Quantized Inverse Design for Photonic Integrated Circuits

TL;DR

This paper introduces a memory-efficient inverse design framework for photonic integrated circuits that leverages straight-through gradient estimation to handle complex constraints and non-differentiable shape parametrizations, enabling the creation of advanced 3D structures.

Contribution

It presents a novel reverse-mode automatic differentiation method tailored for FDTD simulations, improving design flexibility and efficiency for complex PIC structures.

Findings

Successfully designed complex 3D PIC structures

Demonstrated handling of non-differentiable shape constraints

Showed potential for practical PIC applications

Abstract

The inverse design of photonic integrated circuits (PICs) presents distinctive computational challenges, including their large memory requirements. Advancements in the two-photon polymerization (2PP) fabrication process introduce additional complexity, necessitating the development of more flexible optimization algorithms to enable the creation of multi-material 3D structures with unique properties. This paper presents a memory efficient reverse-mode automatic differentiation framework for finite-difference time-domain (FDTD) simulations that is able to handle complex constraints arising from novel fabrication methods. Our method is based on straight-through gradient estimation that enables non-differentiable shape parametrizations. We demonstrate the effectiveness of our approach by creating increasingly complex structures to solve the coupling problem in PICs. The results highlight the potential of our method for future PIC design and practical applications.