Understanding multi-fidelity training of machine-learned force-fields
Machine Learning: Science and Technology · 01.08.2026 · 7 (4) · 045057
DOI: 10.1088/2632-2153/ae8e2e ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning: Science and Technology, online erschienen am 18.08.2026. Verfasst von John L A Gardner und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 90 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study systematically investigates two multi-fidelity strategies used to train machine-learned force fields (MLFFs)—pre-training/fine-tuning and multi-headed training—and elucidates the mechanisms underpinning their success. For pre-training and fine-tuning, we uncover a log–log linear relationship between pre-trained and fine-tuned accuracies that holds across model architectures, model sizes, and quantum-chemical methods. The success of this approach hinges on the quantity and quality of available pre-training data, and, critically, the inclusion of force labels. We demonstrate that pre-trained representations are inherently method-specific, requiring adaptation of the model backbone during fine-tuning. In contrast, multi-headed models learn method-independent backbone representations, where again the heads’ accuracies are log–log linearly related. Relative to pre-training and fine-tuning, these shared representations marginally reduce model performance in most cases. However, this trade-off is offset by practical advantages: multi-headed training extends naturally to multiple labelling methods and enables partial replacement of expensive labels with cheaper alternatives, paving the way towards cost-efficient universal MLFFs.
Zitieren
APA
Gardner, J. L. A., Schulz, H., Hélie, J., Sun, L., & Simm, G. N. C. (2026). Understanding multi-fidelity training of machine-learned force-fields. Machine Learning: Science and Technology, 7(4), 045057. https://doi.org/10.1088/2632-2153/ae8e2e
Vancouver
Gardner JLA, Schulz H, Hélie J, Sun L, Simm GNC. Understanding multi-fidelity training of machine-learned force-fields. Machine Learning: Science and Technology. 2026;7(4):045057. doi:10.1088/2632-2153/ae8e2e
BibTeX
@article{Gardner2026,
author = {Gardner, John L A and Schulz, Hannes and Hélie, Jean and Sun, Lixin and Simm, Gregor N C},
title = {Understanding multi-fidelity training of machine-learned force-fields},
journal = {Machine Learning: Science and Technology},
year = {2026},
volume = {7},
number = {4},
pages = {045057},
doi = {10.1088/2632-2153/ae8e2e}
}
EUVIMED