Spectral operator representations

Abstract Machine learning in atomistic materials science has grown to become a powerful tool, with most approaches focusing on atomic geometry, typically decomposed into local atomic environments. This approach, while well-suited for machine-learned interatomic potentials, is conceptually at odds wi...

Full description

Saved in:
Bibliographic Details
Main Authors: Austin Zadoks, Antimo Marrazzo, Nicola Marzari
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:npj Computational Materials
Online Access:https://doi.org/10.1038/s41524-024-01446-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846136944457678848
author Austin Zadoks
Antimo Marrazzo
Nicola Marzari
author_facet Austin Zadoks
Antimo Marrazzo
Nicola Marzari
author_sort Austin Zadoks
collection DOAJ
description Abstract Machine learning in atomistic materials science has grown to become a powerful tool, with most approaches focusing on atomic geometry, typically decomposed into local atomic environments. This approach, while well-suited for machine-learned interatomic potentials, is conceptually at odds with learning complex intrinsic properties of materials, often driven by spectral properties commonly represented in reciprocal space (e.g., band gaps or mobilities) which cannot be readily partitioned in real space. For such applications, methods that represent the electronic rather than the atomic structure could be more promising. In this work, we present a general framework focused on electronic-structure descriptors that take advantage of the natural symmetries and inherent interpretability of physical models. We apply this framework first to material similarity and then to accelerated screening, where a model trained on 217 materials correctly labels 75% of entries in the Materials Cloud 3D database, which meet common screening criteria for promising transparent-conducting materials.
format Article
id doaj-art-db01f22b7cea4d7bb0430892b6fd7b80
institution Kabale University
issn 2057-3960
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series npj Computational Materials
spelling doaj-art-db01f22b7cea4d7bb0430892b6fd7b802024-12-08T12:37:36ZengNature Portfolionpj Computational Materials2057-39602024-12-0110111210.1038/s41524-024-01446-9Spectral operator representationsAustin Zadoks0Antimo Marrazzo1Nicola Marzari2Theory and Simulation of Materials (THEOS), École Polytechnique Fédérale de LausanneDipartimento di Fisica, Università di TriesteTheory and Simulation of Materials (THEOS), École Polytechnique Fédérale de LausanneAbstract Machine learning in atomistic materials science has grown to become a powerful tool, with most approaches focusing on atomic geometry, typically decomposed into local atomic environments. This approach, while well-suited for machine-learned interatomic potentials, is conceptually at odds with learning complex intrinsic properties of materials, often driven by spectral properties commonly represented in reciprocal space (e.g., band gaps or mobilities) which cannot be readily partitioned in real space. For such applications, methods that represent the electronic rather than the atomic structure could be more promising. In this work, we present a general framework focused on electronic-structure descriptors that take advantage of the natural symmetries and inherent interpretability of physical models. We apply this framework first to material similarity and then to accelerated screening, where a model trained on 217 materials correctly labels 75% of entries in the Materials Cloud 3D database, which meet common screening criteria for promising transparent-conducting materials.https://doi.org/10.1038/s41524-024-01446-9
spellingShingle Austin Zadoks
Antimo Marrazzo
Nicola Marzari
Spectral operator representations
npj Computational Materials
title Spectral operator representations
title_full Spectral operator representations
title_fullStr Spectral operator representations
title_full_unstemmed Spectral operator representations
title_short Spectral operator representations
title_sort spectral operator representations
url https://doi.org/10.1038/s41524-024-01446-9
work_keys_str_mv AT austinzadoks spectraloperatorrepresentations
AT antimomarrazzo spectraloperatorrepresentations
AT nicolamarzari spectraloperatorrepresentations