MS Teams
Speaker
Bamdad HosseiniÌý|ÌýÌýCalifornia Institute of Technology
Title
ÌýInverse Problems and Machine Learning
Abstract
The fields of inverse problems and machine learning have a lot in common. In fact, manyÌýof the algorithms and problems in each of these fields can be analyzed and developed fromÌýthe perspective of the other. In this talk I give an overview of some of my research at the intersectionÌýof these fields. In the first part I will discuss the asymptotic consistency of a graphical semi-supervisedÌýlearning algorithm. Using ideas from theory of inverse problems and spectral analysis of elliptic operatorsÌýto develop a deeper understandingÌýof how and why graphical algorithms work. In the second part of theÌýtalk, I will present an algorithm for data-driven solution of Bayesian inverse problems by combining toolsÌýfrom machine learning, such as generative adversarial networks, with ideas in measure transport. ThisÌýapproach leads to a model agnostic method for conditional sampling and in turn the solution of inverseÌýproblems.