Literatur vom gleichen Autor/der gleichen Autor*in
plus bei Google Scholar

Bibliografische Daten exportieren
 

Multi-fidelity transfer learning for quantum chemical data using a robust density functional tight binding baseline

Titelangaben

Cui, Mengnan ; Reuter, Karsten ; Margraf, Johannes T.:
Multi-fidelity transfer learning for quantum chemical data using a robust density functional tight binding baseline.
In: Machine Learning: Science and Technology. Bd. 6 (2025) . - 015071.
ISSN 2632-2153
DOI: https://doi.org/10.1088/2632-2153/adc222

Volltext

Link zum Volltext (externe URL): Volltext

Abstract

Machine learning has revolutionized the development of interatomic potentials over the past decade, offering unparalleled computational speed without compromising accuracy. However, the performance of these models is highly dependent on the quality and amount of training data. Consequently, the current scarcity of high-fidelity datasets (i.e. beyond semilocal density functional theory) represents a significant challenge for further improvement. To address this, this study investigates the performance of transfer learning (TL) across multiple fidelities for both molecules and materials. Crucially, we disentangle the effects of multiple fidelities and different configuration/chemical spaces for pre-training and fine-tuning, in order to gain a deeper understanding of TL for chemical applications. This reveals that negative transfer, driven by noise from low-fidelity methods such as a density functional tight binding baseline, can significantly impact fine-tuned models. Despite this, the multi-fidelity approach demonstrates superior performance compared to single-fidelity learning. Interestingly, it even outperforms TL based on foundation models in some cases, by leveraging an optimal overlap of pre-training and fine-tuning chemical spaces.

Weitere Angaben

Publikationsform: Artikel in einer Zeitschrift
Begutachteter Beitrag: Ja
Institutionen der Universität: Fakultäten > Fakultät für Biologie, Chemie und Geowissenschaften > Fachgruppe Chemie > Lehrstuhl Physikalische Chemie V - Theorie und Maschinelles Lernen
Fakultäten > Fakultät für Biologie, Chemie und Geowissenschaften > Fachgruppe Chemie > Lehrstuhl Physikalische Chemie V - Theorie und Maschinelles Lernen > Lehrstuhl Physikalische Chemie V - Theorie und Maschinelles Lernen - Univ.-Prof. Dr. Johannes Theo Margraf
Forschungseinrichtungen > Zentrale wissenschaftliche Einrichtungen > Bayerisches Zentrum für Batterietechnik - BayBatt
Titel an der UBT entstanden: Ja
Themengebiete aus DDC: 500 Naturwissenschaften und Mathematik > 540 Chemie
Eingestellt am: 31 Mär 2025 05:54
Letzte Änderung: 31 Mär 2025 05:54
URI: https://eref.uni-bayreuth.de/id/eprint/93049