The Computational Modeling and Analysis of Systems (CMAS) Lab focuses on the development of advanced methods in electronic design automation (EDA). Our research centers around building efficient tools and models that help designers analyze, prototype, and optimize circuits and devices without relying on expensive and time-consuming fabrication cycles.

We work across multiple abstraction layers, from the physics of nanoscale devices to the behavior of large electronic systems by applying principles from computational science, signal integrity, and statistical modeling. A core area of emphasis is scientific machine learning, particularly physics-informed neural networks and neural operator-based models, which we use to simulate and understand high-frequency analog and RF circuits under uncertainty.

Our current efforts include uncertainty quantification, stochastic circuit modeling, CAD for spintronics and photonics, and machine learning for device-circuit co-design. Whether it’s analyzing variability in silicon-photonic systems or developing compact models for emerging FET architectures, we aim to push the boundaries of how computation and learning can transform electronic system design.