Séance Séminaire

Séminaire de Probabilités et Statistique

Learning with signatures

Modern applications of artificial intelligence lead to high-dimensional multivariate temporal data that pose many challenges. Through a geometric approach to data flows, the notion of signature, a representation of a process as an infinite vector of its iterated integrals, is a promising tool. Its properties, developed in the context of rough path theory, make it a good candidate to play the role of features, then injected in learning algorithms. If the definition of signatures goes back to the work of Chen (1960), its use in machine learning is recent, and many questions remain to be explored. In this talk, we will give a quick overview of the signature definition and properties, continue with a discussion on how it can be used in a learning context, and conclude by focusing on one specific problem: learning the dynamics of a non-parametric system linking a target and a feature time series.

Séminaire en salle 109, également retransmis sur zoom : https://umontpellier-fr.zoom.us/j/94087408185