Orb: A Fast, Scalable Neural Network Potential

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Neumann, Mark, Gin, James, Rhodes, Benjamin, Bennett, Steven, Li, Zhiyi, Choubisa, Hitarth, Hussey, Arthur, Godwin, Jonathan
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913567846957056
author Neumann, Mark
Gin, James
Rhodes, Benjamin
Bennett, Steven
Li, Zhiyi
Choubisa, Hitarth
Hussey, Arthur
Godwin, Jonathan
author_facet Neumann, Mark
Gin, James
Rhodes, Benjamin
Bennett, Steven
Li, Zhiyi
Choubisa, Hitarth
Hussey, Arthur
Godwin, Jonathan
contents We introduce Orb, a family of universal interatomic potentials for atomistic modelling of materials. Orb models are 3-6 times faster than existing universal potentials, stable under simulation for a range of out of distribution materials and, upon release, represented a 31% reduction in error over other methods on the Matbench Discovery benchmark. We explore several aspects of foundation model development for materials, with a focus on diffusion pretraining. We evaluate Orb as a model for geometry optimization, Monte Carlo and molecular dynamics simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2410_22570
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Orb: A Fast, Scalable Neural Network Potential
Neumann, Mark
Gin, James
Rhodes, Benjamin
Bennett, Steven
Li, Zhiyi
Choubisa, Hitarth
Hussey, Arthur
Godwin, Jonathan
Materials Science
Machine Learning
We introduce Orb, a family of universal interatomic potentials for atomistic modelling of materials. Orb models are 3-6 times faster than existing universal potentials, stable under simulation for a range of out of distribution materials and, upon release, represented a 31% reduction in error over other methods on the Matbench Discovery benchmark. We explore several aspects of foundation model development for materials, with a focus on diffusion pretraining. We evaluate Orb as a model for geometry optimization, Monte Carlo and molecular dynamics simulations.
title Orb: A Fast, Scalable Neural Network Potential
topic Materials Science
Machine Learning
url https://arxiv.org/abs/2410.22570