Virtual screening of inorganic materials synthesis parameters with deep learning

被引:152
作者
Kim, Edward [1 ]
Huang, Kevin [1 ]
Jegelka, Stefanie [2 ]
Olivetti, Elsa [1 ]
机构
[1] MIT, Dept Mat Sci & Engn, Cambridge, MA 02139 USA
[2] MIT, Dept EECS & CSAIL, 77 Massachusetts Ave, Cambridge, MA 02139 USA
基金
美国国家科学基金会; 加拿大自然科学与工程研究理事会;
关键词
BROOKITE; NANOCRYSTALS; TEMPERATURE; DISCOVERY; TITANATE; ANATASE; BATIO3; PHASES; OXIDES; RUTILE;
D O I
10.1038/s41524-017-0055-6
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Virtual materials screening approaches have proliferated in the past decade, driven by rapid advances in first-principles computational techniques, and machine-learning algorithms. By comparison, computationally driven materials synthesis screening is still in its infancy, and is mired by the challenges of data sparsity and data scarcity: Synthesis routes exist in a sparse, high-dimensional parameter space that is difficult to optimize over directly, and, for some materials of interest, only scarce volumes of literature-reported syntheses are available. In this article, we present a framework for suggesting quantitative synthesis parameters and potential driving factors for synthesis outcomes. We use a variational autoencoder to compress sparse synthesis representations into a lower dimensional space, which is found to improve the performance of machine-learning tasks. To realize this screening framework even in cases where there are few literature data, we devise a novel data augmentation methodology that incorporates literature synthesis data from related materials systems. We apply this variational autoencoder framework to generate potential SrTiO3 synthesis parameter sets, propose driving factors for brookite TiO2 formation, and identify correlations between alkali-ion intercalation and MnO2 polymorph selection.
引用
收藏
页数:9
相关论文
共 76 条
[1]  
Abadi M., 2016, TENSORFLOW LARGE SCA
[2]   Low Data Drug Discovery with One-Shot Learning [J].
Altae-Tran, Han ;
Ramsundar, Bharath ;
Pappu, Aneesh S. ;
Pande, Vijay .
ACS CENTRAL SCIENCE, 2017, 3 (04) :283-293
[3]  
[Anonymous], 2012, ADV NEURAL INF PROCE
[4]  
[Anonymous], 2017, ABS170401212 CORR
[5]  
[Anonymous], 2002, Data Sci. J., DOI DOI 10.2481/DSJ.1.19
[6]  
[Anonymous], 2013, 1 INT C LEARN REPR I
[7]  
[Anonymous], 2011, J. Mach. Learn. Res.
[8]  
[Anonymous], 2014, P IEEE COMP SOC C CO
[9]  
[Anonymous], 2013, ADV NEURAL INF PROCE
[10]  
[Anonymous], AUTOMATIC CHEM DESIG