The role of hyperparameters in machine learning models and how to tune them

Arnold, Christian ORCID: https://orcid.org/0000-0002-7042-594X, Biedebach, Luka, Küpfer, Andreas and Neunhoeffer, Marcel 2024. The role of hyperparameters in machine learning models and how to tune them. Political Science Research and Methods 12 (4) , pp. 841-848. 10.1017/psrm.2023.61

Preview

PDF - Published Version Available under License Creative Commons Attribution. Download (145kB) | Preview

Abstract

Hyperparameters critically influence how well machine learning models perform on unseen, out-of-sample data. Systematically comparing the performance of different hyperparameter settings will often go a long way in building confidence about a model's performance. However, analyzing 64 machine learning related manuscripts published in three leading political science journals (APSR, PA, and PSRM) between 2016 and 2021, we find that only 13 publications (20.31 percent) report the hyperparameters and also how they tuned them in either the paper or the appendix. We illustrate the dangers of cursory attention to model and tuning transparency in comparing machine learning models’ capability to predict electoral violence from tweets. The tuning of hyperparameters and their documentation should become a standard component of robustness checks for machine learning models.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Cardiff Law & Politics
Publisher:	Cambridge University Press
ISSN:	2049-8470
Date of First Compliant Deposit:	13 July 2023
Date of Acceptance:	5 July 2023
Last Modified:	11 Nov 2024 14:51
URI:	https://orca.cardiff.ac.uk/id/eprint/160937

Actions (repository staff only)

Edit Item