Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures

We demonstrate a tandem neural network architecture that tolerates inconsistent training instances in inverse design of nanophotonic devices. It provides a way to train large neural networks for the inverse design of complex photonic structures. © 2019 The Author(s)

[1]  Li Jing,et al.  Nanophotonic Inverse Design Using Artificial Neural Network , 2017 .

[2]  A. Levi,et al.  Optimization of aperiodic dielectric structures , 2006 .

[3]  N. Yu,et al.  Flat optics with designer metasurfaces. , 2014, Nature materials.

[4]  Michal Lipson,et al.  Low modal volume dipole-like dielectric slab resonator. , 2008, Optics express.

[5]  B. Shen,et al.  An integrated-nanophotonics polarization beamsplitter with 2.4 × 2.4 μm2 footprint , 2015, Nature Photonics.

[6]  Joni Dambre,et al.  Trainable hardware for dynamical computing using error backpropagation through physical media , 2014, Nature Communications.

[7]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[8]  John J. Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities , 1999 .

[9]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[10]  S. Prasad,et al.  Reverse modeling of microwave circuits with bidirectional neural network models , 1998 .

[11]  Guoxing Zheng,et al.  Metasurface holograms reaching 80% efficiency. , 2015, Nature nanotechnology.

[12]  Giuseppe Pelosi,et al.  Neural network applications in microwave device design , 2002 .

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

[14]  D Psaltis,et al.  Optical implementation of the Hopfield model. , 1985, Applied optics.

[15]  Alexander Y. Piggott,et al.  Inverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer , 2015, Nature Photonics.

[16]  S. Osher,et al.  Maximizing band gaps in two-dimensional photonic crystals by using level set methods , 2005 .

[17]  Shanhui Fan,et al.  S4 : A free electromagnetic solver for layered periodic structures , 2012, Comput. Phys. Commun..

[18]  Qi-Jun Zhang,et al.  Neural Network Inverse Modeling and Applications to Microwave Filter Design , 2008, IEEE Transactions on Microwave Theory and Techniques.

[19]  Steven G. Johnson,et al.  Robust optimization of adiabatic tapers for coupling to slow-light photonic-crystal waveguides. , 2012, Optics express.