Neural Posterior Unfolding

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Acosta, Fernando Torales, Chan, Jay, Desai, Krish, Mikuni, Vinicius, Nachman, Benjamin, Pan, Jingjing, Rubbo, Francesco
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911142557777920
author Acosta, Fernando Torales
Chan, Jay
Desai, Krish
Mikuni, Vinicius
Nachman, Benjamin
Pan, Jingjing
Rubbo, Francesco
author_facet Acosta, Fernando Torales
Chan, Jay
Desai, Krish
Mikuni, Vinicius
Nachman, Benjamin
Pan, Jingjing
Rubbo, Francesco
contents Differential cross section measurements are the currency of scientific exchange in particle and nuclear physics. A key challenge for these analyses is the correction for detector distortions, known as deconvolution or unfolding. Binned unfolding of cross section measurements traditionally rely on the regularized inversion of the response matrix that represents the detector response, mapping pre-detector (`particle level') observables to post-detector (`detector level') observables. In this paper we introduce Neural Posterior Unfolding, a modern, Bayesian approach that leverages normalizing flows for unfolding. By using normalizing flows for neural posterior estimation, NPU offers several key advantages including implicit regularization through the neural network architecture, fast amortized inference that eliminates the need for repeated retraining, and direct access to the full uncertainty in the unfolded result. In addition to introducing NPU, we implement a classical Bayesian unfolding method called Fully Bayesian Unfolding (FBU) in modern Python so it can also be studied. These tools are validated on simple Gaussian examples and then tested on simulated jet substructure examples from the Large Hadron Collider (LHC). We find that the Bayesian methods are effective and worth additional development to be analysis ready for cross section measurements at the LHC and beyond.
format Preprint
id arxiv_https___arxiv_org_abs_2509_06370
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neural Posterior Unfolding
Acosta, Fernando Torales
Chan, Jay
Desai, Krish
Mikuni, Vinicius
Nachman, Benjamin
Pan, Jingjing
Rubbo, Francesco
High Energy Physics - Phenomenology
High Energy Physics - Experiment
Differential cross section measurements are the currency of scientific exchange in particle and nuclear physics. A key challenge for these analyses is the correction for detector distortions, known as deconvolution or unfolding. Binned unfolding of cross section measurements traditionally rely on the regularized inversion of the response matrix that represents the detector response, mapping pre-detector (`particle level') observables to post-detector (`detector level') observables. In this paper we introduce Neural Posterior Unfolding, a modern, Bayesian approach that leverages normalizing flows for unfolding. By using normalizing flows for neural posterior estimation, NPU offers several key advantages including implicit regularization through the neural network architecture, fast amortized inference that eliminates the need for repeated retraining, and direct access to the full uncertainty in the unfolded result. In addition to introducing NPU, we implement a classical Bayesian unfolding method called Fully Bayesian Unfolding (FBU) in modern Python so it can also be studied. These tools are validated on simple Gaussian examples and then tested on simulated jet substructure examples from the Large Hadron Collider (LHC). We find that the Bayesian methods are effective and worth additional development to be analysis ready for cross section measurements at the LHC and beyond.
title Neural Posterior Unfolding
topic High Energy Physics - Phenomenology
High Energy Physics - Experiment
url https://arxiv.org/abs/2509.06370