Litcius/Paper detail

High-efficient <i>ab initio</i> Bayesian active learning method and applications in prediction of two-dimensional functional materials

Xingyu Ma, Hou-Yi Lyu, Kuan-Rong Hao, Zhen‐Gang Zhu, Qing‐Bo Yan, Gang Su

2021Nanoscale19 citationsDOIOpen Access PDF

Abstract

calculations to accelerate the prediction of desired functional materials with ultrahigh efficiency and accuracy. We apply it as an instance to a large family (3119) of two-dimensional hexagonal binary compounds with unbalanced materials properties, and accurately screen out the materials with maximal electric polarization and proper photovoltaic band gaps, respectively, whereas the computational costs are significantly reduced by only calculating a few tenths of the possible candidates in comparison with a random search. This approach shows the enormous advantages for the cases with unbalanced distribution of target properties. It can be readily applied to seek a broad range of advanced materials.

Topics & Concepts

Ab initioComputer scienceProperty (philosophy)Binary numberRange (aeronautics)Artificial intelligenceBayesian probabilityMachine learningDensity functional theoryMaterials scienceAlgorithmComputational chemistryMathematicsPhysicsChemistryPhilosophyEpistemologyComposite materialArithmeticQuantum mechanicsMachine Learning in Materials ScienceElectronic and Structural Properties of Oxides2D Materials and Applications