Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials

Data science and artificial intelligence are playing an increasingly important role in the physical sciences. Unfortunately, in the field of energetic materials data scarcity limits the accuracy and even applicability of ML tools. To address data limitations, we compiled multi‐modal data: both exper...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Propellants, explosives, pyrotechnics explosives, pyrotechnics, 2024-11
Hauptverfasser: Appleton, Robert J., Klinger, Daniel, Lee, Brian H., Taylor, Michael, Kim, Sohee, Blankenship, Samuel, Barnes, Brian C., Son, Steven F., Strachan, Alejandro
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title Propellants, explosives, pyrotechnics
container_volume
creator Appleton, Robert J.
Klinger, Daniel
Lee, Brian H.
Taylor, Michael
Kim, Sohee
Blankenship, Samuel
Barnes, Brian C.
Son, Steven F.
Strachan, Alejandro
description Data science and artificial intelligence are playing an increasingly important role in the physical sciences. Unfortunately, in the field of energetic materials data scarcity limits the accuracy and even applicability of ML tools. To address data limitations, we compiled multi‐modal data: both experimental and computational results for several properties. We find that multi‐task neural networks can learn from multi‐modal data and outperform single‐task models trained for specific properties. As expected, the improvement is more significant for data‐scarce properties. These models are trained using descriptors built from simple molecular information and can be readily applied for large‐scale materials screening to explore multiple properties simultaneously. This approach is widely applicable to fields outside energetic materials.
doi_str_mv 10.1002/prep.202400248
format Article
fullrecord <record><control><sourceid>crossref</sourceid><recordid>TN_cdi_crossref_primary_10_1002_prep_202400248</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>10_1002_prep_202400248</sourcerecordid><originalsourceid>FETCH-LOGICAL-c164t-1668f2decf98c11c5c87abc86312ea79f16b70256c993e9c774c21899fdf70fc3</originalsourceid><addsrcrecordid>eNo9kLtOwzAYhS0EEqGwMvsFEvw7iS8jqlpASgVDmSPX-V0ZQhLZZujGI_CMPAmpuEzfOcM5w0fINbACGOM3U8Cp4IxXc6nUCcmg5pBXTMlTkjE55xKgPicXMb4wNk8YZKTZvPfJf318bk18pX9l7TvsfTrQBk0Y_LCno6NPYZwwJI-RujHQ1YBhj8lbujEJgzd9vCRnbgZe_XJBnter7fI-bx7vHpa3TW5BVCkHIZTjHVqnlQWwtVXS7KwSJXA0UjsQO8l4LazWJWorZWU5KK1d5yRztlyQ4ufXhjHGgK6dgn8z4dACa48u2qOL9t9F-Q2E0FUH</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials</title><source>Access via Wiley Online Library</source><creator>Appleton, Robert J. ; Klinger, Daniel ; Lee, Brian H. ; Taylor, Michael ; Kim, Sohee ; Blankenship, Samuel ; Barnes, Brian C. ; Son, Steven F. ; Strachan, Alejandro</creator><creatorcontrib>Appleton, Robert J. ; Klinger, Daniel ; Lee, Brian H. ; Taylor, Michael ; Kim, Sohee ; Blankenship, Samuel ; Barnes, Brian C. ; Son, Steven F. ; Strachan, Alejandro</creatorcontrib><description>Data science and artificial intelligence are playing an increasingly important role in the physical sciences. Unfortunately, in the field of energetic materials data scarcity limits the accuracy and even applicability of ML tools. To address data limitations, we compiled multi‐modal data: both experimental and computational results for several properties. We find that multi‐task neural networks can learn from multi‐modal data and outperform single‐task models trained for specific properties. As expected, the improvement is more significant for data‐scarce properties. These models are trained using descriptors built from simple molecular information and can be readily applied for large‐scale materials screening to explore multiple properties simultaneously. This approach is widely applicable to fields outside energetic materials.</description><identifier>ISSN: 0721-3115</identifier><identifier>EISSN: 1521-4087</identifier><identifier>DOI: 10.1002/prep.202400248</identifier><language>eng</language><ispartof>Propellants, explosives, pyrotechnics, 2024-11</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c164t-1668f2decf98c11c5c87abc86312ea79f16b70256c993e9c774c21899fdf70fc3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Appleton, Robert J.</creatorcontrib><creatorcontrib>Klinger, Daniel</creatorcontrib><creatorcontrib>Lee, Brian H.</creatorcontrib><creatorcontrib>Taylor, Michael</creatorcontrib><creatorcontrib>Kim, Sohee</creatorcontrib><creatorcontrib>Blankenship, Samuel</creatorcontrib><creatorcontrib>Barnes, Brian C.</creatorcontrib><creatorcontrib>Son, Steven F.