Enhancing Adjoint Optimization-based Photonics Inverse Design with Explainable Machine Learning

A fundamental challenge in the design of photonic devices and electromagnetic structures more generally is the optimization of their overall architecture to achieve a desired response. To this end, topology or shape optimizers based on the adjoint variables method have been widely adopted due to their high computational efficiency and ability to create complex freeform geometries. However, the functional understanding of such freeform structures remains black box. Moreover, unless a design space of high-performance devices is known in advance, such gradient-based optimizers can get trapped in local minima, limiting performance achievable through this inverse design process. To elucidate the relationships between device performance and nanoscale structuring, while mitigating the effects of local minima trapping, we present an inverse design framework that combines adjoint optimization, automated machine learning (AutoML), and explainable artificial intelligence (XAI). Integrated with a numerical electromagnetics simulation method, our framework reveals structural contributions towards a figure-of-merit (FOM) of interest, then leverages this information to minimize the FOM further than that obtained through adjoint optimization alone, overcoming local minima. We demonstrate our framework in the context of waveguide design and achieve between 43% and 74% increases in device performance relative to state-of-the-art adjoint optimization-based inverse design across a range of telecom wavelengths. Results of this work thus highlight machine learning strategies that can substantially extend and improve the capabilities of a conventional, optimization-based inverse design algorithm while revealing deeper insights into the algorithm's designs.