Menu Close

Inverse design of photonic integrated devices: Optimization and machine learning

Diagram of a fully connected neural network with an input layer, two hidden layers and an output layer

A tutorial on inverse design for photonic integrated devices, covering the fundamental concepts behind both iterative optimization and machine-learning approaches. Using the design of an integrated mode exchanger as a running example, it compares the strengths and limitations of the different techniques and discusses how to choose the right design tool.

Read the paper: APL Photonics 10, 101101 (2025), doi: 10.1063/5.0283179