Neural network-based approach to phase space integration

Monte Carlo methods are widely used in particle physics to integrate and sample probability distributions (differential cross sections or decay rates) on multi-dimensional phase spaces. We present a Neural Network (NN) algorithm optimized to perform this task. The algorithm has been applied to se...

Full description

Bibliographic Details
Main Author: Matthew D. Klimek, Maxim Perelstein
Format: Article
Language:English
Published: SciPost 2020-10-01
Series:SciPost Physics
Online Access:https://scipost.org/SciPostPhys.9.4.053
id doaj-5335e0d385ae44aea11f81afdc308fc7
record_format Article
spelling doaj-5335e0d385ae44aea11f81afdc308fc72020-11-25T03:41:37ZengSciPostSciPost Physics2542-46532020-10-019405310.21468/SciPostPhys.9.4.053Neural network-based approach to phase space integrationMatthew D. Klimek, Maxim PerelsteinMonte Carlo methods are widely used in particle physics to integrate and sample probability distributions (differential cross sections or decay rates) on multi-dimensional phase spaces. We present a Neural Network (NN) algorithm optimized to perform this task. The algorithm has been applied to several examples of direct relevance for particle physics, including situations with non-trivial features such as sharp resonances and soft/collinear enhancements. Excellent performance has been demonstrated in all examples, with the properly trained NN achieving unweighting efficiencies of between 30% and 75%. In contrast to traditional Monte Carlo algorithms such as VEGAS, the NN-based approach does not require that the phase space coordinates be aligned with resonant or other features in the cross section.https://scipost.org/SciPostPhys.9.4.053
collection DOAJ
language English
format Article
sources DOAJ
author Matthew D. Klimek, Maxim Perelstein
spellingShingle Matthew D. Klimek, Maxim Perelstein
Neural network-based approach to phase space integration
SciPost Physics
author_facet Matthew D. Klimek, Maxim Perelstein
author_sort Matthew D. Klimek, Maxim Perelstein
title Neural network-based approach to phase space integration
title_short Neural network-based approach to phase space integration
title_full Neural network-based approach to phase space integration
title_fullStr Neural network-based approach to phase space integration
title_full_unstemmed Neural network-based approach to phase space integration
title_sort neural network-based approach to phase space integration
publisher SciPost
series SciPost Physics
issn 2542-4653
publishDate 2020-10-01
description Monte Carlo methods are widely used in particle physics to integrate and sample probability distributions (differential cross sections or decay rates) on multi-dimensional phase spaces. We present a Neural Network (NN) algorithm optimized to perform this task. The algorithm has been applied to several examples of direct relevance for particle physics, including situations with non-trivial features such as sharp resonances and soft/collinear enhancements. Excellent performance has been demonstrated in all examples, with the properly trained NN achieving unweighting efficiencies of between 30% and 75%. In contrast to traditional Monte Carlo algorithms such as VEGAS, the NN-based approach does not require that the phase space coordinates be aligned with resonant or other features in the cross section.
url https://scipost.org/SciPostPhys.9.4.053
work_keys_str_mv AT matthewdklimekmaximperelstein neuralnetworkbasedapproachtophasespaceintegration
_version_ 1724529163081089024