JAIST Repository >
a. 知識科学研究科・知識科学系 >
a11. 会議発表論文 >
a11-1. 会議発表論文 >

このアイテムの引用には次の識別子を使用してください: http://hdl.handle.net/10119/14785

タイトル: Novel mixture model for the representation of potential energy surfaces
著者: Pham, Tien Lam
Kino, Hiori
Terakura, Kiyoyuki
Miyake, Takashi
Dam, Hieu Chi
キーワード: Materials informatics
Data mining
Machine learning
Potential energy surface
発行日: 2016-10-17
出版者: American Institute of Physics
誌名: The Journal of Chemical Physics
巻: 145
号: 15
開始ページ: 154103-1
終了ページ: 154103-6
DOI: 10.1063/1.4964318
抄録: We demonstrate that knowledge of chemical physics on a materials system can be automatically extracted from first-principles calculations using a data mining technique; this information can then be utilized to construct a simple empirical atomic potential model. By using unsupervised learning of the generative Gaussian mixture model, physically meaningful patterns of atomic local chemical environments can be detected automatically. Based on the obtained information regarding these atomic patterns, we propose a chemical-structure-dependent linear mixture model for estimating the atomic potential energy. Our experiments show that the proposed mixture model significantly improves the accuracy of the prediction of the potential energy surface for complex systems that possess a large diversity in their local structures.
Rights: Copyright 2016 American Institute of Physics. This article may be downloaded for personal use only. Any other use requires prior permission of the author and the American Institute of Physics. The following article appeared in Tien Lam Pham, Hiori Kino, Kiyoyuki Terakura, Takashi Miyake and Hieu Chi Dam, The Journal of Chemical Physics, 145(15), 154103 (2016) and may be found at http://dx.doi.org/10.1063/1.4964318
URI: http://hdl.handle.net/10119/14785
資料タイプ: publisher
出現コレクション:a11-1. 会議発表論文 (Conference Papers)


ファイル 記述 サイズ形式
22980.pdf2044KbAdobe PDF見る/開く



お問い合わせ先 : 北陸先端科学技術大学院大学 研究推進課図書館情報係