4-11、从数据到发现：加速搜索材料

From Data to Discovery: Accelerated search for materials with targeted properties

Turab Lookman^*

Los Alamos National Laboratory

摘要：Finding new materials with targeted properties with as few experiments as possible is a key goal of accelerated materials discovery and the materials genome initiative. The enormous complexity due to the interplay of structural, chemical and microstructural degrees of freedom in materials makes the rational design of new materials rather difficult. Machine learning and optimization, used in industry for solving complex problems, are increasingly being adapted for the design of new materials by learning from past data. However, the number of well characterized samples available as sources of data to learn from is often typically small. I will review some of the applications we have examined, including finding alloys and ceramics with desired properties, via an active learning loop using surrogate modeling and design.

关键词：Machine Learning，Active Learning， Design of experiment

DOI:10.12110/secondfmge.20181014.411