Enabling stable preservation of ML algorithms in high-energy physics with petrifyML

2025/09/15 by A. G. Buckley, Buckley, Andy, Louie Corpe +5
Computer Science · Decision Sciences · #Advanced Data Storage Technologies #Computational Physics and Python Applications #Scientific Computing and Data Management

paper · pdf · doi:10.48550/arxiv.2509.11830

Abstract

Machine learning (ML) in high-energy physics (HEP) has moved in the LHC era from an internal detail of experiment software, to an unavoidable public component of many physics data analyses. Scientific reproducibility thus requires that it be possible to accurately and stably preserve the behaviours of these, sometimes very complex algorithms. We present and document the petrifyML package, which provides missing mechanisms to convert configurations from commonly used HEP ML tools to either the industry-standard ONNX format or to native Python or C++ code, enabling future re-use and re-interpretation of many ML-based experimental studies.

Citations

• Rivet
• Array programming with NumPy
• Fast inference of deep neural networks in FPGAs for particle physics

Related

• HEP ML Lab: An end-to-end framework for applying machine learning into phenomenology studies
• Snowmass 2021 Computational Frontier CompF03 Topical Group Report: Machine Learning
• Machine Learning for Columnar High Energy Physics Analysis
• Machine Learning Solutions for High Energy Physics: Applications to\n Electromagnetic Shower Generation, Flavor Tagging, and the Search for\n di-Higgs Production
• MLaaS4HEP: Machine Learning as a Service for HEP