We develop and apply advanced artificial intelligence techniques to solve complex problems in electromagnetics and RF/microwave engineering. Our research focuses on integrating machine learning methodologies—including active learning, deep neural networks (DNN), diffusion models, and differentiable physics—into electromagnetic wave analysis. By replacing or augmenting conventional full-wave simulations (e.g., FDTD, RCWA), we design high-performance meta-devices, such as frequency selective rasorbers (FSR) for stealth radomes and 6G RF active metasurfaces.

• Key Topics: Active Learning, Differentiable Physics, Generative AI for Meta-structures, Surrogate Modeling for RF/Microwave.

We explore the intersection of artificial intelligence and quantum computing to overcome the limitations of current noisy intermediate-scale quantum (NISQ) devices and to pave the way for fault-tolerant quantum computing (FTQC). Our research designs physics-informed quantum encoders that directly map machine learning weights—such as those from Higher-Order Factorization Machines (HOFM)—to quantum circuits. We also utilize AI to mitigate stochastic quantum noise (T1/T2) in quantum sensing and develop quantum-centric hybrid compilers to achieve practical quantum advantage.

• Key Topics: Quantum-centric Hybrid Computing, Physics-informed Quantum Encoders, AI-assisted Quantum Sensing, Quantum Advantage.

We bridge the critical gap between computational simulation and physical fabrication. Relying solely on simulation data often leads to performance degradation in actual manufacturing. To resolve this, we employ Bayesian optimization and active learning to integrate experimental measurements directly with simulation models. This data-driven methodology significantly minimizes trial-and-error in fabrication. Our core applications span next-generation semiconductor manufacturing—including IGZO thin-film processes developed through collaborative research—as well as nanophotonic structures and SiOx battery electrodes.

• Key Topics: Semiconductor Manufacturing (IGZO), Active Learning, Bayesian Optimization, Fabrication-Aware Design, Process Optimization.