Math 76.02 Computational Inverse Problems

Estimating unknown variables (or states) of interest from measurement or observation data is called an inverse problem. In this course, you will learn computational tools to solve inverse problems. Among various tools, we will focus on the Bayesian approach (a class of probability-based methods). The course aims to tackle challenges in i) PDE-constrained optimization, ii) Data assimilation, iii) Image reconstruction, iv) motion learning problems. Students are required to be confident with at least one of the following skills

Probability and Statistics (Math 40, 50, 60, 70)
Linear Algebra (Math 22 or 24)
Differential Equations (Math 23, 53)
Time series analysis (Math 106)
Programming in Python (CS 1, 10; we will use OOP)

Students will work as a team (there will be two or three teams) based on their expertises. Other skills you will learn in this course include

Shell programming
Git
Numerical PDEs (FEM)
Stochastic process