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:2060@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20171208T140000
DTEND;TZID=Europe/Paris:20171208T150000
DTSTAMP:20171123T130000Z
URL:https://www.i2m.univ-amu.fr/evenements/new-theoretical-and-algorithmic
 -advances-in-machine-learning-with-optimal-transport/
SUMMARY: (...): New theoretical and algorithmic advances in machine learnin
 g with optimal transport
DESCRIPTION:: Optimal transportation problem is a powerful technique that h
 as recently found its application in various areas of machine learning inc
 luding\, for instance\, computer vision\, information retrieval and music 
 unmixing. In this talk\, I will introduce the basic concepts of optimal tr
 ansportation theory as well as some algorithmic ideas that were proposed i
 n machine learning based on it. These algorithmic ideas include a recently
  proposed unsupervised learning algorithm based on regularized optimal tra
 nsport and a new method for feature selection applied in the context of do
 main adaptation for prostate cancer mapping. From that point I will furthe
 r present some theoretical insights that were obtained based on the optima
 l transportation theory for domain adaptation in order to justify its use 
 in this context and to analyze the a priori success of adaptation represen
 ted by the existence of the joint hypothesis between source and target dom
 ains.https://sites.google.com/site/redkoievgen/
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20171029T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR