Peng Rixi, Ren Simiao, Malof Jordan, Padilla Willie J
Electrical and Computer Engineering, Duke University, Durham, NC, USA.
Computer Science, University of Montana, Missoula, MT, USA.
Nanophotonics. 2024 Mar 22;13(13):2323-2334. doi: 10.1515/nanoph-2023-0691. eCollection 2024 May.
We demonstrate transfer learning as a tool to improve the efficacy of training deep learning models based on residual neural networks (ResNets). Specifically, we examine its use for study of multi-scale electrically large metasurface arrays under open boundary conditions in electromagnetic metamaterials. Our aim is to assess the efficiency of transfer learning across a range of problem domains that vary in their resemblance to the original base problem for which the ResNet model was initially trained. We use a quasi-analytical discrete dipole approximation (DDA) method to simulate electrically large metasurface arrays to obtain ground truth data for training and testing of our deep neural network. Our approach can save significant time for examining novel metasurface designs by harnessing the power of transfer learning, as it effectively mitigates the pervasive data bottleneck issue commonly encountered in deep learning. We demonstrate that for the best case when the transfer task is sufficiently similar to the target task, a new task can be effectively trained using only a few data points yet still achieve a test mean absolute relative error of 3 % with a pre-trained neural network, realizing data reduction by a factor of 1000.
我们展示了迁移学习作为一种工具，用于提高基于残差神经网络（ResNets）训练深度学习模型的效率。具体而言，我们研究了其在电磁超材料中开放边界条件下多尺度电大超表面阵列研究中的应用。我们的目标是评估迁移学习在一系列问题域中的效率，这些问题域与最初训练ResNet模型的原始基础问题的相似程度各不相同。我们使用准解析离散偶极子近似（DDA）方法来模拟电大超表面阵列，以获得用于训练和测试我们深度神经网络的真实数据。我们的方法通过利用迁移学习的力量，可以节省大量时间来研究新型超表面设计，因为它有效地缓解了深度学习中常见的普遍存在的数据瓶颈问题。我们证明，在迁移任务与目标任务足够相似的最佳情况下，仅使用几个数据点就可以有效地训练新任务，并且使用预训练的神经网络仍能实现3%的测试平均绝对相对误差，实现了1000倍的数据缩减。
