Loading…
Nudged elastic band calculations accelerated with Gaussian process regression
Minimum energy paths for transitions such as atomic and/or spin rearrangements in thermalized systems are the transition paths of largest statistical weight. Such paths are frequently calculated using the nudged elastic band method, where an initial path is iteratively shifted to the nearest minimum...
Saved in:
Published in: | The Journal of chemical physics 2017-10, Vol.147 (15), p.152720-152720 |
---|---|
Main Authors: | , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63 |
---|---|
cites | cdi_FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63 |
container_end_page | 152720 |
container_issue | 15 |
container_start_page | 152720 |
container_title | The Journal of chemical physics |
container_volume | 147 |
creator | Koistinen, Olli-Pekka Dagbjartsdóttir, Freyja B. Ásgeirsson, Vilhjálmur Vehtari, Aki Jónsson, Hannes |
description | Minimum energy paths for transitions such as atomic and/or spin rearrangements in thermalized systems are the transition paths of largest statistical weight. Such paths are frequently calculated using the nudged elastic band method, where an initial path is iteratively shifted to the nearest minimum energy path. The computational effort can be large, especially when ab initio or electron density functional calculations are used to evaluate the energy and atomic forces. Here, we show how the number of such evaluations can be reduced by an order of magnitude using a Gaussian process regression approach where an approximate energy surface is generated and refined in each iteration. When the goal is to evaluate the transition rate within harmonic transition state theory, the evaluation of the Hessian matrix at the initial and final state minima can be carried out beforehand and used as input in the minimum energy path calculation, thereby improving stability and reducing the number of iterations needed for convergence. A Gaussian process model also provides an uncertainty estimate for the approximate energy surface, and this can be used to focus the calculations on the lesser-known part of the path, thereby reducing the number of needed energy and force evaluations to a half in the present calculations. The methodology is illustrated using the two-dimensional Müller-Brown potential surface and performance assessed on an established benchmark involving 13 rearrangement transitions of a heptamer island on a solid surface. |
doi_str_mv | 10.1063/1.4986787 |
format | article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2116062218</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1954416877</sourcerecordid><originalsourceid>FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63</originalsourceid><addsrcrecordid>eNp90MFLwzAYBfAgipvTg_-AFLyo0PkladLkKEOnMPWi55Al32ZH12rSIv73Rjc9ePD0Lj8ej0fIMYUxBckv6bjQSpaq3CFDCkrnpdSwS4YAjOZaghyQgxhXAEBLVuyTAdMgBAcxJPcPvV-iz7C2satcNreNz5ytXV_brmqbmFnnsMZgu6Teq-4lm9o-xso22WtoHcaYBVyGlEkfkr2FrSMebXNEnm-unya3-exxeje5muWOF6rLi4XHgiuh0TuLglnkmnErNGdSKEmdokCl8pYJ5RAkckAppFtwCXPpJR-Rs01vmvDWY-zMuoppZm0bbPtoqBZFkRrKMtHTP3TV9qFJ6wyjNH3DGFVJnW-UC22MARfmNVRrGz4MBfP1saFm-3GyJ9vGfr5G_yt_Tk3gYgOiq7rvF_9p-wSgY4Kj</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2116062218</pqid></control><display><type>article</type><title>Nudged elastic band calculations accelerated with Gaussian process regression</title><source>American Institute of Physics:Jisc Collections:Transitional Journals Agreement 2021-23 (Reading list)</source><source>American Institute of Physics</source><creator>Koistinen, Olli-Pekka ; Dagbjartsdóttir, Freyja B. ; Ásgeirsson, Vilhjálmur ; Vehtari, Aki ; Jónsson, Hannes</creator><creatorcontrib>Koistinen, Olli-Pekka ; Dagbjartsdóttir, Freyja B. ; Ásgeirsson, Vilhjálmur ; Vehtari, Aki ; Jónsson, Hannes</creatorcontrib><description>Minimum energy paths for transitions such as atomic and/or spin rearrangements in thermalized systems are the transition paths of largest statistical weight. Such paths are frequently calculated using the nudged elastic band method, where an initial path is iteratively shifted to the nearest minimum energy path. The computational effort can be large, especially when ab initio or electron density functional calculations are used to evaluate the energy and atomic forces. Here, we show how the number of such evaluations can be reduced by an order of magnitude using a Gaussian process regression approach where an approximate energy surface is generated and refined in each iteration. When the goal is to evaluate the transition rate within harmonic transition state theory, the evaluation of the Hessian matrix at the initial and final state minima can be carried out beforehand and used as input in the minimum energy path calculation, thereby improving stability and reducing the number of iterations needed for convergence. A Gaussian process model also provides an uncertainty estimate for the approximate energy surface, and this can be used to focus the calculations on the lesser-known part of the path, thereby reducing the number of needed energy and force evaluations to a half in the present calculations. The methodology is illustrated using the two-dimensional Müller-Brown potential surface and performance assessed on an established benchmark involving 13 rearrangement transitions of a heptamer island on a solid surface.