Applied Math Seminar | Bamdad Hosseini, Inverse Problems and Machine Learning

Wednesday, January 27, 2021 12:00 pm - 12:00 pm EST (GMT -05:00)

MS Teams

Speaker

Bamdad Hosseini |  California Institute of Technology

Title

 Inverse Problems and Machine Learning

Abstract

The fields of inverse problems and machine learning have a lot in common. In fact, many of the algorithms and problems in each of these fields can be analyzed and developed from the perspective of the other. In this talk I give an overview of some of my research at the intersection of these fields. In the first part I will discuss the asymptotic consistency of a graphical semi-supervised learning algorithm. Using ideas from theory of inverse problems and spectral analysis of elliptic operators to develop a deeper understanding of how and why graphical algorithms work. In the second part of the talk, I will present an algorithm for data-driven solution of Bayesian inverse problems by combining tools from machine learning, such as generative adversarial networks, with ideas in measure transport. This approach leads to a model agnostic method for conditional sampling and in turn the solution of inverse problems.