Revisiting Computational Thermodynamics through Machine Learning of High-Dimensional Data

Author

SRINIVASAN, SUDARSHAN ; Rajan, K.

Volume

15

Issue

5

fYear

2013

fDate

Sept.-Oct. 2013

Firstpage

22

Lastpage

31

Abstract

A new perspective on alloy thermodynamics computation uses data-driven analysis and machine learning for the design and discovery of materials. The focus is on an integrated machine-learning framework, coupling different genres of supervised and unsupervised informatics techniques, and bridging two distinct viewpoints: continuum representations based on solid solution thermodynamics and discrete high-dimensional elemental descriptions.

Keywords

alloys; data analysis; learning (artificial intelligence); materials science computing; thermal stability; alloy thermodynamics computation; computational thermodynamics; continuum representations; data-driven analysis; discrete high-dimensional elemental descriptions; high-dimensional data; integrated machine learning framework; material design; material discovery; solid solution thermodynamics; supervised informatics techniques; unsupervised informatics techniques; Atomic measurements; Computational modeling; Informatics; Machine learning; Principal component analysis; Semiconductor materials; Thermodynamics; bandgap engineering; compound semiconductors; computational thermodynamics; data mining; high-dimensional model representation; machine learning; materials informatics;

fLanguage

English

Journal_Title

Computing in Science & Engineering

Publisher

ieee

ISSN

1521-9615

Type

jour

DOI

10.1109/MCSE.2013.76

Filename

6576112