</description><identifier>ISSN: 0021-9606</identifier><identifier>EISSN: 1089-7690</identifier><identifier>DOI: 10.1063/1.4986787</identifier><identifier>PMID: 29055305</identifier><identifier>CODEN: JCPSA6</identifier><language>eng</language><publisher>United States: American Institute of Physics</publisher><subject>Atomic properties ; Electron density ; Energy ; Gaussian process ; Hessian matrices ; Iterative methods ; Physics ; Solid surfaces ; Statistical analysis ; Weight</subject><ispartof>The Journal of chemical physics, 2017-10, Vol.147 (15), p.152720-152720</ispartof><rights>Author(s)</rights><rights>2017 Author(s). Published by AIP Publishing.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63</citedby><cites>FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63</cites><orcidid>0000-0002-0810-7369 ; 0000-0003-2164-9469 ; 0000-0001-8285-5421 ; 0000000182855421 ; 0000000208107369 ; 0000000321649469</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://pubs.aip.org/jcp/article-lookup/doi/10.1063/1.4986787$$EHTML$$P50$$Gscitation$$H</linktohtml><link.rule.ids>314,780,782,784,795,27924,27925,76255</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/29055305$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Koistinen, Olli-Pekka</creatorcontrib><creatorcontrib>Dagbjartsdóttir, Freyja B.</creatorcontrib><creatorcontrib>Ásgeirsson, Vilhjálmur</creatorcontrib><creatorcontrib>Vehtari, Aki</creatorcontrib><creatorcontrib>Jónsson, Hannes</creatorcontrib><title>Nudged elastic band calculations accelerated with Gaussian process regression</title><title>The Journal of chemical physics</title><addtitle>J Chem Phys</addtitle><description>Minimum energy paths for transitions such as atomic and/or spin rearrangements in thermalized systems are the transition paths of largest statistical weight. Such paths are frequently calculated using the nudged elastic band method, where an initial path is iteratively shifted to the nearest minimum energy path. The computational effort can be large, especially when ab initio or electron density functional calculations are used to evaluate the energy and atomic forces. Here, we show how the number of such evaluations can be reduced by an order of magnitude using a Gaussian process regression approach where an approximate energy surface is generated and refined in each iteration. When the goal is to evaluate the transition rate within harmonic transition state theory, the evaluation of the Hessian matrix at the initial and final state minima can be carried out beforehand and used as input in the minimum energy path calculation, thereby improving stability and reducing the number of iterations needed for convergence. A Gaussian process model also provides an uncertainty estimate for the approximate energy surface, and this can be used to focus the calculations on the lesser-known part of the path, thereby reducing the number of needed energy and force evaluations to a half in the present calculations. The methodology is illustrated using the two-dimensional Müller-Brown potential surface and performance assessed on an established benchmark involving 13 rearrangement transitions of a heptamer island on a solid surface.</description><subject>Atomic properties</subject><subject>Electron density</subject><subject>Energy</subject><subject>Gaussian process</subject><subject>Hessian matrices</subject><subject>Iterative methods</subject><subject>Physics</subject><subject>Solid surfaces</subject><subject>Statistical analysis</subject><subject>Weight</subject><issn>0021-9606</issn><issn>1089-7690</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><recordid>eNp90MFLwzAYBfAgipvTg_-AFLyo0PkladLkKEOnMPWi55Al32ZH12rSIv73Rjc9ePD0Lj8ej0fIMYUxBckv6bjQSpaq3CFDCkrnpdSwS4YAjOZaghyQgxhXAEBLVuyTAdMgBAcxJPcPvV-iz7C2satcNreNz5ytXV_brmqbmFnnsMZgu6Teq-4lm9o-xso22WtoHcaYBVyGlEkfkr2FrSMebXNEnm-unya3-exxeje5muWOF6rLi4XHgiuh0TuLglnkmnErNGdSKEmdokCl8pYJ5RAkckAppFtwCXPpJR-Rs01vmvDWY-zMuoppZm0bbPtoqBZFkRrKMtHTP3TV9qFJ6wyjNH3DGFVJnW-UC22MARfmNVRrGz4MBfP1saFm-3GyJ9vGfr5G_yt_Tk3gYgOiq7rvF_9p-wSgY4Kj</recordid><startdate>20171021</startdate><enddate>20171021</enddate><creator>Koistinen, Olli-Pekka</creator><creator>Dagbjartsdóttir, Freyja B.