Le laboratoire SAMOVAR (UMR 5157) organise sa journée doctorants le jeudi 5 octobre 2017 à partir de 14h au bâtiment Étoile. A cette occasion nous organisons un prix des meilleurs posters : un poster sera récompensé par un jury composé de chercheurs du laboratoire et un autre poster sera distingué par les votes du public.

Programme et titres des présentations

Le programme :

Vous êtes cordialement invités à participer à cette journée.

Pour des raisons logistiques, nous vous remercions de bien vouloir vous inscrire à l’adresse suivante

https://form.jotformeu.com/51536018441348

L’inscription est gratuite mais obligatoire.

Nous aurons le plaisir d’accueillir Bertrand Thirion (équipe projet Parietal, INRIA Saclay) qui fera un exposé invité sur les problématiques de machine learning.

Title : Machine learning: from generic to specific machine learning tools.




Abstract
Besides the deep learning tsunami, data analysts rely on a daily basis on simple machine learning operations: linear of tree-based classification, clustering, data decompositions. The scikit-learn Python library makes these available through an homogeneous API, in an open-source setting. In this talk, I will give an overview of this library, and discuss the features that make it a high-quality tool. I will conclude by a quick description of some specific developments that share some concepts, yet are well-suited to specific needs, such as high-dimensional small-sample problems encountered in brain imaging.




Biographie:
Bertrand Thirion is the principal investigator of the Parietal team (INRIA Saclay-Île-de-France) situated within the Neurospin research center (CEA, DSV, I2BM) at Saclay, France. After graduating from Ecole Polytechnique (1998), he got specialized in applied mathematics, with applications to computer vision. He did his PhD with the Odyssée team (Sophia-Antipolis, France) on the statistical analysis of functional brain images, under the direction of Olivier Faugeras (2000 2003). His post doc at SHFJ, Orsay (2003-2006) made him an active contributor of image analysis methods for between-subject comparison; since then he has been involved in neuroimaging platforms (SHFJ, then Neurospin). B. Thirion has been part of INRIA Saclay since 2006. His main research interests are related to the use of machine learning and statistical analysis techniques for neuroimaging, e.g. the modeling of between brain variability in group studies, the mathematical study of functional connectivity and brain activity decoding; he addresses various applications such as the study of vision through neuroimaging, the classification of brain images for diagnosis or brain mapping and the study of correlations between neuroimaging and genetics information. He has co-authored about 70 papers in neuroimaging, medical image analysis and machine learning journals or conferences. He is part of the scientific board of the French neuroscience institute.