MAPS: Multi-Fidelity AI-Augmented Photonic Simulation and Inverse Design Infrastructure

TL;DR

MAPS tackles the bottlenecks in AI-assisted photonic inverse design, including data scarcity, computational inefficiency, manufacturability, and lack of standardized benchmarks. It presents a cohesive, open-source infrastructure consisting of MAPS-Data for multi-fidelity, richly labeled datasets; MAPS-Train for flexible, physics- and data-driven training with standardized evaluation; and MAPS-InvDes for fabrication-aware adjoint inverse design that can incorporate pre-trained neural solvers. The framework enables universal training, fair benchmarking, and practical inverse-design workflows across diverse photonic devices, with case studies showing improved data diversity, gradient fidelity, and design speedups. Collectively, MAPS provides a scalable platform to accelerate AI-enabled photonic hardware development and collaboration across disciplines, paving the way for more robust and manufacturable PICs.

Abstract

Inverse design has emerged as a transformative approach for photonic device optimization, enabling the exploration of high-dimensional, non-intuitive design spaces to create ultra-compact devices and advance photonic integrated circuits (PICs) in computing and interconnects. However, practical challenges, such as suboptimal device performance, limited manufacturability, high sensitivity to variations, computational inefficiency, and lack of interpretability, have hindered its adoption in commercial hardware. Recent advancements in AI-assisted photonic simulation and design offer transformative potential, accelerating simulations and design generation by orders of magnitude over traditional numerical methods. Despite these breakthroughs, the lack of an open-source, standardized infrastructure and evaluation benchmark limits accessibility and cross-disciplinary collaboration. To address this, we introduce MAPS, a multi-fidelity AI-augmented photonic simulation and inverse design infrastructure designed to bridge this gap. MAPS features three synergistic components: (1) MAPS-Data: A dataset acquisition framework for generating multi-fidelity, richly labeled devices, providing high-quality data for AI-for-optics research. (2) MAPS-Train: A flexible AI-for-photonics training framework offering a hierarchical data loading pipeline, customizable model construction, support for data- and physics-driven losses, and comprehensive evaluations. (3) MAPS-InvDes: An advanced adjoint inverse design toolkit that abstracts complex physics but exposes flexible optimization steps, integrates pre-trained AI models, and incorporates fabrication variation models. This infrastructure MAPS provides a unified, open-source platform for developing, benchmarking, and advancing AI-assisted photonic design workflows, accelerating innovation in photonic hardware optimization and scientific machine learning.

Figures (6)

• Figure 1: Illustration of MAPS infrastructure based on three synergistic sub-modules: MAPS-Data, MAPS-InvDes, and MAPS-Train, targeting AI-assisted photonic simulation and inverse design.
• Figure 2: MAPS-Data framework with various devices and sampling strategies.
• Figure 3: MAPS-Train framework with standardized input, flexible neural network model definition, and diverse data-driven/physics-driven loss function calculation.
• Figure 4: MAPS-InvDes framework with full support of simulation and adjoint-based inverse design.
• Figure 5: Comparison of sampling strategies: (a) Transmission ratio histogram for different strategies; (b) t-SNE showing separate distributions of low- and high-performance patterns, with perturbed opt-traj sampling covering both.