Data Generation for Machine Learning Interatomic Potentials and Beyond

Kulichenko, Maksim; Nebgen, Benjamin; Lubbers, Nicholas; Smith, Justin S.; Barros, Kipton; Allen, Alice E. A.; Habib, Adela; Shinkle, Emily; Fedik, Nikita; Li, Ying Wai; Messerly, Richard A.; Tretiak, Sergei

doi:10.1021/acs.chemrev.4c00572

reviewChemical ReviewsNov 21, 2024GREEN OA

Data Generation for Machine Learning Interatomic Potentials and Beyond

MKMaksim Kulichenko BNBenjamin Nebgen NLNicholas Lubbers JSJustin S. Smith KBKipton Barros

Los Alamos National Laboratory · Statistical Service · +2 more institutions

PubMed

Indexed incrossrefpubmed

Abstract

The field of data-driven chemistry is undergoing an evolution, driven by innovations in machine learning models for predicting molecular properties and behavior. Recent strides in ML-based interatomic potentials have paved the way for accurate modeling of diverse chemical and structural properties at the atomic level. The key determinant defining MLIP reliability remains the quality of the training data. A paramount challenge lies in constructing training sets that capture specific domains in the vast chemical and structural space. This Review navigates the intricate landscape of essential components and integrity of training data that ensure the extensibility and transferability of the resulting models. We…

Citation impact

122

total citations

FWCI: 13.24
Percentile: 100%
References: 345

Citations per year

Authors

12

Topics & keywords

Topics

Keywords

Chemical space
Transferability
Reliability (semiconductor)
Field (mathematics)
Space (punctuation)
Computer science
Data acquisition
Data science

No related works found for this paper.