Abstract
Many existing well-known multilayer design methods are based on so-called greedy algorithms. New deep search algorithms developed for needle optimization, gradual evolution, and design cleaner methods are presented. The algorithms possess machine learning features. The advantages of the deep search methods are demonstrated on a set of examples including the OIC Design Contest 2019.
© 2019 Optical Society of America
Full Article | PDF ArticleMore Like This
J. A. Dobrowolski and R. A. Kemp
Appl. Opt. 31(31) 6747-6756 (1992)
Jennifer D. T. Kruschwitz, Vladimir Pervak, Jason Keck, Eric Baltz, Amit Deliwala, Zachary Gerig, Fabien Lemarchand, Samuel Pellicori, Kageyuki Sato, William Southwell, Michael Trubetskov, Kunihiko Yano, and Wenjia Yuan
Appl. Opt. 59(5) A206-A212 (2020)
Jennifer D. T. Kruschwitz, Michael Trubetskov, and Jason Keck
Appl. Opt. 62(7) B43-B54 (2023)