Project Grant 2435746

This $500,000 Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) will support the development of an innovative deep learning framework that combines physics-informed principles with scientific domain-adapted generative diffusion models. The goal is to overcome key challenges in scientific inverse design and accelerate scientific discovery, particularly in the field of nanophotonics. The research aims to establish the first model of generalizable, AI-assisted inverse design of complex nanophotonic systems through cross-disciplinary collaboration. The award will fund the creation of domain-specific generative diffusion models that efficiently capture photonic structure data priors with limited training data, as well as the integration of a novel physics-informed machine learning approach to ensure accurate predictions of physical properties and reliable conditional inverse design. The framework will be validated by inverse design of advanced topological photonic systems. The project also includes educational and outreach components to enhance engineering and physics curricula for undergraduate and graduate students, and engage high school students in Southeast Michigan.