Materials Design Through Batch Bayesian Optimization with Multisource Information Fusion
Materials Design Through Batch Bayesian Optimization with Multisource Information Fusion
复制标题
DOI：
10.1007/s11837-020-04396-x
复制
发表时间：
2020-10
期刊：
影响因子：
2.6
通讯作者：
Richard Couperthwaite;Abhilash Molkeri;Danial Khatamsaz;Ankit Srivastava;D. Allaire;R. Arróyave
中科院分区：
文献类型：
--
作者：
Richard Couperthwaite;Abhilash Molkeri;Danial Khatamsaz;Ankit Srivastava;D. Allaire;R. Arróyave
综合计算材料工程（ICME）要求将模拟工具和实验相结合，以加速材料的发展。ICME方法往往计算成本高，最近，贝叶斯优化（BO）被提出作为一种使ICME更有效地利用资源的方法。然而，传统的BO本质上是连续的（即一次一个），当材料设计选择的评估成本很高时，这使得它非常耗时。虽然传统的高通量方法能够以并行方式合成和表征（或模拟）材料，但它们往往是“开环”的，并且一旦并行实验/模拟进行和分析，就无法提供下一步尝试的建议。在这里，我们通过引入一个批量BO框架来解决这个问题，该框架能够以并行的方式探索材料的设计空间。我们通过合并能够集成多个信息源的信息融合框架来增强这种方法。在双相钢的计算设计中展示了所提出的方法，我们表明批量BO可以显著减少执行设计任务所需的时间和资源。所建议的方法具有更广泛的适用性，远远超出了用于演示它的ICME示例。
Integrated computational materials engineering (ICME) calls for the integration of simulation tools and experiments to accelerate the development of materials. ICME approaches tend to be computationally costly, and recently, Bayesian optimization (BO) has been proposed as a way to make ICME more resource efficient. Conventional BO, however, is sequential (i.e., one-at-a-time) in nature, which makes it very time-consuming when the evaluation of a materials design choice is costly. While conventional high-throughput approaches enable the synthesis and characterization (or simulation) of materials in a parallel manner, they tend to be “open loop” and are unable to provide recommendations of what to try next once the parallel experiment/simulation has been carried out and analyzed. Here, we address this problem by introducing a batch BO framework that enables the exploration of the material’s design space in a parallel fashion. We augment this approach by incorporating information fusion frameworks capable of integrating multiple information sources. Demonstrating the proposed approach in the computational design of dual-phase steel, we show that batch BO can result in a significant reduction in the time and resources needed to carry out the design task. The proposed approach has wider applicability, well beyond the ICME example used to demonstrate it.