</creatorcontrib><creatorcontrib>Strachan, Alejandro</creatorcontrib><title>Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials</title><title>Propellants, explosives, pyrotechnics</title><description>Data science and artificial intelligence are playing an increasingly important role in the physical sciences. Unfortunately, in the field of energetic materials data scarcity limits the accuracy and even applicability of ML tools. To address data limitations, we compiled multi‐modal data: both experimental and computational results for several properties. We find that multi‐task neural networks can learn from multi‐modal data and outperform single‐task models trained for specific properties. As expected, the improvement is more significant for data‐scarce properties. These models are trained using descriptors built from simple molecular information and can be readily applied for large‐scale materials screening to explore multiple properties simultaneously. This approach is widely applicable to fields outside energetic materials.</description><issn>0721-3115</issn><issn>1521-4087</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNo9kLtOwzAYhS0EEqGwMvsFEvw7iS8jqlpASgVDmSPX-V0ZQhLZZujGI_CMPAmpuEzfOcM5w0fINbACGOM3U8Cp4IxXc6nUCcmg5pBXTMlTkjE55xKgPicXMb4wNk8YZKTZvPfJf318bk18pX9l7TvsfTrQBk0Y_LCno6NPYZwwJI-RujHQ1YBhj8lbujEJgzd9vCRnbgZe_XJBnter7fI-bx7vHpa3TW5BVCkHIZTjHVqnlQWwtVXS7KwSJXA0UjsQO8l4LazWJWorZWU5KK1d5yRztlyQ4ufXhjHGgK6dgn8z4dACa48u2qOL9t9F-Q2E0FUH</recordid><startdate>20241123</startdate><enddate>20241123</enddate><creator>Appleton, Robert J.</creator><creator>Klinger, Daniel</creator><creator>Lee, Brian H.</creator><creator>Taylor, Michael</creator><creator>Kim, Sohee</creator><creator>Blankenship, Samuel</creator><creator>Barnes, Brian C.</creator><creator>Son, Steven F.</creator><creator>Strachan, Alejandro</creator><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20241123</creationdate><title>Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials</title><author>Appleton, Robert J. ; Klinger, Daniel ; Lee, Brian H. ; Taylor, Michael ; Kim, Sohee ; Blankenship, Samuel ; Barnes, Brian C. ; Son, Steven F. ; Strachan, Alejandro</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c164t-1668f2decf98c11c5c87abc86312ea79f16b70256c993e9c774c21899fdf70fc3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Appleton, Robert J.</creatorcontrib><creatorcontrib>Klinger, Daniel</creatorcontrib><creatorcontrib>Lee, Brian H.</creatorcontrib><creatorcontrib>Taylor, Michael</creatorcontrib><creatorcontrib>Kim, Sohee</creatorcontrib><creatorcontrib>Blankenship, Samuel</creatorcontrib><creatorcontrib>Barnes, Brian C.</creatorcontrib><creatorcontrib>Son, Steven F.</creatorcontrib><creatorcontrib>Strachan, Alejandro</creatorcontrib><collection>CrossRef</collection><jtitle>Propellants, explosives, pyrotechnics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Appleton, Robert J.</au><au>Klinger, Daniel</au><au>Lee, Brian H.</au><au>Taylor, Michael</au><au>Kim, Sohee</au><au>Blankenship, Samuel</au><au>Barnes, Brian C.</au><au>Son, Steven F.</au><au>Strachan, Alejandro</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials</atitle><jtitle>Propellants, explosives, pyrotechnics</jtitle><date>2024-11-23</date><risdate>2024</risdate><issn>0721-3115</issn><eissn>1521-4087</eissn><abstract>Data science and artificial intelligence are playing an increasingly important role in the physical sciences. Unfortunately, in the field of energetic materials data scarcity limits the accuracy and even applicability of ML tools. To address data limitations, we compiled multi‐modal data: both experimental and computational results for several properties. We find that multi‐task neural networks can learn from multi‐modal data and outperform single‐task models trained for specific properties. As expected, the improvement is more significant for data‐scarce properties. These models are trained using descriptors built from simple molecular information and can be readily applied for large‐scale materials screening to explore multiple properties simultaneously. This approach is widely applicable to fields outside energetic materials.</abstract><doi>10.1002/prep.202400248</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0721-3115
ispartof Propellants, explosives, pyrotechnics, 2024-11
issn 0721-3115
1521-4087
language eng
recordid cdi_crossref_primary_10_1002_prep_202400248
source Access via Wiley Online Library
title Multi‐Task Multi‐Fidelity Learning of Properties for Energetic Materials
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T00%3A26%3A51IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-crossref&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Multi%E2%80%90Task%20Multi%E2%80%90Fidelity%20Learning%20of%20Properties%20for%20Energetic%20Materials&rft.jtitle=Propellants,%20explosives,%20pyrotechnics&rft.au=Appleton,%20Robert%20J.&rft.date=2024-11-23&rft.issn=0721-3115&rft.eissn=1521-4087&rft_id=info:doi/10.1002/prep.202400248&rft_dat=%3Ccrossref%3E10_1002_prep_202400248%3C/crossref%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true