Machine Learning for Future Fiber-Optic Communication Systems

Chapter Eleven: Machine learning for design and optimization of photonic devices

Keisuke Kojima^a; Toshiaki Koike-Akino^a; Yingheng Tang^a,b; Ye Wang^a; Matthew Brand^a    aMitsubishi Electric Research Laboratories (MERL), Cambridge, MA, United States^bElectrical and Computer Engineering Dept., Purdue University, West Lafeyette, IN, United States

Abstract

This chapter discusses how we can apply deep learning to photonic device design problems, by focusing on three types of modeling approaches. In forward modeling, the trained deep neural network (DNN) is integrated in the optimization loops of classical optimizers. In inverse modeling, the trained DNN directly constructs a structure with the desired target performance given as input. Generative modeling ...