Boost invariant polynomials for efficient jet tagging

Munoz, Jose M and Batatia, Ilyes and Ortner, Christoph (2022) Boost invariant polynomials for efficient jet tagging. Machine Learning: Science and Technology, 3 (4). 04LT05. ISSN 2632-2153

[thumbnail of Munoz_2022_Mach._Learn.__Sci._Technol._3_04LT05.pdf] Text
Munoz_2022_Mach._Learn.__Sci._Technol._3_04LT05.pdf - Published Version

Download (657kB)

Abstract

Given the vast amounts of data generated by modern particle detectors, computational efficiency is essential for many data-analysis jobs in high-energy physics. We develop a new class of physically interpretable boost invariant polynomial (BIP) features for jet tagging that achieves such efficiency. We show that, for both supervised and unsupervised tasks, integrating BIPs with conventional classification techniques leads to models achieving high accuracy on jet tagging benchmarks while being orders of magnitudes faster to train and evaluate than contemporary deep learning systems.

Item Type: Article
Subjects: Eurolib Press > Multidisciplinary
Depositing User: Managing Editor
Date Deposited: 14 Jul 2023 04:20
Last Modified: 29 Sep 2023 12:43
URI: http://info.submit4journal.com/id/eprint/2245

Actions (login required)

View Item
View Item