Les Houches guide to reusable ML models in LHC analyses

With the increasing usage of machine-learning in high-energy physics analyses, the publication of the trained models in a reusable form has become a crucial question for analysis preservation and reuse. The complexity of these models creates practical issues for both reporting them accurately and for ensuring the stability of their behaviours in different environments and over extended timescales. In this note we discuss the current state of affairs, highlighting specific practical issues and focusing on the most promising technical and strategic approaches to ensure trustworthy analysis-preservation. This material originated from discussions in the LHC Reinterpretation Forum and the 2023 PhysTeV workshop at Les Houches.

Authors

Institutions

Publication Details

Journal
SciPost Physics Community Reports
Published
2024-11-06
DOI
https://doi.org/10.21468/scipostphyscommrep.3
Citations
5
Primary Topic
Particle physics theoretical and experimental studies
Type
article
Field-Weighted Citation Impact
3.30

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Les Houches guide to reusable ML models in LHC analyses

Jack Y. Araz, Gregor Kasieczka, Humberto Reyes-González, T. Procter et al.
5 citations
SciPost Physics Community Reports
Particle physics theoretical and experimental studies
3.30
article

Les Houches guide to reusable ML models in LHC analyses

Jack Y. Araz, Gregor Kasieczka, Humberto Reyes-González, T. Procter, S. Sekmen, Krzysztof Rolbiecki, André Lessa, Anders Kvellestad, Sabine Kraml, A. G. Buckley, Are Raklev, G. Ünel, J. Kieseler
article en
5 citations

Abstract

With the increasing usage of machine-learning in high-energy physics analyses, the publication of the trained models in a reusable form has become a crucial question for analysis preservation and reuse. The complexity of these models creates practical issues for both reporting them accurately and for ensuring the stability of their behaviours in different environments and over extended timescales. In this note we discuss the current state of affairs, highlighting specific practical issues and focusing on the most promising technical and strategic approaches to ensure trustworthy analysis-preservation. This material originated from discussions in the LHC Reinterpretation Forum and the 2023 PhysTeV workshop at Les Houches.

SciPost Physics Community Reports
Karlsruhe Institute of Technology (DE), Institut polytechnique de Grenoble (FR), Centre National de la Recherche Scientifique (FR), Universität Hamburg (DE), University of Oslo (NO), University of California, Irvine (US), Thomas Jefferson National Accelerator Facility (US), Kyungpook National University (KR), Istituto Nazionale di Fisica Nucleare, Sezione di Genova (IT), Laboratoire de Physique Subatomique et de Cosmologie (FR), University of Warsaw (PL), Universidade Federal do ABC (BR), University of Glasgow (GB), University of Genoa (IT), RWTH Aachen University (DE), Université Grenoble Alpes (FR)
Alexander von Humboldt-Stiftung, National Research Foundation, Deutsche Forschungsgemeinschaft, Fundação de Amparo à Pesquisa do Estado de São Paulo, Ministero dell’Istruzione, dell’Università e della Ricerca, National Research Foundation of Korea, Narodowe Centrum Nauki, Narodowym Centrum Nauki, Centre National de la Recherche Scientifique, Norges Forskningsråd, Conseil National de la Recherche Scientifique, Engineering and Physical Sciences Research Council, Science and Technology Facilities Council, Norway Grants, Institut National de Physique Nucléaire et de Physique des Particules
Openalex Percentile: Top 6%
Particle physics theoretical and experimental studies
3.30
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.