The lack of reliable methods for identifying descriptors - the sets of
parameters capturing the underlying mechanisms of a materials property - is one
of the key factors hindering efficient materials development. Here, we propose
a systematic approach for discovering descriptors for materials properties,
within the framework of compressed-sensing based dimensionality reduction.
SISSO (sure independence screening and sparsifying operator) tackles immense
and correlated features spaces, and converges to the optimal solution from a
combination of features relevant to the materials' property of interest. In
addition, SISSO gives stable results also with small training sets. The
methodology is benchmarked with the quantitative prediction of the ground-state
enthalpies of octet binary materials (using ab initio data) and applied to the
showcase example of predicting the metal/insulator classification of binaries
(with experimental data). Accurate, predictive models are found in both cases.
For the metal-insulator classification model, the predictive capability are
tested beyond the training data: It rediscovers the available pressure-induced
insulator->metal transitions and it allows for the prediction of yet unknown
transition candidates, ripe for experimental validation. As a step forward with
respect to previous model-identification methods, SISSO can become an effective
tool for automatic materials development.
Description
SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates
%0 Generic
%1 ouyang2017sisso
%A Ouyang, Runhai
%A Curtarolo, Stefano
%A Ahmetcik, Emre
%A Scheffler, Matthias
%A Ghiringhelli, Luca M.
%D 2017
%K machinelearning
%R 10.1103/PhysRevMaterials.2.083802
%T SISSO: a compressed-sensing method for identifying the best
low-dimensional descriptor in an immensity of offered candidates
%U http://arxiv.org/abs/1710.03319
%X The lack of reliable methods for identifying descriptors - the sets of
parameters capturing the underlying mechanisms of a materials property - is one
of the key factors hindering efficient materials development. Here, we propose
a systematic approach for discovering descriptors for materials properties,
within the framework of compressed-sensing based dimensionality reduction.
SISSO (sure independence screening and sparsifying operator) tackles immense
and correlated features spaces, and converges to the optimal solution from a
combination of features relevant to the materials' property of interest. In
addition, SISSO gives stable results also with small training sets. The
methodology is benchmarked with the quantitative prediction of the ground-state
enthalpies of octet binary materials (using ab initio data) and applied to the
showcase example of predicting the metal/insulator classification of binaries
(with experimental data). Accurate, predictive models are found in both cases.
For the metal-insulator classification model, the predictive capability are
tested beyond the training data: It rediscovers the available pressure-induced
insulator->metal transitions and it allows for the prediction of yet unknown
transition candidates, ripe for experimental validation. As a step forward with
respect to previous model-identification methods, SISSO can become an effective
tool for automatic materials development.
@misc{ouyang2017sisso,
abstract = {The lack of reliable methods for identifying descriptors - the sets of
parameters capturing the underlying mechanisms of a materials property - is one
of the key factors hindering efficient materials development. Here, we propose
a systematic approach for discovering descriptors for materials properties,
within the framework of compressed-sensing based dimensionality reduction.
SISSO (sure independence screening and sparsifying operator) tackles immense
and correlated features spaces, and converges to the optimal solution from a
combination of features relevant to the materials' property of interest. In
addition, SISSO gives stable results also with small training sets. The
methodology is benchmarked with the quantitative prediction of the ground-state
enthalpies of octet binary materials (using ab initio data) and applied to the
showcase example of predicting the metal/insulator classification of binaries
(with experimental data). Accurate, predictive models are found in both cases.
For the metal-insulator classification model, the predictive capability are
tested beyond the training data: It rediscovers the available pressure-induced
insulator->metal transitions and it allows for the prediction of yet unknown
transition candidates, ripe for experimental validation. As a step forward with
respect to previous model-identification methods, SISSO can become an effective
tool for automatic materials development.},
added-at = {2019-04-06T00:57:25.000+0200},
author = {Ouyang, Runhai and Curtarolo, Stefano and Ahmetcik, Emre and Scheffler, Matthias and Ghiringhelli, Luca M.},
biburl = {https://www.bibsonomy.org/bibtex/2cd9591e6ff99f0720b0dbfc545df7284/thiago.buzelli},
description = {SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates},
doi = {10.1103/PhysRevMaterials.2.083802},
interhash = {eb25ead062e2e53ead33867450097c94},
intrahash = {cd9591e6ff99f0720b0dbfc545df7284},
keywords = {machinelearning},
note = {cite arxiv:1710.03319Comment: 11 pages, 5 figures, in press in Phys. Rev. Materials},
timestamp = {2019-04-06T00:57:25.000+0200},
title = {SISSO: a compressed-sensing method for identifying the best
low-dimensional descriptor in an immensity of offered candidates},
url = {http://arxiv.org/abs/1710.03319},
year = 2017
}