Iterative optimization of photonic crystal nanocavity designs by using deep neural networks

Abstract Devices based on two-dimensional photonic-crystal nanocavities, which are defined by their air hole patterns, usually require a high quality (Q) factor to achieve high performance. We demonstrate that hole patterns with very high Q factors can be efficiently found by the iteration procedure consisting of machine learning of the relation between the hole pattern and the corresponding Q factor and new dataset generation based on the regression function obtained by machine learning. First, a dataset comprising randomly generated cavity structures and their first principles Q factors is prepared. Then a deep neural network is trained using the initial dataset to obtain a regression function that approximately predicts the Q factors from the structural parameters. Several candidates for higher Q factors are chosen by searching the parameter space using the regression function. After adding these new structures and their first principles Q factors to the training dataset, the above process is repeated. As an example, a standard silicon-based L3 cavity is optimized by this method. A cavity design with a high Q factor exceeding 11 million is found within 101 iteration steps and a total of 8070 cavity structures. This theoretical Q factor is more than twice the previously reported record values of the cavity designs detected by the evolutionary algorithm and the leaky mode visualization method. It is found that structures with higher Q factors can be detected within less iteration steps by exploring not only the parameter space near the present highest-Q structure but also that distant from the present dataset.

[1]  Jelena Vuckovic,et al.  Inverse design of nanophotonic structures using complementary convex optimization , 2010 .

[2]  Zongfu Yu,et al.  Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures , 2017, 2019 Conference on Lasers and Electro-Optics (CLEO).

[3]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[4]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[5]  T. Asano,et al.  Photonic crystal nanocavity with a Q-factor of ~9 million. , 2014, Optics express.

[6]  Jelena Vucković,et al.  Inverse design in nanophotonics , 2018, Nature Photonics.

[7]  Yi Yang,et al.  Nanophotonic particle simulation and inverse design using artificial neural networks , 2018, Science Advances.

[8]  Dario Gerace,et al.  Genetically designed L3 photonic crystal nanocavities with measured quality factor exceeding one million , 2014 .

[9]  Siegfried Janz,et al.  Mapping the global design space of nanophotonic components using machine learning pattern recognition , 2018, Nature Communications.

[10]  Zin Lin,et al.  Topology optimization of multi-track ring resonators and 2D microcavities for nonlinear frequency conversion. , 2017, Optics letters.

[11]  M. Notomi,et al.  Ultrahigh-Q two-dimensional photonic crystal slab nanocavities in very thin barriers , 2008 .

[12]  Kent D. Choquette,et al.  Optimization of a single defect photonic crystal laser cavity , 2008 .

[13]  Oskar Painter,et al.  Momentum space design of high-Q photonic crystal optical cavities. , 2002, Optics express.

[14]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[15]  Prabhat,et al.  Scalable Bayesian Optimization Using Deep Neural Networks , 2015, ICML.

[16]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[17]  Susumu Noda,et al.  Photonic crystal nanocavity with a Q factor exceeding eleven million. , 2017, Optics express.

[18]  Vincenzo Savona,et al.  Automated optimization of photonic crystal slab cavities , 2014, Scientific Reports.

[19]  Lawrence D. Jackel,et al.  Handwritten Digit Recognition with a Back-Propagation Network , 1989, NIPS.

[20]  Steven G. Johnson,et al.  Formulation for scalable optimization of microcavities via the frequency-averaged local density of states. , 2013, Optics express.

[21]  Susumu Noda,et al.  Analysis of high-Q photonic crystal L3 nanocavities designed by visualization of the leaky components. , 2017, Optics express.

[22]  Susumu Noda,et al.  Analysis of the experimental Q factors (~ 1 million) of photonic crystal nanocavities. , 2006, Optics express.

[23]  Mitsuru Kitamura,et al.  Cross-Sectional Shape Optimization of Whispering-Gallery Ring Resonators , 2012, Journal of Lightwave Technology.

[24]  Yoshinori Tanaka,et al.  Design of Photonic Crystal Nanocavity , 2008 .

[25]  Nando de Freitas,et al.  Bayesian Optimization in a Billion Dimensions via Random Embeddings , 2013, J. Artif. Intell. Res..

[26]  Cheng Li,et al.  High Dimensional Bayesian Optimization with Elastic Gaussian Process , 2017, ICML.

[27]  土屋 一郎,et al.  Ultra-High-Q Photonic Double-Heterostructure Nanocavity , 2005 .

[28]  W. Cai,et al.  A Generative Model for Inverse Design of Metamaterials , 2018, Nano letters.

[29]  Ilya Fushman,et al.  General recipe for designing photonic crystal cavities. , 2005, Optics express.

[30]  Masaya Notomi,et al.  Ultrahigh-Q photonic crystal nanocavities realized by the local width modulation of a line defect , 2006 .

[31]  Yang Long,et al.  Inverse design of photonic topological state via machine learning , 2019, Applied Physics Letters.

[32]  Susumu Noda,et al.  Photonic Crystal Devices in Silicon Photonics , 2018, Proceedings of the IEEE.

[33]  Susumu Noda,et al.  Trapping and emission of photons by a single defect in a photonic bandgap structure , 2000, Nature.

[34]  Dario Gerace,et al.  Photonic crystal slab cavity simultaneously optimized for ultra-high Q/V and vertical radiation coupling , 2017 .

[35]  Steven G. Johnson,et al.  Cavity-enhanced second-harmonic generation via nonlinear-overlap optimization , 2015, 1505.02880.

[36]  T. Asano,et al.  Improvement in the quality factors for photonic crystal nanocavities via visualization of the leaky components. , 2016, Optics express.

[37]  Philippe Boucaud,et al.  Optimized design for 2 × 106 ultra-high Q silicon photonic crystal cavities , 2010 .

[38]  T. Asano,et al.  Optimization of photonic crystal nanocavities based on deep learning. , 2018, Optics express.

[39]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[40]  Yoshinori Tanaka,et al.  High-Q nanocavity with a 2-ns photon lifetime. , 2007, Optics express.

[41]  S. Noda,et al.  Design of Photonic Crystal Nanocavity With $Q$-Factor of ${{\sim}10^{9}}$ , 2008, Journal of Lightwave Technology.

[42]  Giulia Marcucci,et al.  Machine learning inverse problem for topological photonics , 2018, Communications Physics.

[43]  Boris Polyak.  Some methods of speeding up the convergence of iteration methods , 1964 .

[44]  Anders Krogh,et al.  A Simple Weight Decay Can Improve Generalization , 1991, NIPS.

[45]  T. Asano,et al.  High-Q photonic nanocavity in a two-dimensional photonic crystal , 2003, Nature.