Abstract
Conventional neural network surrogate solvers are severely limited by their inability to function outside of fixed simulation parameters. We present a foundational method for conditioning on arbitrary parameters and demonstrate optimizing high-efficiency aperiodic superpixel deflectors.
© 2023 The Author(s)
PDF Article | Presentation VideoMore Like This
Akira Ueno, Hung-I Lin, Fan Yang, Sensong An, Louis Martin-Monier, Mikhail Y. Shalanginov, Tian Gu, and Juejun Hu
FTu1E.3 Frontiers in Optics (FiO) 2023
Shiyu Li, Ho-Chun Lin, and Chia Wei Hsu
FW4H.5 CLEO: Fundamental Science (CLEO:FS) 2023
Julie Buquet, Jean-François Lalonde, and Simon Thibault
FM6C.6 Frontiers in Optics (FiO) 2023