Gradient-Informed Bayesian and Interior Point Optimization for Efficient Inverse Design in Nanophotonics

TL;DR

This paper introduces BONNI, a novel optimization method combining Bayesian optimization with neural network surrogates and interior point techniques, effectively improving inverse design efficiency in nanophotonics by avoiding local optima and reducing spectral error.

Contribution

The paper presents BONNI, a new gradient-informed Bayesian optimization framework that enhances inverse design in nanophotonics by integrating neural network surrogates with interior point methods.

Findings

BONNI achieved a 4.5% spectral error with 10-layer Bragg reflector, outperforming previous 7.8% with 16 layers.

BONNI effectively designed broadband waveguide taper and photonic crystal waveguide transition.

The method outperforms existing optimization algorithms in nanophotonic device design.

Abstract

Inverse design, particularly geometric shape optimization, provides a systematic approach for developing high-performance nanophotonic devices. While numerous optimization algorithms exist, previous global approaches exhibit slow convergence and conversely local search strategies frequently become trapped in local optima. To address the limitations inherent to both local and global approaches, we introduce BONNI: Bayesian optimization through neural network ensemble surrogates with interior point optimization. It augments global optimization with an efficient incorporation of gradient information to determine optimal sampling points. This capability allows BONNI to circumvent the local optima found in many nanophotonic applications, while capitalizing on the efficiency of gradient-based optimization. We demonstrate BONNI's capabilities in the design of a distributed Bragg reflector as well as a dual-layer grating coupler through an exhaustive comparison against other optimization algorithms commonly used in literature. Using BONNI, we were able to design a 10-layer distributed Bragg reflector with only 4.5% mean spectral error, compared to the previously reported results of 7.8% error with 16 layers. Further designs of a broadband waveguide taper and photonic crystal waveguide transition validate the capabilities of BONNI.