

Abstract for the talk on 11.07.2018 (15:30 h)
Mathematics of Data SeminarHarald Oberhauser (University of Oxford)
Learning laws of stochastic processes
The sequence of moments characterizes the law of (sufficiently nice) random variables in finite-dimensional spaces. I will talk about an analogous result for path-valued random variables, that is stochastic processes, and develop applications in statistical learning. Our starting point is the so-called signature of a path, that identifies a path as an element in a tensor algebra. (Joint work with Ilya Chevyrev).