Instead of using FDTD simulations for all the inverse design steps, we proposed to use neural network-based fitting to estimate the output of the FDTD simulations, and improve the design. We observed clear acceleration in the improvement of metric.

© 2017 Optical Society of America

PDF Article


You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
Login to access OSA Member Subscription