. and Ekman, P.A. and Hedberg, V. and Herde, H. and Konya, B. and Lytken, E. and Poettgen, R. and Smirnova, O. and Wallin, E.J. and Zwalinski, L.}},
  issn         = {{0034-4885}},
  keywords     = {{frequentist statistics; likelihood-free inference; machine learning; neural simulation-based inference; parameter inference}},
  language     = {{eng}},
  number       = {{6}},
  publisher    = {{IOP Publishing}},
  series       = {{Reports on Progress in Physics}},
  title        = {{An implementation of neural simulation-based inference for parameter estimation in ATLAS}},
  url          = {{http://dx.doi.org/10.1088/1361-6633/add370}},
  doi          = {{10.1088/1361-6633/add370}},
  volume       = {{88}},
  year         = {{2025}},
}