[Paper Review] A Living Review of Machine Learning for Particle Physics

M. Feickert, Benjamin Nachman|arXiv (Cornell University)|Feb 2, 2021

Particle physics theoretical and experimental studies1 references79 citations

TL;DR

This paper presents a living, continuously updated review that catalogs how machine learning is applied in high energy physics, organized by categories and open to community contributions.

ABSTRACT

Modern machine learning techniques, including deep learning, are rapidly being applied, adapted, and developed for high energy physics. Given the fast pace of this research, we have created a living review with the goal of providing a nearly comprehensive list of citations for those developing and applying these approaches to experimental, phenomenological, or theoretical analyses. As a living document, it will be updated as often as possible to incorporate the latest developments. A list of proper (unchanging) reviews can be found within. Papers are grouped into a small set of topics to be as useful as possible. Suggestions and contributions are most welcome, and we provide instructions for participating.

Motivation & Objective

• Provide a nearly comprehensive, up-to-date catalog of ML papers in high energy physics (HEP).
• Organize papers into topics and sub-categories to enhance discoverability for researchers.
• Encourage community contributions and maintainability through a GitHub-based workflow.
• Automate deployment of updated PDF/Markdown versions and provide BibTeX references.

Proposed method

• Create a living document that is continuously updated as new ML-HEP papers are published.
• Organize papers into categories and sub-categories with descriptions for discoverability.
• Automate generation and deployment of PDF/Markdown versions from a GitHub repository.
• Provide a BibTeX file for easy citation by authors.
• Offer guidelines and processes for community contributions via PRs (pull requests).
• Link updates to arXiv and potential Inspire synchronization for references in the future.

Experimental results

Research questions

• RQ1 How can ML papers in HEP be organized to maximize discoverability for researchers?
• RQ2 What processes enable a living, up-to-date review that incorporates new publications rapidly?
• RQ3 How can the community contribute effectively to a living review while maintaining quality and consistency?
• RQ4 What future automation is feasible to keep references current and synchronized with external databases?

Key findings

• The Living Review provides a nearly comprehensive list of citations for papers applying ML to experimental, phenomenological, or theoretical HEP analyses.
• Papers are grouped into topics and sub-categories to improve search efficiency and discoverability.
• The review is maintained as a living document with ongoing updates and community contributions.
• It is generated from a GitHub repository using LaTeX-based CI to produce PDF and Markdown versions deployed publicly.
• A BibTeX file is provided to facilitate citation in new publications.
• Contributions are guided by a documented PR workflow to ensure smooth collaboration.