BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:862@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20150924T140000
DTEND;TZID=Europe/Paris:20150924T150000
DTSTAMP:20240524T072505Z
URL:https://www.i2m.univ-amu.fr/evenements/soutenance-de-these-de-s-takerk
 art-lif-int-a-multi-source-perspective-on-inter-subject-learning-contribut
 ions-to-neuroimaging-2/
SUMMARY: (...): Soutenance de thèse de S. Takerkart (LIF\, INT) : A multi-
 source perspective on inter-subject learning. Contributions to neuroimagin
 g.
DESCRIPTION:: Soutenance de thèse de Sylvain Takerkart\n\nmulti-source per
 spective on inter-subject learning. Contributions to neuroimaging.\n\nSall
 e Henri Gastaut\nCampus santé Timone\, boulevard Jean Moulin\n13385 Marse
 ille cedex 5\n\nJury:\n- Pr. Rainer Goebel (Maastricht University\, The Ne
 therlands)\, rapporteur\n- Pr. Patrick Gallinari (Université Pierre et Ma
 rie Curie\, Paris)\, rapporteur\n- Dr. Jean-François Mangin (CEA\, Saclay
 )\, examinateur\n- Dr. Bertrand Thirion (INRIA\, Saclay)\, examinateur\n- 
 Dr. Olivier Coulon (CNRS\, Marseille)\, co-directeur de thèse\n- Pr. Liva
  Ralaivola (Aix-Marseille Université\, Marseille)\, directeur de thèse\n
 \nAbstract: Inter-subject learning consists in giving predictions on data 
 from a subject not present in the training database\, as with computer-aid
 ed diagnosis where the computer has to guess wether an unknown individual 
 is healthy or sick. In this thesis\, we argue that inter-subject learning 
 should be handled in the multi-source framework where each subject is a di
 fferent source of data. We then introduce three original contributions for
  neuroimaging applications. The first one is a method for inter-subject pr
 edictions of fMRI data. Because of the inter-subject variability\, the ori
 ginal feature spaces are all different. Using graphs and a graph kernel\, 
 the input patterns are implicitly projected into a common reproducing kern
 el hilbert space. We show the effectiveness of this method on tonotopy dat
 a recorded in the auditory cortex. The second one is a cortical morphometr
 y method. We design graphs from the deepest points of cortical sulci\, and
  we project them into a common space using a graph kernel. A spatial infer
 ence method is then proposed to perform the detection of cortical zones wh
 ere populations are different. Using this method\, we study cortical asymm
 etries and gender differences.nnThe third contribution of this thesis is a
  multi-source domain adaptation technique. Our method is an extension of t
 he kernel mean matching for the multi-source case. We present preliminary 
 results on a inter-subject prediction task used to analyse data from a mag
 neto-encephalography experiment.\n\n&nbsp\;
CATEGORIES:Séminaire
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20150329T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR