A topological encoding method for data-driven photonics inverse design

TL;DR

This paper introduces a topological encoding technique that converts binary images of photonic structures into a continuous sparse form, facilitating data-driven inverse design and optimization of photonic devices.

Contribution

The novel encoding method enables effective dimensionality reduction and dataset generation for topology optimization in photonics using machine learning.

Findings

Successfully applied to design 2D non-paraxial diffractive optical elements

Enhanced accuracy and global optimization in inverse design tasks

Facilitated dataset creation for machine learning in photonics

Abstract

Data-driven approaches have been proposed as effective strategies for the inverse design and optimization of photonic structures in recent years. In order to assist data-driven methods for the design of topology of photonic devices, we propose a topological encoding method that transforms photonic structures represented by binary images to a continuous sparse representation. This sparse representation can be utilized for dimensionality reduction and dataset generation, enabling effective analysis and optimization of photonic topologies with data-driven approaches. As a proof of principle, we leverage our encoding method for the design of two dimensional non-paraxial diffractive optical elements with various diffraction intensity distributions. We proved that our encoding method is able to assist machine-learning-based inverse design approach for accurate and global optimization.