MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zamaraeva, Elena, Collins, Christopher M., Darling, George R., Dyer, Matthew S., Peng, Bei, Savani, Rahul, Antypov, Dmytro, Gusev, Vladimir V., Clymo, Judith, Spirakis, Paul G., Rosseinsky, Matthew J.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916779035459584
author Zamaraeva, Elena
Collins, Christopher M.
Darling, George R.
Dyer, Matthew S.
Peng, Bei
Savani, Rahul
Antypov, Dmytro
Gusev, Vladimir V.
Clymo, Judith
Spirakis, Paul G.
Rosseinsky, Matthew J.
author_facet Zamaraeva, Elena
Collins, Christopher M.
Darling, George R.
Dyer, Matthew S.
Peng, Bei
Savani, Rahul
Antypov, Dmytro
Gusev, Vladimir V.
Clymo, Judith
Spirakis, Paul G.
Rosseinsky, Matthew J.
contents Geometry optimization of atomic structures is a common and crucial task in computational chemistry and materials design. Following the learning to optimize paradigm, we propose a new multi-agent reinforcement learning method called Multi-Agent Crystal Structure optimization (MACS) to address periodic crystal structure optimization. MACS treats geometry optimization as a partially observable Markov game in which atoms are agents that adjust their positions to collectively discover a stable configuration. We train MACS across various compositions of reported crystalline materials to obtain a policy that successfully optimizes structures from the training compositions as well as structures of larger sizes and unseen compositions, confirming its excellent scalability and zero-shot transferability. We benchmark our approach against a broad range of state-of-the-art optimization methods and demonstrate that MACS optimizes periodic crystal structures significantly faster, with fewer energy calculations, and the lowest failure rate.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04195
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures
Zamaraeva, Elena
Collins, Christopher M.
Darling, George R.
Dyer, Matthew S.
Peng, Bei
Savani, Rahul
Antypov, Dmytro
Gusev, Vladimir V.
Clymo, Judith
Spirakis, Paul G.
Rosseinsky, Matthew J.
Machine Learning
Artificial Intelligence
68T05
I.2.6; I.2.11
Geometry optimization of atomic structures is a common and crucial task in computational chemistry and materials design. Following the learning to optimize paradigm, we propose a new multi-agent reinforcement learning method called Multi-Agent Crystal Structure optimization (MACS) to address periodic crystal structure optimization. MACS treats geometry optimization as a partially observable Markov game in which atoms are agents that adjust their positions to collectively discover a stable configuration. We train MACS across various compositions of reported crystalline materials to obtain a policy that successfully optimizes structures from the training compositions as well as structures of larger sizes and unseen compositions, confirming its excellent scalability and zero-shot transferability. We benchmark our approach against a broad range of state-of-the-art optimization methods and demonstrate that MACS optimizes periodic crystal structures significantly faster, with fewer energy calculations, and the lowest failure rate.
title MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures
topic Machine Learning
Artificial Intelligence
68T05
I.2.6; I.2.11
url https://arxiv.org/abs/2506.04195