English   español  
Please use this identifier to cite or link to this item: http://hdl.handle.net/10261/150439
Share/Impact:
Statistics
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Exportar a otros formatos:

Title

Beyond High throughput: Optimal Computational Materials Discovery

AuthorsRaymundo Arróyave
KeywordsComputational Materials Science
Issue DateJun-2017
AbstractThe goal-oriented discovery of materials requires the identification of the composition and process history necessary to achieve specific multi-scale structural features that in turn bring about the desired properties. To accelerate the materials discovery process, high-throughput (HT) computational and experimental methods have been proposed. Unfortunately, current HT computational and experimental approaches have severe limitations as they: (1) are incapable of dealing with the high dimensionality (composition, configurational, and microstructural degrees of freedom) and complexity (e.g., multi-physics) of the materials design space; (2) employ hardcoded workflows and lack flexibility to iteratively learn and adapt based on the knowledge acquired to assure balanced exploration and exploitation of the materials design space; (4) are suboptimal in resource allocation as experimental decisions do not account for the cost and time of experimentation. In this talk, we present some preliminary work in which we have adapted ideas from fields as diverse as Artificial Intelligence, Optimal Experimental Design, Global Optimization and Game Theory to develop a framework capable of optimally exploring the materials design space in order to attain an optimal materials response. Specifically, we use variants of the Efficient Global Optimization algorithm to deploy an autonomous computational materials discovery platform capable of performing optimal sequential computational experiments in order to find optimal materials. We demonstrate single and multi-objective optimization and use the so-called MAX phases as examples. Moreover, we show how this framework can be made robust against selection of non-informative features by using so-called Bayesian Model Averaging approaches.
DescriptionCiclo de Conferencias "Embajadores" del CENIM
URIhttp://hdl.handle.net/10261/150439
Appears in Collections:(CENIM) Cursos-Material didáctico
Files in This Item:
File Description SizeFormat 
Raimundo Arroyave.pdf52,19 kBAdobe PDFThumbnail
View/Open
Show full item record
Review this work
 


WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.