摘要
由一名新闻记者-机器人与机器学习每日新闻编辑-调查人员讨论人工智能的新发现。根据NewsRx记者从帝国理工学院伦敦分校传来的消息，研究表明：“缺陷决定了许多功能材料的性能。”这项研究的资助者包括Rcuk|工程和物理科学研究委员会。我们的新闻记者从伦敦帝国理工学院的研究中获得了一句话：“为了了解缺陷的行为及其对物理性能的影响，有必要确定最稳定的缺陷几何形状。然而，对于高通量缺陷研究或具有复杂缺陷景观的材料，如合金或disor dered固体，全局结构搜索在计算上具有挑战性。这里，我们利用机器学习的urrogate模型来定性地探索中性点缺陷的结构景观，通过学习一族相关金属硫属化合物和混合阴离子晶体中的缺陷基元，该模型成功地预测了90%的情况下看不见的成分中看不见的缺陷的有利重构，而第一性原理计算的数量减少了73%。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators discuss new findings in artificial intelligence. According to news originating from Imperial College Lon don by NewsRx correspondents, research stated, "Defects dictate the properties o f many functional materials." Funders for this research include Rcuk | Engineering And Physical Sciences Resea rch Council. Our news correspondents obtained a quote from the research from Imperial College London: "To understand the behaviour of defects and their impact on physical pr operties, it is necessary to identify the most stable defect geometries. However , global structure searching is computationally challenging for high-throughput defect studies or materials with complex defect landscapes, like alloys or disor dered solids. Here, we tackle this limitation by harnessing a machine-learning s urrogate model to qualitatively explore the structural landscape of neutral poin t defects. By learning defect motifs in a family of related metal chalcogenide a nd mixed anion crystals, the model successfully predicts favourable reconstructi ons for unseen defects in unseen compositions for 90% of cases, th ereby reducing the number of first-principles calculations by 73%."
NETL
NSTL