MLatom 2: An Integrative Platform for Atomistic Machine Learning

IF 7.1 2区 化学 Q1 CHEMISTRY, MULTIDISCIPLINARY Topics in Current Chemistry Pub Date : 2021-06-08 DOI:10.1007/s41061-021-00339-5
Pavlo O. Dral, Fuchun Ge, Bao-Xin Xue, Yi-Fan Hou, Max Pinheiro Jr, Jianxing Huang, Mario Barbatti
{"title":"MLatom 2: An Integrative Platform for Atomistic Machine Learning","authors":"Pavlo O. Dral,&nbsp;Fuchun Ge,&nbsp;Bao-Xin Xue,&nbsp;Yi-Fan Hou,&nbsp;Max Pinheiro Jr,&nbsp;Jianxing Huang,&nbsp;Mario Barbatti","doi":"10.1007/s41061-021-00339-5","DOIUrl":null,"url":null,"abstract":"<p>Atomistic machine learning (AML) simulations are used in chemistry at an ever-increasing pace. A large number of AML models has been developed, but their implementations are scattered among different packages, each with its own conventions for input and output. Thus, here we give an overview of our MLatom 2 software package, which provides an integrative platform for a wide variety of AML simulations by implementing from scratch and interfacing existing software for a range of state-of-the-art models. These include kernel method-based model types such as KREG (native implementation), sGDML, and GAP-SOAP as well as neural-network-based model types such as ANI, DeepPot-SE, and PhysNet. The theoretical foundations behind these methods are overviewed too. The modular structure of MLatom allows for easy extension to more AML model types. MLatom 2 also has many other capabilities useful for AML simulations, such as the support of custom descriptors, farthest-point and structure-based sampling, hyperparameter optimization, model evaluation, and automatic learning curve generation. It can also be used for such multi-step tasks as Δ-learning, self-correction approaches, and absorption spectrum simulation within the machine-learning nuclear-ensemble approach. Several of these MLatom 2 capabilities are showcased in application examples.</p>","PeriodicalId":54344,"journal":{"name":"Topics in Current Chemistry","volume":"379 4","pages":""},"PeriodicalIF":7.1000,"publicationDate":"2021-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1007/s41061-021-00339-5","citationCount":"29","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Topics in Current Chemistry","FirstCategoryId":"92","ListUrlMain":"https://link.springer.com/article/10.1007/s41061-021-00339-5","RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 29

Abstract

Atomistic machine learning (AML) simulations are used in chemistry at an ever-increasing pace. A large number of AML models has been developed, but their implementations are scattered among different packages, each with its own conventions for input and output. Thus, here we give an overview of our MLatom 2 software package, which provides an integrative platform for a wide variety of AML simulations by implementing from scratch and interfacing existing software for a range of state-of-the-art models. These include kernel method-based model types such as KREG (native implementation), sGDML, and GAP-SOAP as well as neural-network-based model types such as ANI, DeepPot-SE, and PhysNet. The theoretical foundations behind these methods are overviewed too. The modular structure of MLatom allows for easy extension to more AML model types. MLatom 2 also has many other capabilities useful for AML simulations, such as the support of custom descriptors, farthest-point and structure-based sampling, hyperparameter optimization, model evaluation, and automatic learning curve generation. It can also be used for such multi-step tasks as Δ-learning, self-correction approaches, and absorption spectrum simulation within the machine-learning nuclear-ensemble approach. Several of these MLatom 2 capabilities are showcased in application examples.

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
MLatom 2:原子机器学习的集成平台
原子机器学习(AML)模拟在化学中的应用速度越来越快。已经开发了大量的AML模型,但是它们的实现分散在不同的包中,每个包都有自己的输入和输出约定。因此,我们在这里概述了我们的MLatom 2软件包,该软件包通过从头开始实现并为一系列最先进的模型连接现有软件,为各种AML模拟提供了一个集成平台。这些包括基于内核方法的模型类型,如KREG(本机实现)、sGDML和GAP-SOAP,以及基于神经网络的模型类型,如ANI、DeepPot-SE和PhysNet。本文还概述了这些方法背后的理论基础。MLatom的模块化结构允许轻松扩展到更多的AML模型类型。MLatom 2还具有许多其他对AML模拟有用的功能,例如支持自定义描述符、最远点和基于结构的采样、超参数优化、模型评估和自动学习曲线生成。它还可以用于机器学习核系综方法中的Δ-learning、自我校正方法和吸收光谱模拟等多步骤任务。在应用程序示例中展示了其中几个MLatom 2功能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Topics in Current Chemistry
Topics in Current Chemistry Chemistry-General Chemistry
CiteScore
13.70
自引率
1.20%
发文量
48
期刊介绍: Topics in Current Chemistry is a journal that presents critical reviews of present and future trends in modern chemical research. It covers all areas of chemical science, including interactions with related disciplines like biology, medicine, physics, and materials science. The articles in this journal are organized into thematic collections, offering a comprehensive perspective on emerging research to non-specialist readers in academia or industry. Each review article focuses on one aspect of the topic and provides a critical survey, placing it in the context of the collection. Selected examples highlight significant developments from the past 5 to 10 years. Instead of providing an exhaustive summary or extensive data, the articles concentrate on methodological thinking. This approach allows non-specialist readers to understand the information fully and presents the potential prospects for future developments.
期刊最新文献
Structure-Based Drug Design of RdRp Inhibitors against SARS-CoV-2 The Intramolecular Povarov Tool in the Construction of Fused Nitrogen-Containing Heterocycles Photothermal Catalytic CO2 Conversion: Beyond Catalysis and Photocatalysis Multicomponent Reactions Using C,N-Binucleophilic Nature of Aminopyrazoles: Construction of Pyrazole-Fused Heterocycles Laser-Induced Transfer of Functional Materials
×
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