Data-driven design of metasurface systems (Rising Researcher) (Conference Presentation)

Jonathan A. Fan

doi:10.1117/12.2557973

27 April 2020 Data-driven design of metasurface systems (Rising Researcher) (Conference Presentation)

Jonathan A. Fan

Proceedings Volume 11389, Micro- and Nanotechnology Sensors, Systems, and Applications XII; 113890V (2020) https://doi.org/10.1117/12.2557973
Event: SPIE Defense + Commercial Sensing, 2020, Online Only

Abstract

I will discuss the development and utility of the application of Global Topology Optimization Networks (GLOnets) to metasurface design. GLOnets are generative neural networks that can generate a distribution of metasurface designs from random noise inputs, and they train by using adjoint variables calculations as terms for backpropagation. With metagratings operating across a range of wavelengths and angles as a model system, we show that devices produced from the trained generative network have efficiencies comparable to the best devices produced by gradient-based topology optimization. Our reframing of adjoint-based optimization to the training of a generative neural network applies generally to physical systems that support performance improvements by gradient descent.

Conference Presentation

Citation Download Citation

Jonathan A. Fan "Data-driven design of metasurface systems (Rising Researcher) (Conference Presentation)", Proc. SPIE 11389, Micro- and Nanotechnology Sensors, Systems, and Applications XII, 113890V (27 April 2020); https://doi.org/10.1117/12.2557973

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available