</creator><creator>Ásgeirsson, Vilhjálmur</creator><creator>Vehtari, Aki</creator><creator>Jónsson, Hannes</creator><general>American Institute of Physics</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>8FD</scope><scope>H8D</scope><scope>L7M</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-0810-7369</orcidid><orcidid>https://orcid.org/0000-0003-2164-9469</orcidid><orcidid>https://orcid.org/0000-0001-8285-5421</orcidid><orcidid>https://orcid.org/0000000182855421</orcidid><orcidid>https://orcid.org/0000000208107369</orcidid><orcidid>https://orcid.org/0000000321649469</orcidid></search><sort><creationdate>20171021</creationdate><title>Nudged elastic band calculations accelerated with Gaussian process regression</title><author>Koistinen, Olli-Pekka ; Dagbjartsdóttir, Freyja B. ; Ásgeirsson, Vilhjálmur ; Vehtari, Aki ; Jónsson, Hannes</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Atomic properties</topic><topic>Electron density</topic><topic>Energy</topic><topic>Gaussian process</topic><topic>Hessian matrices</topic><topic>Iterative methods</topic><topic>Physics</topic><topic>Solid surfaces</topic><topic>Statistical analysis</topic><topic>Weight</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Koistinen, Olli-Pekka</creatorcontrib><creatorcontrib>Dagbjartsdóttir, Freyja B.</creatorcontrib><creatorcontrib>Ásgeirsson, Vilhjálmur</creatorcontrib><creatorcontrib>Vehtari, Aki</creatorcontrib><creatorcontrib>Jónsson, Hannes</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>Technology Research Database</collection><collection>Aerospace Database</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>MEDLINE - Academic</collection><jtitle>The Journal of chemical physics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Koistinen, Olli-Pekka</au><au>Dagbjartsdóttir, Freyja B.</au><au>Ásgeirsson, Vilhjálmur</au><au>Vehtari, Aki</au><au>Jónsson, Hannes</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Nudged elastic band calculations accelerated with Gaussian process regression</atitle><jtitle>The Journal of chemical physics</jtitle><addtitle>J Chem Phys</addtitle><date>2017-10-21</date><risdate>2017</risdate><volume>147</volume><issue>15</issue><spage>152720</spage><epage>152720</epage><pages>152720-152720</pages><issn>0021-9606</issn><eissn>1089-7690</eissn><coden>JCPSA6</coden><abstract>Minimum energy paths for transitions such as atomic and/or spin rearrangements in thermalized systems are the transition paths of largest statistical weight. Such paths are frequently calculated using the nudged elastic band method, where an initial path is iteratively shifted to the nearest minimum energy path. The computational effort can be large, especially when ab initio or electron density functional calculations are used to evaluate the energy and atomic forces. Here, we show how the number of such evaluations can be reduced by an order of magnitude using a Gaussian process regression approach where an approximate energy surface is generated and refined in each iteration. When the goal is to evaluate the transition rate within harmonic transition state theory, the evaluation of the Hessian matrix at the initial and final state minima can be carried out beforehand and used as input in the minimum energy path calculation, thereby improving stability and reducing the number of iterations needed for convergence. A Gaussian process model also provides an uncertainty estimate for the approximate energy surface, and this can be used to focus the calculations on the lesser-known part of the path, thereby reducing the number of needed energy and force evaluations to a half in the present calculations. The methodology is illustrated using the two-dimensional Müller-Brown potential surface and performance assessed on an established benchmark involving 13 rearrangement transitions of a heptamer island on a solid surface.</abstract><cop>United States</cop><pub>American Institute of Physics</pub><pmid>29055305</pmid><doi>10.1063/1.4986787</doi><tpages>14</tpages><orcidid>https://orcid.org/0000-0002-0810-7369</orcidid><orcidid>https://orcid.org/0000-0003-2164-9469</orcidid><orcidid>https://orcid.org/0000-0001-8285-5421</orcidid><orcidid>https://orcid.org/0000000182855421</orcidid><orcidid>https://orcid.org/0000000208107369</orcidid><orcidid>https://orcid.org/0000000321649469</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0021-9606 |
ispartof | The Journal of chemical physics, 2017-10, Vol.147 (15), p.152720-152720 |
issn | 0021-9606 1089-7690 |
language | eng |
recordid | cdi_proquest_journals_2116062218 |
source | American Institute of Physics:Jisc Collections:Transitional Journals Agreement 2021-23 (Reading list); American Institute of Physics |
subjects | Atomic properties Electron density Energy Gaussian process Hessian matrices Iterative methods Physics Solid surfaces Statistical analysis Weight |
title | Nudged elastic band calculations accelerated with Gaussian process regression |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T23%3A17%3A33IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Nudged%20elastic%20band%20calculations%20accelerated%20with%20Gaussian%20process%20regression&rft.jtitle=The%20Journal%20of%20chemical%20physics&rft.au=Koistinen,%20Olli-Pekka&rft.date=2017-10-21&rft.volume=147&rft.issue=15&rft.spage=152720&rft.epage=152720&rft.pages=152720-152720&rft.issn=0021-9606&rft.eissn=1089-7690&rft.coden=JCPSA6&rft_id=info:doi/10.1063/1.4986787&rft_dat=%3Cproquest_cross%3E1954416877%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c348t-4fde43859edcae52ae3923a593265861c810168da258ce06e30e656cf360b6d63%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2116062218&rft_id=info:pmid/29055305&rfr_iscdi=true |