检索结果-南通市图书馆

Extra.ting structura. motifs from pa.r distribution function da.a.of na.ostructures using expla.na.le ma.hine lea.ning

引用

npj Computa.iona. Ma.eria.s 2022年第1期8卷 2053-2063页

作者： a.dy S.a.ker Emil T.S.Kjær Mikkel Juelsholt Troels Linda.l Christia.sen Susa.ne Linn Skjærvø Ma.s Ry Vogel Jørgensen Innokenty Ka.tor Da.iel Risskov Sørensen Simon J.L.billinge.Ra.ha.endra.Selva. Kirsten M.Ø.Jensen Depa.tment of Chemistry a.d Na.o-Science Center University of Copenhagen2100CopenhagenDenmark Depa.tment of Ma.eria.s University of OxfordParks RoadOxfordUK Depa.tment of Chemistry&iNa.O Aarhus University8000AarhusDenmark Ma. IV La.ora.ory Lund University22484LundSweden Depa.tment of Physics Technical University of Denmark2880LyngbyDenmark Depa.tment of a.plied Physics a.d a.plied Ma.hema.ics Columbia UniversityNew YorkNY10027USA Condensed Ma.ter Physics a.d Ma.eria.s Science Depa.tment Brookhaven National LaboratoryUptonNY11973USA Depa.tment of Computer Science University of Copenhagen2100CopenhagenDenmark Depa.tment of Neuroscience University of Copenhagen2200CopenhagenDenmark

Cha.a.teriza.ion of ma.eria. structure with X-ra. or neutron sca.tering using *** Distribution Function(PDF)a.a.ysis most often rely on refining a.structure model a.a.nst a. experimenta. ***,identifying a.suita.le model is often a.***,a.toma.ed a.proa.hes ha.e ma.e it possible to test thousa.ds of models for ea.h da.a.et,but these methods a.e computa.iona.ly expensive a.d a.a.ysing the output,*** structura. informa.ion from the resulting fits in a.mea.ingful wa.,is *** Ma.hine Lea.ning ba.ed Motif Extra.tor(ML-MotEx)tra.ns a. ML a.gorithm on thousa.ds of fits,a.d uses SHa.(SHa.ley a.ditive exPla.a.ion)va.ues to identify which model fea.ures a.e importa.t for the fit *** use the method for 4 different chemica. systems,including disordered na.oma.eria.s a.d ***-MotEx opens for a.type of modelling where ea.h fea.ure in a.model is a.signed a. importa.ce va.ue for the fit qua.ity ba.ed on expla.na.le ML.

关键词： expla.n structura. mea.ingful

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Recent a.va.ces a.d a.plica.ions of deep lea.ning methods in ma.eria.s science

引用

npj Computa.iona. Ma.eria.s 2022年第1期8卷 548-573页

作者： Ka.a. Choudha.y Bria. DeCost Chi Chen a.ubha. Ja.n Fra.cesca.Ta.a.za.Rya. Cohn Cheol Woo Pa.k a.ok Choudha.y a.kit a.ra.a. Simon J.L.billinge.Eliza.eth Holm Shyue Ping Ong Chris Wolverton Ma.eria.s Science a.d Engineering Division National Institute of Standards and TechnologyGaithersburgMD20899USA Theiss Resea.ch La JollaCA92037USA DeepMa.eria.s LLC Silver SpringMD20906USA Ma.eria. Mea.urement Science Division National Institute of Standards and TechnologyGaithersburgMD20899USA Depa.tment of Na.oEngineering University of California San DiegoSan DiegoCA92093USA Energy Technologies a.ea Lawrence Berkeley National LaboratoryBerkeleyCAUSA Depa.tment of Ma.eria.s Science a.d Engineering Carnegie Mellon UniversityPittsburghPA15213USA Depa.tment of Ma.eria.s Science a.d Engineering Northwestern UniversityEvanstonIL60208USA Depa.tment of Electrica. a.d Computer Engineering Northwestern UniversityEvanstonIL60208USA Depa.tment of a.plied Physics a.d a.plied Ma.hema.ics a.d the Da.a.Science Institute Fu Foundation School of Engineering and Applied SciencesColumbia UniversityNew YorkNY10027USA

Deep lea.ning(DL)is one of the fa.test-growing topics in ma.eria.s da.a.science,with ra.idly emerging a.plica.ions spa.ning a.omistic,ima.e-ba.ed,spectra.,a.d textua. da.a.*** a.lows a.a.ysis of unstructured da.a.a.d a.toma.ed identifica.ion of *** recent development of la.ge ma.eria.s da.a.a.es ha. fueled the a.plica.ion of DL methods in a.omistic prediction in *** contra.t,a.va.ces in ima.e a.d spectra. da.a.ha.e la.gely levera.ed synthetic da.a.ena.led by high-qua.ity forwa.d models a. well a. by genera.ive unsupervised DL *** this a.ticle,we present a.high-level overview of deep lea.ning methods followed by a.deta.led discussion of recent developments of deep lea.ning in a.omistic simula.ion,ma.eria.s ima.ing,spectra. a.a.ysis,a.d na.ura. la.gua.e *** ea.h moda.ity we discuss a.plica.ions involving both theoretica. a.d experimenta. da.a.typica. modeling a.proa.hes with their strengths a.d limita.ions,a.d releva.t publicly a.a.la.le softwa.e a.d *** conclude the review with a.discussion of recent cross-cutting work rela.ed to uncerta.nty qua.tifica.ion in this field a.d a.brief perspective on limita.ions,cha.lenges,a.d potentia. growth a.ea. for DL methods in ma.eria.s science.

关键词： lea.ning limita.ions textua.

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：