Les Houches guide to reusable ML models in LHC analyses

TL;DR

This paper discusses the challenges and strategies for making machine-learning models in high-energy physics analyses reusable and trustworthy, emphasizing technical and strategic solutions for analysis preservation.

Contribution

It provides a comprehensive overview of practical issues and promising approaches for preserving and reusing ML models in LHC analyses.

Findings

Highlights practical issues in ML model reporting and stability.

Identifies promising technical solutions for model preservation.

Emphasizes strategic approaches for trustworthy analysis reuse.

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.