论文标题
一种基于物理的神经网络的方法,可以设计全dielectric metasurfaces
A Physics Based Approach for Neural Networks Enabled Design of All-Dielectric Metasurfaces
论文作者
论文摘要
机器学习方法在各种学科中找到了新的应用领域,因为它们为复杂问题提供了低计算的成本解决方案。最近,Metasurface Design在这些应用程序中加入了,神经网络在短时间内实现了重大改进。但是,仍然需要克服杰出的挑战。在这里,我们根据物理问题的管理定律提出了一种数据预处理方法,以消除高维光学响应与跨索面的低维特征空间之间的维度不匹配。我们向前训练和反向模型,以预测圆柱元原子的光学响应,并分别检索其几何参数以产生所需的光学响应。我们的方法即使在训练频谱范围之外也可以提供准确的预测能力。最后,使用我们的反向模型,我们设计并演示了焦点金属作为概念验证应用程序,从而验证了我们提出的方法的能力。我们认为,我们的方法将为实用的基于学习的模型铺平道路,以解决更复杂的光子设计问题。
Machine learning methods have found novel application areas in various disciplines as they offer low-computational cost solutions to complex problems. Recently, metasurface design has joined among these applications, and neural networks enabled significant improvements within a short period of time. However, there are still outstanding challenges that needs to be overcome. Here, we propose a data pre-processing approach based on the governing laws of the physical problem to eliminate dimensional mismatch between high dimensional optical response and low dimensional feature space of metasurfaces. We train forward and inverse models to predict optical responses of cylindrical meta-atoms and to retrieve their geometric parameters for a desired optical response, respectively. Our approach provides accurate prediction capability even outside the training spectral range. Finally, using our inverse model, we design and demonstrate a focusing metalens as a proof-of-concept application, thus validating the capability of our proposed approach. We believe our method will pave the way towards practical learning-based models to solve more complicated photonic design problems.