Nanophotonic particle simulation and inverse design using artificial neural networks

被引：732

作者：

Peurifoy, John ^{[1
]}

Shen, Yichen ^{[1
]}

Jing, Li ^{[1
]}

Yang, Yi ^{[1
,2
]}

Cano-Renteria, Fidel ^{[3
]}

DeLacy, Brendan G. ^{[4
]}

Joannopoulos, John D. ^{[1
]}

Tegmark, Max ^{[1
]}

Soljacic, Marin ^{[1
]}

机构：

[1] MIT, Dept Phys, Cambridge, MA 02139 USA

[2] MIT, Dept Elect Engn & Comp Sci, Cambridge, MA 02139 USA

[3] MIT, Dept Math, Cambridge, MA 02139 USA

[4] US Army, Edgewood Chem Biol Ctr, Aberdeen Proving Ground, MD 21010 USA

来源：

SCIENCE ADVANCES | 2018年 / 4卷 / 06期

基金：

美国国家科学基金会;

关键词：

OPTIMIZATION;

D O I：

10.1126/sciadv.aar4206

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

We propose a method to use artificial neural networks to approximate light scattering by multilayer nanoparticles. We find that the network needs to be trained on only a small sampling of the data to approximate the simulation to high precision. Once the neural network is trained, it can simulate such optical processes orders of magnitude faster than conventional simulations. Furthermore, the trained neural network can be used to solve nanophotonic inverse design problems by using back propagation, where the gradient is analytical, not numerical.

引用

页数：7

共 21 条

[1]

[Anonymous], FRONTIERS OPTICS

[2]

Apagyi B, 1996, INVERSE ALGEBRAIC QU

[3]

Bohren C. F., 1983, Absorption and Scattering of Light by Small Particles, DOI 10.1002/9783527618156

[4] Solving the quantum many-body problem with artificial neural networks [J].