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.

翻译：机器学习模型可以通过接近于计算成本昂贵的模拟器或解决反向设计问题来帮助设计元材料。然而,过去的工作通常依赖于黑盒深神经网络,其推理过程不透明,需要大量昂贵的数据集才能获得。在这项工作中,我们开发了两种新颖的机器学习方法来发现元材料,这些方法没有这些缺点。这些方法被称为形状-频率特征和单元-细胞模板,可以发现2D元材料存在用户指定的频带差距。我们的方法为元材料单元-细胞提供了逻辑的基于规则的条件,允许进行可解释的推理过程,并在不同分辨率的设计空间中全面推广。模板还提供了设计灵活性,用户可以几乎自由地设计一个单元-细胞的精细分辨率特征,而不会影响用户想要的波段差距。