Annonce en lien avec l’Action/le Réseau : aucun

Laboratoire/Entreprise : Institut Elie Cartan de Lorraine
Durée : 6 mois
Contact : marianne.clausel@univ-lorraine.fr
Date limite de publication : 2020-03-01

Contexte :
The analysis of multi-dimensional time series is a fundamental problem in most
areas of science and industry. Often, linear models are insufficient to capture the
structure present in data.
The internship shall focus on the improvement of machine learning techniques
for multivariate time series analysis based on specific feautures encoding dependencies between the components and known as the iterated-integrals signature
(IIS) [1]. Equipped with mathematical guarantees, the IIS is a means to extract
(almost all) multilinear features of a time series. The IIS can then be combined
with Kernel methods as in [2] to perform classical machine learning tasks as classification. We intend to extend classical kernel approaches for statistical testing
and change point detection with this new framework.

Sujet :
The analysis of multi-dimensional time series is a fundamental problem in most
areas of science and industry. Often, linear models are insufficient to capture the
structure present in data.
The internship shall focus on the improvement of machine learning techniques
for multivariate time series analysis based on specific feautures encoding dependencies between the components and known as the iterated-integrals signature
(IIS) [1]. Equipped with mathematical guarantees, the IIS is a means to extract
(almost all) multilinear features of a time series. The IIS can then be combined
with Kernel methods as in [2] to perform classical machine learning tasks as classification. We intend to extend classical kernel approaches for statistical testing
and change point detection with this new framework.
The internship will be divided into two parts : understanding of the IIS features
and the kernelized framework, and thereafter application to statistical testing.

Profil du candidat :
Master 2 students with good background in statistical learning, strong programming skills in Python

Formation et compétences requises :
Master 2

Adresse d’emploi :
Institut Élie Cartan de Lorraine
Université de Lorraine, Site de Nancy
B.P. 70239, F-54506 Vandoeuvre-lès-Nancy Cedex

Document attaché : stage-IECL-CRAN.pdf