Comparison of Bayesian and particle swarm algorithms for hyperparameter optimisation in machine learning applications in high energy physics

When using machine learning (ML) techniques, users typically need to choose a plethora of algorithm-specific parameters, referred to as hyperparameters. In this paper, we compare the performance of two algorithms, particle swarm optimisation (PSO) and Bayesian optimisation (BO), for the autonomous d...

Full description

Saved in:
Bibliographic Details
Published inComputer physics communications Vol. 294; p. 108955
Main Authors Tani, Laurits, Veelken, Christian
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.01.2024
Subjects
Evolutionary algorithms
High energy physics
Hyperparameter optimization
Machine learning
Hyperparameter optimization
High energy physics
Evolutionary algorithms
Machine learning
Online AccessGet full text
ISSN0010-4655
1879-2944
DOI10.1016/j.cpc.2023.108955

Cover

More Information
Summary:When using machine learning (ML) techniques, users typically need to choose a plethora of algorithm-specific parameters, referred to as hyperparameters. In this paper, we compare the performance of two algorithms, particle swarm optimisation (PSO) and Bayesian optimisation (BO), for the autonomous determination of these hyperparameters in applications to different ML tasks typical for the field of high energy physics (HEP). Our evaluation of the performance includes a comparison of the capability of the PSO and BO algorithms to make efficient use of the highly parallel computing resources that are characteristic of contemporary HEP experiments.
ISSN:0010-4655
1879-2944
DOI:10.1016/j.cpc.2023.108955