Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments

A major task in particle physics is the measurement of rare signal processes. Even modest improvements in background rejection, at a fixed signal efficiency, can significantly enhance the measurement sensitivity. Building on prior research by others that incorporated physical symmetries into neural...

Full description

Saved in:
Bibliographic Details
Main Author: Luc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi Zhang
Format: Article
Language:English
Published: SciPost 2025-07-01
Series:SciPost Physics
Online Access:https://scipost.org/SciPostPhys.19.1.028
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849251658349936640
author Luc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi Zhang
author_facet Luc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi Zhang
author_sort Luc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi Zhang
collection DOAJ
description A major task in particle physics is the measurement of rare signal processes. Even modest improvements in background rejection, at a fixed signal efficiency, can significantly enhance the measurement sensitivity. Building on prior research by others that incorporated physical symmetries into neural networks, this work extends those ideas to include additional physics-motivated features. Specifically, we introduce energy-dependent particle interaction strengths, derived from leading-order SM predictions, into modern deep learning architectures, including Transformer Architectures (Particle Transformer), and Graph Neural Networks (Particle Net). These interaction strengths, represented as the SM interaction matrix, are incorporated into the attention matrix (transformers) and edges (graphs). Our results in event classification show that the integration of all physics-motivated features improves background rejection by $10\%-40\%$ over baseline models, with an additional gain of up to $9\%$ due to the SM interaction matrix. This study also provides one of the broadest comparisons of event classifiers to date, demonstrating how various architectures perform across this task. A simplified statistical analysis demonstrates that these enhanced architectures yield significant improvements in signal significance compared to a graph network baseline.
format Article
id doaj-art-3cdf7fa308a5413bb0d8fbe70e7f04c3
institution Kabale University
issn 2542-4653
language English
publishDate 2025-07-01
publisher SciPost
record_format Article
series SciPost Physics
spelling doaj-art-3cdf7fa308a5413bb0d8fbe70e7f04c32025-08-20T03:56:54ZengSciPostSciPost Physics2542-46532025-07-0119102810.21468/SciPostPhys.19.1.028Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experimentsLuc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi ZhangA major task in particle physics is the measurement of rare signal processes. Even modest improvements in background rejection, at a fixed signal efficiency, can significantly enhance the measurement sensitivity. Building on prior research by others that incorporated physical symmetries into neural networks, this work extends those ideas to include additional physics-motivated features. Specifically, we introduce energy-dependent particle interaction strengths, derived from leading-order SM predictions, into modern deep learning architectures, including Transformer Architectures (Particle Transformer), and Graph Neural Networks (Particle Net). These interaction strengths, represented as the SM interaction matrix, are incorporated into the attention matrix (transformers) and edges (graphs). Our results in event classification show that the integration of all physics-motivated features improves background rejection by $10\%-40\%$ over baseline models, with an additional gain of up to $9\%$ due to the SM interaction matrix. This study also provides one of the broadest comparisons of event classifiers to date, demonstrating how various architectures perform across this task. A simplified statistical analysis demonstrates that these enhanced architectures yield significant improvements in signal significance compared to a graph network baseline.https://scipost.org/SciPostPhys.19.1.028
spellingShingle Luc Builtjes, Sascha Caron, Polina Moskvitina, Clara Nellist, Roberto Ruiz de Austri, Rob Verheyen, Zhongyi Zhang
Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
SciPost Physics
title Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
title_full Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
title_fullStr Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
title_full_unstemmed Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
title_short Attention to the strengths of physical interactions: Transformer and graph-based event classification for particle physics experiments
title_sort attention to the strengths of physical interactions transformer and graph based event classification for particle physics experiments
url https://scipost.org/SciPostPhys.19.1.028
work_keys_str_mv AT lucbuiltjessaschacaronpolinamoskvitinaclaranellistrobertoruizdeaustrirobverheyenzhongyizhang attentiontothestrengthsofphysicalinteractionstransformerandgraphbasedeventclassificationforparticlephysicsexperiments