On the reproducibility of free energy surfaces in machine-learned collective variable spaces
The Journal of Chemical Physics · 14.10.2025 · 163 (14)
DOI: 10.1063/5.0287912 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 13.10.2025. Verfasst von Florian M Dietrich und 1 weiteren Autor:innen. Beteiligte Einrichtung: Thomas Young Centre and Department of Chemical Engineering, University College London, London WC1E 7JE, United Kingdom. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 41 Literaturangaben. Gefördert von: Engineering and Physical Sciences Research Council, XtalPi.
Abstract (Originalsprache)
As Machine-Learned Collective Variables (MLCVs) are becoming increasingly relevant in the molecular simulation literature, we discuss the necessary conditions to enable reproducibility in calculating and representing free energy surfaces. We note that the variability of the training process and the roughness of the hyperparameter space impose inherent limits on the reproducibility of results even when the mathematical structure of the model defining a collective variable is consistent. To this end, we propose the adoption of a geometric (gauge invariant) free energy representation to obtain consistent free energy differences across training instances and architectures. Furthermore, we introduce a normalization factor to model gradients for biased enhanced sampling. This factor effectively unifies free energy definitions and addresses practical issues preventing the widespread use and deployment of MLCVs.
Autor:innen und Institutionen
- Thomas Young Centre and Department of Chemical Engineering, University College London, London WC1E 7JE, United Kingdom.
ORCID-Profile: Florian M Dietrich, Matteo Salvalaglio
Förderung
- Engineering and Physical Sciences Research Council
- XtalPi
Zitieren
APA
Dietrich, F. M., & Salvalaglio, M. (2025). On the reproducibility of free energy surfaces in machine-learned collective variable spaces. The Journal of Chemical Physics, 163(14). https://doi.org/10.1063/5.0287912
Vancouver
Dietrich FM, Salvalaglio M. On the reproducibility of free energy surfaces in machine-learned collective variable spaces. The Journal of Chemical Physics. 2025;163(14). doi:10.1063/5.0287912
BibTeX
@article{Dietrich2025,
author = {Dietrich, Florian M and Salvalaglio, Matteo},
title = {On the reproducibility of free energy surfaces in machine-learned collective variable spaces},
journal = {The Journal of Chemical Physics},
year = {2025},
volume = {163},
number = {14},
doi = {10.1063/5.0287912}
}
EUVIMED