Cost-Effective Strategy of Enhancing Machine Learning Potentials by Transfer Learning from a Multicomponent Data Set on ænet-PyTorch

IF 3.3 3区 化学 Q2 CHEMISTRY, PHYSICAL The Journal of Physical Chemistry C Pub Date : 2024-12-27 DOI:10.1021/acs.jpcc.4c06235
An Niza El Aisnada, Kajjana Boonpalit, Robin van der Kruit, Koen M. Draijer, Jon López-Zorrilla, Masahiro Miyauchi, Akira Yamaguchi, Nongnuch Artrith
{"title":"Cost-Effective Strategy of Enhancing Machine Learning Potentials by Transfer Learning from a Multicomponent Data Set on ænet-PyTorch","authors":"An Niza El Aisnada, Kajjana Boonpalit, Robin van der Kruit, Koen M. Draijer, Jon López-Zorrilla, Masahiro Miyauchi, Akira Yamaguchi, Nongnuch Artrith","doi":"10.1021/acs.jpcc.4c06235","DOIUrl":null,"url":null,"abstract":"Machine learning potentials (MLPs) offer efficient and accurate material simulations, but constructing the reference ab initio database remains a significant challenge, particularly for catalyst-adsorbate systems. Training an MLP with a small data set can lead to overfitting, thus limiting its practical applications. This study explores the feasibility of developing computationally cost-effective and accurate MLPs for catalyst-adsorbate systems with a limited number of ab initio references by leveraging a transfer learning strategy from subsets of a comprehensive public database. Using the Open Catalyst Project 2020 (OC20)─a data set closely related to our system of interest─we pretrained MLP models on OC20 subsets using the ænet-PyTorch framework. We compared several strategies for database subset selection. Our findings indicate that MLPs constructed via transfer learning exhibit better generalizability than those constructed from scratch, as demonstrated by the consistency in the dynamics simulations. Remarkably, transfer learning enhances the stability and accuracy of MLPs for the CuAu/H<sub>2</sub>O system with approximately 600 reference data points. This approach achieved excellent extrapolation performance in molecular dynamics simulations for the larger CuAu/6H<sub>2</sub>O system, maintaining stable and accurate predictions for up to 250 ps, whereas MLPs without transfer learning become unstable before reaching 50 ps. We also examine the potential limitations of this strategy. This work proposes an alternative, cost-effective approach for constructing MLPs for the challenging simulation of catalytic systems. Finally, we anticipate that this methodology will pave the way for broader applications in materials science and catalysis research, facilitating more efficient and accurate simulations across various systems.","PeriodicalId":61,"journal":{"name":"The Journal of Physical Chemistry C","volume":"224 1","pages":""},"PeriodicalIF":3.3000,"publicationDate":"2024-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"The Journal of Physical Chemistry C","FirstCategoryId":"1","ListUrlMain":"https://doi.org/10.1021/acs.jpcc.4c06235","RegionNum":3,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
引用次数: 0

Abstract

Machine learning potentials (MLPs) offer efficient and accurate material simulations, but constructing the reference ab initio database remains a significant challenge, particularly for catalyst-adsorbate systems. Training an MLP with a small data set can lead to overfitting, thus limiting its practical applications. This study explores the feasibility of developing computationally cost-effective and accurate MLPs for catalyst-adsorbate systems with a limited number of ab initio references by leveraging a transfer learning strategy from subsets of a comprehensive public database. Using the Open Catalyst Project 2020 (OC20)─a data set closely related to our system of interest─we pretrained MLP models on OC20 subsets using the ænet-PyTorch framework. We compared several strategies for database subset selection. Our findings indicate that MLPs constructed via transfer learning exhibit better generalizability than those constructed from scratch, as demonstrated by the consistency in the dynamics simulations. Remarkably, transfer learning enhances the stability and accuracy of MLPs for the CuAu/H2O system with approximately 600 reference data points. This approach achieved excellent extrapolation performance in molecular dynamics simulations for the larger CuAu/6H2O system, maintaining stable and accurate predictions for up to 250 ps, whereas MLPs without transfer learning become unstable before reaching 50 ps. We also examine the potential limitations of this strategy. This work proposes an alternative, cost-effective approach for constructing MLPs for the challenging simulation of catalytic systems. Finally, we anticipate that this methodology will pave the way for broader applications in materials science and catalysis research, facilitating more efficient and accurate simulations across various systems.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
求助全文
约1分钟内获得全文 去求助
来源期刊
The Journal of Physical Chemistry C
The Journal of Physical Chemistry C 化学-材料科学:综合
CiteScore
6.50
自引率
8.10%
发文量
2047
审稿时长
1.8 months
期刊介绍: The Journal of Physical Chemistry A/B/C is devoted to reporting new and original experimental and theoretical basic research of interest to physical chemists, biophysical chemists, and chemical physicists.
期刊最新文献
Influence of the Annealing Temperature on the Structural and Electrical Properties of SiOx(Si)&FeyOz(Fe) Films CatFlow: An Automated Workflow for Training Machine Learning Potentials to Compute Free Energies in Dynamic Catalysis Characterization of an Unexpected μ3 Adsorption of Molecular Oxygen on Ag(100) with Low-Temperature STM Interplay of Product Selectivity and Isolated Ru-Oxo Sites in the Solvent-Free Oxidization of Benzyl Alcohol Designing a Third Component for Ternary Organic Solar Cells by Tailoring the Acceptor Structure
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1