ECE 228 Machine Learning for Physical Applications
Description: This course provides an introduction to physics-guided deep learning and its applications in physical systems and control. The course includes both the practical and theoretical aspects of the following topics: linear regression and classification, LASSO and ridge regressions, feedforward neural networks, recurrent neural networks, Transformers, physics-informed neural networks, neural operators, Neural ODEs, incorporating optimization layers into learning, Gaussian processes and uncertainty quantification. It involves a group-based course project that provides a hands-on opportunity in conducting machine learning projects for physical applications!
ECE 171B Linear Control System Theory
Description: This course provides an introduction to linear system control theory with applications in physical, biological, and engineering systems. t focuses on the fundamental principles of linear feedback control design in state space. Both practical and theoretical aspects are covered, including: review of ODEs and linear algebra; system modeling; equilibrium points and linearization; stability of linear time-invariant (LTI) systems; reachability, controllability, and stabilizability of LTI systems; state feedback controller design; observability and detectability; observer and output feedback design; linear quadratic regulator (LQR); Kalman filtering; and an introduction to Lyapunov stability analysis and reinforcement learning, with emphasis on their applications in linear control systems.