Gen-Fab: A Variation-Aware Generative Model for Predicting Fabrication Variations in Nanophotonic Devices

TL;DR

Gen-Fab is a novel conditional GAN model that predicts fabrication variations in nanophotonic devices from design layouts, providing diverse high-resolution outcomes and uncertainty estimates, outperforming existing methods.

Contribution

We introduce Gen-Fab, a variation-aware generative model that captures fabrication uncertainties and predicts diverse outcomes from design layouts, advancing digital twin capabilities for nanophotonics.

Findings

Gen-Fab achieves an IoU of 89.8%, surpassing baselines.

It better models distributional uncertainty in fabrication outcomes.

Gen-Fab generalizes well to unseen geometries.

Abstract

Silicon photonic devices often exhibit fabrication-induced variations such as over-etching, underetching, and corner rounding, which can significantly alter device performance. These variations are non-uniform and are influenced by feature size and shape. Accurate digital twins are therefore needed to predict the range of possible fabricated outcomes for a given design. In this paper, we introduce Gen-Fab, a conditional generative adversarial network (cGAN) based on Pix2Pix to predict and model uncertainty in photonic fabrication outcomes. The proposed method takes a design layout (in GDS format) as input and produces diverse high-resolution predictions similar to scanning electron microscope (SEM) images of fabricated devices, capturing the range of process variations at the nanometer scale. To enable one-to-many mapping, we inject a latent noise vector at the model bottleneck. We compare Gen-Fab against three baselines: (1) a deterministic U-Net predictor, (2) an inference-time Monte Carlo Dropout U-Net, and (3) an ensemble of varied U-Nets. Evaluations on an out-of-distribution dataset of fabricated photonic test structures demonstrate that Gen-Fab outperforms all baselines in both accuracy and uncertainty modeling. An additional distribution shift analysis further confirms its strong generalization to unseen fabrication geometries. Gen-Fab achieves the highest intersection-over-union (IoU) score of 89.8%, outperforming the deterministic U-Net (85.3%), the MC-Dropout U-Net (83.4%), and varying U-Nets (85.8%). It also better aligns with the distribution of real fabrication outcomes, achieving lower Kullback-Leibler divergence and Wasserstein distance.