Transfer Learning to Accelerate Interface Structure Searches

Interfaces have atomic structures that are significantly different from those in the bulk, and play crucial roles in material properties. The central structures at the interfaces that provide properties have been extensively investigated. However, determination of even one interface structure requires searching for the stable configuration among many thousands of candidates. Here, a powerful combination of machine learning techniques based on kriging and transfer learning (TL) is proposed as a method for unveiling the interface structures. Using the kriging+TL method, thirty-three grain boundaries were systematically determined from 1,650,660 candidates in only 462 calculations, representing an increase in efficiency over conventional all-candidate calculation methods, by a factor of approximately 3,600.

[1]  Hak-Sung Lee,et al.  Controlling Interface Intermixing and Properties of SrTiO3‐Based Superlattices , 2011 .

[2]  Julian D. Gale,et al.  GULP: A computer program for the symmetry-adapted simulation of solids , 1997 .

[3]  Y. Tokura,et al.  Interface-driven topological Hall effect in SrRuO3-SrIrO3 bilayer , 2016, Science Advances.

[4]  Taro Hitosugi,et al.  Rapid prediction of molecule arrangements on metal surfaces via Bayesian optimization , 2017 .

[5]  Howard L. Heinisch,et al.  The effects of grain boundary structure on binding of He in Fe , 2004 .

[6]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[7]  J. Mannhart,et al.  Tunable Quasi-Two-Dimensional Electron Gases in Oxide Heterostructures , 2006, Science.

[8]  J. Buban,et al.  Grain Boundary Strengthening in Alumina by Rare Earth Impurities , 2006, Science.

[9]  Yuichi Ikuhara,et al.  Direct measurements of grain boundary sliding in yttrium-doped alumina bicrystals , 2003 .

[10]  Koji Tsuda,et al.  Acceleration of stable interface structure searching using a kriging approach , 2016 .

[11]  M. Finnis,et al.  A simple empirical N-body potential for transition metals , 1984 .

[12]  Hideharu Nakashima,et al.  Grain Boundary Energy and Structure of α-Fe Symmetric Tilt Boundary , 2000 .

[13]  Hideo Hosono,et al.  Giant thermoelectric Seebeck coefficient of a two-dimensional electron gas in SrTiO3. , 2007, Nature materials.

[14]  Robert O. Ritchie,et al.  Effects of Grain‐Boundary Structure on the Strength, Toughness, and Cyclic‐Fatigue Properties of a Monolithic Silicon Carbide , 2004 .

[15]  L. Zhang,et al.  Relationship between misorientation and bismuth induced embrittlement of [001] tilt boundary in copper bicrystal , 1995 .

[16]  Michael W. Finnis,et al.  Bismuth embrittlement of copper is an atomic size effect , 2004, Nature.

[17]  Akira Ohtomo,et al.  A high-mobility electron gas at the LaAlO3/SrTiO3 heterointerface , 2004, Nature.

[18]  Shin Kiyohara,et al.  Prediction of interface structures and energies via virtual screening , 2016, Science Advances.

[19]  Yi-Feng Lin,et al.  Insight into the grain boundary effect on the ionic transport of yttria-stabilized zirconia at eleva , 2011 .

[20]  Koji Tsuda,et al.  COMBO: An efficient Bayesian optimization library for materials science , 2016 .