Scientific Machine Learning for PDEs
Course Description
This course covers modern computational methods for learning and solving partial differential equations (PDEs), combining classical numerical analysis with machine learning approaches such as Physics-Informed Neural Networks (PINNs). Students will study numerical aspects of variational formulations and gradient flows for solving PDEs, focusing on key properties such as stability, consistency, and convergence. The course will provide techniques for tackling challenging problems, such as nonlinear PDEs, discontinuous solutions, and inverse problems, using tools from variational calculus, functional analysis, and approximation theory.
The course is targeted at Master’s students in mathematics. Basic knowledge of Functional Analysis and Numerical Methods is recommended
Lecturer: Dr. Juan Esteban Suarez
Schedule and Venue
Monday 14:00 - 16:00 B047
Thursday 14:00 - 16:00 B133
Creditable Modules
Master of Science in Finanz- und Versicherungsmathematik: Modul WP23 “Advanced Topics in Computer and Data Science B”
Master of Science in Mathematik: Modul WP42 "Überblick über ein aktuelles Forschungsgebiet B”
Registration