Dirty engineering data-driven inverse prediction machine learning model

Abstract Most data-driven machine learning (ML) approaches established in metallurgy research fields are focused on a build-up of reliable quantitative models that predict a material property from a given set of material conditions. In general, the input feature dimension (the number of material con...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jin-Woong Lee, Woon Bae Park, Byung Do Lee, Seonghwan Kim, Nam Hoon Goo, Kee-Sun Sohn
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/97f45af18b5640c5a76a5470b7e82b31
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:97f45af18b5640c5a76a5470b7e82b31
record_format	dspace
spelling	oai:doaj.org-article:97f45af18b5640c5a76a5470b7e82b312021-12-02T11:42:13ZDirty engineering data-driven inverse prediction machine learning model10.1038/s41598-020-77575-02045-2322https://doaj.org/article/97f45af18b5640c5a76a5470b7e82b312020-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-77575-0https://doaj.org/toc/2045-2322Abstract Most data-driven machine learning (ML) approaches established in metallurgy research fields are focused on a build-up of reliable quantitative models that predict a material property from a given set of material conditions. In general, the input feature dimension (the number of material condition variables) is much higher than the output feature dimension (the number of material properties of concern). Rather than such a forward-prediction ML model, it is necessary to develop so-called inverse-design modeling, wherein required material conditions could be deduced from a set of desired material properties. Here we report a novel inverse design strategy that employs two independent approaches: a metaheuristics-assisted inverse reading of conventional forward ML models and an atypical inverse ML model based on a modified variational autoencoder. These two unprecedented approaches were successful and led to overlapped results, from which we pinpointed several novel thermo-mechanically controlled processed (TMCP) steel alloy candidates that were validated by a rule-based thermodynamic calculation tool (Thermo-Calc.). We also suggested a practical protocol to elucidate how to treat engineering data collected from industry, which is not prepared as independent and identically distributed (IID) random data.Jin-Woong LeeWoon Bae ParkByung Do LeeSeonghwan KimNam Hoon GooKee-Sun SohnNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-14 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Jin-Woong Lee Woon Bae Park Byung Do Lee Seonghwan Kim Nam Hoon Goo Kee-Sun Sohn Dirty engineering data-driven inverse prediction machine learning model
description	Abstract Most data-driven machine learning (ML) approaches established in metallurgy research fields are focused on a build-up of reliable quantitative models that predict a material property from a given set of material conditions. In general, the input feature dimension (the number of material condition variables) is much higher than the output feature dimension (the number of material properties of concern). Rather than such a forward-prediction ML model, it is necessary to develop so-called inverse-design modeling, wherein required material conditions could be deduced from a set of desired material properties. Here we report a novel inverse design strategy that employs two independent approaches: a metaheuristics-assisted inverse reading of conventional forward ML models and an atypical inverse ML model based on a modified variational autoencoder. These two unprecedented approaches were successful and led to overlapped results, from which we pinpointed several novel thermo-mechanically controlled processed (TMCP) steel alloy candidates that were validated by a rule-based thermodynamic calculation tool (Thermo-Calc.). We also suggested a practical protocol to elucidate how to treat engineering data collected from industry, which is not prepared as independent and identically distributed (IID) random data.
format	article
author	Jin-Woong Lee Woon Bae Park Byung Do Lee Seonghwan Kim Nam Hoon Goo Kee-Sun Sohn
author_facet	Jin-Woong Lee Woon Bae Park Byung Do Lee Seonghwan Kim Nam Hoon Goo Kee-Sun Sohn
author_sort	Jin-Woong Lee
title	Dirty engineering data-driven inverse prediction machine learning model
title_short	Dirty engineering data-driven inverse prediction machine learning model
title_full	Dirty engineering data-driven inverse prediction machine learning model
title_fullStr	Dirty engineering data-driven inverse prediction machine learning model
title_full_unstemmed	Dirty engineering data-driven inverse prediction machine learning model
title_sort	dirty engineering data-driven inverse prediction machine learning model
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/97f45af18b5640c5a76a5470b7e82b31
work_keys_str_mv	AT jinwoonglee dirtyengineeringdatadriveninversepredictionmachinelearningmodel AT woonbaepark dirtyengineeringdatadriveninversepredictionmachinelearningmodel AT byungdolee dirtyengineeringdatadriveninversepredictionmachinelearningmodel AT seonghwankim dirtyengineeringdatadriveninversepredictionmachinelearningmodel AT namhoongoo dirtyengineeringdatadriveninversepredictionmachinelearningmodel AT keesunsohn dirtyengineeringdatadriveninversepredictionmachinelearningmodel
_version_	1718395357071671296

Dirty engineering data-driven inverse prediction machine learning model

Ejemplares similares