Learning Controlled Stochastic Differential Equations

Explore the mathematical foundations and applications of controlled stochastic differential equations in this seminar presented by Dr. Luc Brogat-Motte from the Istituto Italiano di Tecnologica. Delve into the theoretical framework for learning and modeling systems governed by stochastic processes with control mechanisms, examining how uncertainty and randomness can be incorporated into differential equation models. Discover the computational methods and statistical approaches used to estimate parameters and control strategies in stochastic systems, with particular emphasis on machine learning techniques for handling prediction uncertainty. Learn about the practical applications of controlled SDEs in various fields including robotics, finance, and engineering systems where both random fluctuations and control inputs play crucial roles. Gain insights into the latest research developments in representing, calibrating, and leveraging prediction uncertainty in mathematical models that bridge statistics and machine learning methodologies.