Dimensionality reduction for the on-chip integration of advanced photonic devices and functionalities

Daniele Melati, Mohsen Kamandar Dezfouli, Yuri Grinberg, Muhammad Al-Digeil, Dan-Xia Xu, Jens H. Schmid, Pavel Cheben, Abi Waqas, Paolo Manfredi, Jianhao Zhang, Laurent Vivien, Carlos Alonso-Ramos

Abstract: Design of modern photonic devices requires to handle a large number of parameters and figures of merit. By scaling down the complexity of the problem, machine learning dimensionality reduction enables the discovery of better performing devices, higher integration scale, and efficient evaluation of fabrication tolerances.