Efficient nanophotonic devices optimization using deep neural network trained with physics-based transfer learning (PBTL) methodology

TL;DR

This paper introduces a physics-based transfer learning framework combined with neural network surrogate models and genetic algorithms to efficiently optimize complex photonic devices like quantum cascade lasers, significantly reducing computational costs.

Contribution

It presents a novel PBTL-enhanced surrogate modeling approach that improves generalization and reduces data needs in photonic device optimization, enabling faster and more accurate design processes.

Findings

Achieved over 80,000 times speed-up in optimization process.

Improved prediction accuracy by 0.69%, reducing training data needs by 50%.

Enhanced device feasibility and evaluation metrics by 60%.

Abstract

We propose a neural network(NN)-based surrogate modeling framework for photonic device optimization, especially in domains with imbalanced feature importance and high data generation costs. Our framework, which comprises physics-based transfer learning (PBTL)-enhanced surrogate modeling and scalarized multi-objective genetic algorithms (GAs), offers a generalizable solution for photonic design automation with minimal data resources.To validate the framework, we optimize mid-infrared quantum cascade laser (QCL) structures consisting of two regions, active and injection, which have different levels of feature importance. The optimization targets include five key QCL performance metrics such as modal gain, emission wavelength, linewidth, and effective injection, extraction energies. To address the challenge of multiple local optima in the output latent space, we integrate a deep neural network total predictor (DNN-TP) with a GA, enabling scalable and nature-inspired optimization. By replacing computationally expensive numerical simulations with the DNN-TP surrogate model, the optimization achieves a speed-up of over 80,000 times, allowing large-scale exploration of the QCL design space.To improve model generalization with limited data, we introduce PBTL, which transfers knowledge from a DNN core predictor (DNN-CP) trained on active-region structures. This approach yields a 0.69 percentage increase in prediction accuracy, equivalent to a 50 percentage reduction in training data requirements, and leads to generate more feasible device structure with 60 percentage improvement in evaluation metric during optimization.