Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains

Abstract Bayesian optimization (BO) has been leveraged for guiding autonomous and high-throughput experiments in materials science. However, few have evaluated the efficiency of BO across a broad range of experimental materials domains. In this work, we quantify the performance of BO with a collecti...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Qiaohao Liang, Aldair E. Gongora, Zekun Ren, Armi Tiihonen, Zhe Liu, Shijing Sun, James R. Deneault, Daniil Bash, Flore Mekki-Berrada, Saif A. Khan, Kedar Hippalgaonkar, Benji Maruyama, Keith A. Brown, John Fisher III, Tonio Buonassisi
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/bfc6d1e9f14b4b8bbc093f32c3360b8c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:bfc6d1e9f14b4b8bbc093f32c3360b8c
record_format	dspace
spelling	oai:doaj.org-article:bfc6d1e9f14b4b8bbc093f32c3360b8c2021-11-21T12:13:27ZBenchmarking the performance of Bayesian optimization across multiple experimental materials science domains10.1038/s41524-021-00656-92057-3960https://doaj.org/article/bfc6d1e9f14b4b8bbc093f32c3360b8c2021-11-01T00:00:00Zhttps://doi.org/10.1038/s41524-021-00656-9https://doaj.org/toc/2057-3960Abstract Bayesian optimization (BO) has been leveraged for guiding autonomous and high-throughput experiments in materials science. However, few have evaluated the efficiency of BO across a broad range of experimental materials domains. In this work, we quantify the performance of BO with a collection of surrogate model and acquisition function pairs across five diverse experimental materials systems. By defining acceleration and enhancement metrics for materials optimization objectives, we find that surrogate models such as Gaussian Process (GP) with anisotropic kernels and Random Forest (RF) have comparable performance in BO, and both outperform the commonly used GP with isotropic kernels. GP with anisotropic kernels has demonstrated the most robustness, yet RF is a close alternative and warrants more consideration because it is free from distribution assumptions, has smaller time complexity, and requires less effort in initial hyperparameter selection. We also raise awareness about the benefits of using GP with anisotropic kernels in future materials optimization campaigns.Qiaohao LiangAldair E. GongoraZekun RenArmi TiihonenZhe LiuShijing SunJames R. DeneaultDaniil BashFlore Mekki-BerradaSaif A. KhanKedar HippalgaonkarBenji MaruyamaKeith A. BrownJohn Fisher IIITonio BuonassisiNature PortfolioarticleMaterials of engineering and construction. Mechanics of materialsTA401-492Computer softwareQA76.75-76.765ENnpj Computational Materials, Vol 7, Iss 1, Pp 1-10 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765
spellingShingle	Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765 Qiaohao Liang Aldair E. Gongora Zekun Ren Armi Tiihonen Zhe Liu Shijing Sun James R. Deneault Daniil Bash Flore Mekki-Berrada Saif A. Khan Kedar Hippalgaonkar Benji Maruyama Keith A. Brown John Fisher III Tonio Buonassisi Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
description	Abstract Bayesian optimization (BO) has been leveraged for guiding autonomous and high-throughput experiments in materials science. However, few have evaluated the efficiency of BO across a broad range of experimental materials domains. In this work, we quantify the performance of BO with a collection of surrogate model and acquisition function pairs across five diverse experimental materials systems. By defining acceleration and enhancement metrics for materials optimization objectives, we find that surrogate models such as Gaussian Process (GP) with anisotropic kernels and Random Forest (RF) have comparable performance in BO, and both outperform the commonly used GP with isotropic kernels. GP with anisotropic kernels has demonstrated the most robustness, yet RF is a close alternative and warrants more consideration because it is free from distribution assumptions, has smaller time complexity, and requires less effort in initial hyperparameter selection. We also raise awareness about the benefits of using GP with anisotropic kernels in future materials optimization campaigns.
format	article
author	Qiaohao Liang Aldair E. Gongora Zekun Ren Armi Tiihonen Zhe Liu Shijing Sun James R. Deneault Daniil Bash Flore Mekki-Berrada Saif A. Khan Kedar Hippalgaonkar Benji Maruyama Keith A. Brown John Fisher III Tonio Buonassisi
author_facet	Qiaohao Liang Aldair E. Gongora Zekun Ren Armi Tiihonen Zhe Liu Shijing Sun James R. Deneault Daniil Bash Flore Mekki-Berrada Saif A. Khan Kedar Hippalgaonkar Benji Maruyama Keith A. Brown John Fisher III Tonio Buonassisi
author_sort	Qiaohao Liang
title	Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
title_short	Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
title_full	Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
title_fullStr	Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
title_full_unstemmed	Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains
title_sort	benchmarking the performance of bayesian optimization across multiple experimental materials science domains
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/bfc6d1e9f14b4b8bbc093f32c3360b8c
work_keys_str_mv	AT qiaohaoliang benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT aldairegongora benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT zekunren benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT armitiihonen benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT zheliu benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT shijingsun benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT jamesrdeneault benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT daniilbash benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT floremekkiberrada benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT saifakhan benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT kedarhippalgaonkar benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT benjimaruyama benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT keithabrown benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT johnfisheriii benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains AT toniobuonassisi benchmarkingtheperformanceofbayesianoptimizationacrossmultipleexperimentalmaterialssciencedomains
_version_	1718419148064686080

Benchmarking the performance of Bayesian optimization across multiple experimental materials science domains

Ejemplares similares