How to see hidden patterns in metamaterials with interpretable machine learning
How to see hidden patterns in metamaterials with interpretable machine learning
复制标题
DOI：
10.1016/j.eml.2022.101895
复制
发表时间：
2022-11-01
影响因子：
4.7
通讯作者：
Rudin, Cynthia
中科院分区：
文献类型：
--
作者：
Chen, Zhi;Ogren, Alexander;Rudin, Cynthia
机器学习模型可以通过近似计算昂贵的模拟器或解决逆向设计问题来辅助超材料设计。然而，过去的工作通常依赖于黑盒深度神经网络，其推理过程是不透明的，需要大量的数据集，获取成本很高。在这项工作中，我们开发了两种新的机器学习方法来发现超材料，这些方法都没有这些缺点。这些方法被称为形状频率特征和单元格模板，可以发现具有用户指定频率带隙的2D超材料。我们的方法提供了逻辑规则为基础的条件，超材料单元格，允许可解释的推理过程，并推广到不同分辨率的设计空间。该模板还提供了设计灵活性，用户几乎可以自由地设计单位单元的精细分辨率特征，而不会影响用户所需的带隙。(c)2022爱思唯尔有限公司保留所有权利。
Machine learning models can assist with metamaterials design by approximating computationally expensive simulators or solving inverse design problems. However, past work has usually relied on black box deep neural networks, whose reasoning processes are opaque and require enormous datasets that are expensive to obtain. In this work, we develop two novel machine learning approaches to metamaterials discovery that have neither of these disadvantages. These approaches, called shape-frequency features and unit-cell templates, can discover 2D metamaterials with user-specified frequency band gaps. Our approaches provide logical rule-based conditions on metamaterial unit-cells that allow for interpretable reasoning processes, and generalize well across design spaces of different resolutions. The templates also provide design flexibility where users can almost freely design the fine resolution features of a unit-cell without affecting the user's desired band gap.(c) 2022 Elsevier Ltd. All rights reserved.