AE 6531: Scientific Deep Learning for Engineers and Scientists

This course introduces theoretical and practical aspects of deep learning, emphasizing applications in areas such as fluid dynamics, heat transfer, and solid mechanics. Topics covered include: foundational deep learning techniques, specialized loss functions, regularization and weight initialization strategies, advanced deep learning algorithms, physics-informed deep learning (PIDL), domain decomposition-based PIDL methods, and various performance enhancement techniques. Students will engage in multiple hands-on coding projects, utilizing prominent machine learning libraries, to solve governing partial differential equations. Students cannot receive credit for this course if they have taken AE 5093 “Special Topics: Scientific Deep Learning for Engineers”.