Divide-and-conquer potentials enable scalable and accurate predictions of forces and energies in atomistic systems

Claudio Zeni*, Andrea Anelli, Aldo Glielmo, Stefano de Gironcoli, K.R. Rossi*

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

6 Downloads (Pure)

Abstract

In committee of experts strategies, small datasets are extracted from a larger one and utilised for the training of multiple models. These models' predictions are then carefully weighted so as to obtain estimates which are dominated by the model(s) that are most informed in each domain of the data manifold. Here, we show how this divide-and-conquer philosophy provides an avenue in the making of machine learning potentials for atomistic systems, which is general across systems of different natures and efficiently scalable by construction. We benchmark this approach on various datasets and demonstrate that divide-and-conquer linear potentials are more accurate than their single model counterparts, while incurring little to no extra computational cost.
Original languageEnglish
Pages (from-to)113-121
Number of pages9
JournalDigital Discovery
Volume3 (2024)
Issue number1
DOIs
Publication statusPublished - 2023

Fingerprint

Dive into the research topics of 'Divide-and-conquer potentials enable scalable and accurate predictions of forces and energies in atomistic systems'. Together they form a unique fingerprint.

Cite this