arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.00710cond-mat.mtrl-sci
收藏

利用 Bayesian optimization 推断能量-成分关系以增强无机材料的探索

Inferring energy-composition relationships with Bayesian optimization enhances exploration of inorganic materials

Andrij Vasylenko, Benjamin Asher, Chris C. Collins, Michael W. Gaultois, George Darling, Matthew S. Dyer, Matthew J. Rosseinsky

AI总结:

提出基于 Bayesian optimization 的学习算法 PhaseBO,通过优化化学计量比实现成分空间的有效采样,在不损害能量评估准确性的前提下加速无机材料的发现。

摘要:

材料的成分空间计算探索可为合成研究提供指导,从而加速新型材料的发现。大多数方法采用高通量采样,侧重于缩短单个成分的能量评估时间,但往往以牺牲准确性为代价。在此,我们提出一种侧重于成分空间有效采样的替代方法。学习算法 PhaseBO 优化潜在目标材料的化学计量比,在不损害能量评估准确性的前提下,提高其发现概率并加速其发现过程。

英文摘要:

Computational exploration of the compositional spaces of materials can provide guidance for synthetic research and thus accelerate the discovery of novel materials. Most approaches employ high-throughput sampling and focus on reducing the time for energy evaluation for individual compositions, often at the cost of accuracy. Here, we present an alternative approach focusing on effective sampling of the compositional space. The learning algorithm PhaseBO optimizes the stoichiometry of the potential target material while improving the probability of and accelerating its discovery without compromising the accuracy of energy evaluation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