Sample-efficient deep learning for accelerating photonic inverse design
Sample-efficient deep learning for accelerating photonic inverse design
复制标题
DOI：
10.1364/osac.420977
复制
发表时间：
2021-03-15
期刊：
影响因子：
1.6
通讯作者：
Hegde, Ravi
中科院分区：
文献类型：
--
作者：
Hegde, Ravi
目前正在探索深度学习（DL）等数据驱动技术来解决光子学（尤其是纳米光子学）中的逆向设计问题，以处理材料和纳米结构的巨大搜索空间。要充分发挥这种方法的潜力，需要克服许多挑战；当前的工作流程特定于预定义的形状，并且需要在数据集创建和模型超参数搜索方面进行大量前期投资。我们报告了一种改进的工作流程，用于基于深度学习的进化优化加速，适用于过去的模拟数据不存在或严重不足的情况，并考虑多层薄膜光学设计的示例问题证明了其实用性。为了在替代训练中获得样本效率，提出了新的训练损失函数，强调模型预测结构相似的光谱响应的能力，而不是最小化局部逼近误差。该工作流程的目的是将基于深度学习的光学设计范围扩展到复杂的结构，其光谱的计算成本很高。 (C) 2021 年美国光学协会根据 OSA 开放获取出版协议的条款
Data-driven techniques like deep learning (DL) are currently being explored for inverse design problems in photonics (especially nanophotonics) to deal with the vast search space of materials and nanostructures. Many challenges need to be overcome to fully realize the potential of this approach; current workflows are specific to predefined shapes and require large upfront investments in dataset creation and model hyperparameter search. We report an improved workflow for DL based acceleration of evolutionary optimizations for scenarios where past simulation data is nonexistent or highly inadequate and demonstrate its utility considering the example problem of multilayered thin-film optics design. For obtaining sample-efficiency in surrogate training, novel training loss functions that emphasize a model's ability to predict a structurally similar spectral response rather than minimizing local approximation error are proposed. The workflow is of interest to extend the ambit of DL based optics design to complicated structures whose spectra are computationally expensive to calculate. (C) 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement