Show simple item record

dc.creatorZhou, Q
dc.creatorTang, P
dc.creatorLiu, S
dc.creatorPan, J
dc.creatorYan, Q
dc.creatorZhang, SC
dc.date.accessioned2020-12-14T20:33:45Z
dc.date.available2020-12-14T20:33:45Z
dc.date.issued2018-07-10
dc.identifier.issn0027-8424
dc.identifier.issn1091-6490
dc.identifier.doihttp://dx.doi.org/10.34944/dspace/4418
dc.identifier.other29946023 (pubmed)
dc.identifier.urihttp://hdl.handle.net/20.500.12613/4436
dc.description.abstract© 2018 National Academy of Sciences. All Rights Reserved. Exciting advances have been made in artificial intelligence (AI) during recent decades. Among them, applications of machine learning (ML) and deep learning techniques brought human-competitive performances in various tasks of fields, including image recognition, speech recognition, and natural language understanding. Even in Go, the ancient game of profound complexity, the AI player has already beat human world champions convincingly with and without learning from the human. In this work, we show that our unsupervised machines (Atom2Vec) can learn the basic properties of atoms by themselves from the extensive database of known compounds and materials. These learned properties are represented in terms of high-dimensional vectors, and clustering of atoms in vector space classifies them into meaningful groups consistent with human knowledge. We use the atom vectors as basic input units for neural networks and other ML models designed and trained to predict materials properties, which demonstrate significant accuracy.
dc.format.extentE6411-E6417
dc.language.isoen
dc.relation.haspartProceedings of the National Academy of Sciences of the United States of America
dc.relation.isreferencedbyProceedings of the National Academy of Sciences
dc.rightsCC BY-NC-ND
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectatomism
dc.subjectmachine learning
dc.subjectmaterials discovery
dc.titleLearning atoms for materials discovery
dc.typeArticle
dc.type.genreJournal Article
dc.relation.doi10.1073/pnas.1801181115
dc.ada.noteFor Americans with Disabilities Act (ADA) accommodation, including help with reading this content, please contact scholarshare@temple.edu
dc.date.updated2020-12-14T20:33:41Z
refterms.dateFOA2020-12-14T20:33:45Z


Files in this item

Thumbnail
Name:
Learning atoms for materials ...
Size:
1.116Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record

CC BY-NC-ND
Except where otherwise noted, this item's license is described as CC BY-NC-ND