Transfer Learning for Modeling Plasmonic Nanowire Waveguides.

Transfer Learning for Modeling Plasmonic Nanowire Waveguides.
复制标题

DOI:
10.3390/nano12203624
复制
发表时间:
2022-10-16
期刊:
Nanomaterials (Basel, Switzerland)
影响因子:
--
通讯作者:
--
中科院分区:
其他
文献类型:
--
作者:

文献摘要

参考文献

相似文献

通过数值模拟恢复等离子体金属纳米线(MNW)的波导特性是耗时和计算资源消耗的,特别是对于那些具有突变几何特征和对称性破坏的纳米线。深度学习提供了一种替代方法,但由于泛化性能不足和需要大量训练数据,使用起来很有挑战性。在这里,我们克服了这些限制,提出了一种迁移学习方法建模MNW物理的指导下。我们表明,等离子体激元模式的基本知识可以首先从计算成本低廉的独立圆形MNW中学习,然后重新使用以显着提高预测具有各种复杂配置的MNW的波导特性的性能,从而实现更小的误差(约23-61%的减少),更少的可训练参数(约42%的减少)和更小的训练数据集(约50-80%的减少)。数值模拟相比,我们的模型减少了五个数量级的计算时间。与其他非深度学习方法(如圆形区域等效方法和对角圆近似)相比,我们的方法不仅具有更高的准确性,而且具有更全面的特征,为研究MNW提供了一个有效和高效的框架,可以大大促进极化激振元件和器件的设计。
Retrieving waveguiding properties of plasmonic metal nanowires (MNWs) through numerical simulations is time- and computational-resource-consuming, especially for those with abrupt geometric features and broken symmetries. Deep learning provides an alternative approach but is challenging to use due to inadequate generalization performance and the requirement of large sets of training data. Here, we overcome these constraints by proposing a transfer learning approach for modeling MNWs under the guidance of physics. We show that the basic knowledge of plasmon modes can first be learned from free-standing circular MNWs with computationally inexpensive data, and then reused to significantly improve performance in predicting waveguiding properties of MNWs with various complex configurations, enabling much smaller errors (~23–61% reduction), less trainable parameters (~42% reduction), and smaller sets of training data (~50–80% reduction) than direct learning. Compared to numerical simulations, our model reduces the computational time by five orders of magnitude. Compared to other non-deep learning methods, such as the circular-area-equivalence approach and the diagonal-circle approximation, our approach enables not only much higher accuracies, but also more comprehensive characterizations, offering an effective and efficient framework to investigate MNWs that may greatly facilitate the design of polaritonic components and devices.
DOI: 10.1364/oe.27.036414
发表时间: 2019-12-09
期刊: OPTICS EXPRESS
影响因子: 3.8
作者:
Chugh, Sunny;Gulistan, Aamir;Rahman, B. M. A.
通讯作者: Rahman, B. M. A.
用于混合纳米光子元件和电路的等离子体和光子纳米线的直接耦合
DOI: 10.1021/nl902860d
发表时间: 2009-12-01
期刊: NANO LETTERS
影响因子: 10.8
作者:
Guo, Xin;Qiu, Min;Tong, Limin
通讯作者: Tong, Limin
DOI: 10.1021/nn405183r
发表时间: 2014-01-01
期刊: ACS NANO
影响因子: 17.1
作者:
Nauert, Scott;Paul, Aniruddha;Link, Stephan
通讯作者: Link, Stephan
DOI: 10.1109/jlt.2019.2946572
发表时间: 2019-12-15
影响因子: 4.7
作者:
Chugh, Sunny;Ghosh, Souvik;Rahman, B. M. A.
通讯作者: Rahman, B. M. A.
用于等离子传播和高灵敏氢检测的单晶钯及其合金纳米线
DOI: 10.1002/adom.201300413
发表时间: 2014-02-01
影响因子: 9
作者:
Gu, Fuxing;Zeng, Heping;Zhuang, Songlin
通讯作者: Zhuang, Songlin