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:2786@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20190308T140000
DTEND;TZID=Europe/Paris:20190308T150000
DTSTAMP:20190221T130000Z
URL:https://www.i2m.univ-amu.fr/evenements/optimal-transport-for-structure
 d-data/
SUMMARY: (...): Optimal transport for structured data
DESCRIPTION:: In this work\, we consider the problem of computing distances
  between structured objects such as undirected graphs\, seen as probabilit
 y distributions in a specific metric space. We consider a new transportati
 on distance (i.e. which minimizes a total cost of transporting probability
  masses) that unveils the geometric nature of the structured objects space
 . After introducing Wasserstein and Gromov-Wasserstein metrics that focus 
 solely and respectively on features (by considering a metric in the featur
 e space) or structure (by seeing structure as a metric space)\, we will pr
 esent our new distance which exploits jointly both information\, and conse
 quently being called Fused Gromov-Wasserstein (FGW). We will discuss its p
 roperties and computational aspects\, we show results on a graph classific
 ation task\, where our method outperforms both graph kernels and deep grap
 h convolutional networks. Exploiting further on the metric properties of F
 GW\, interesting geometric objects such as Fréchet means or barycenters o
 f graphs are illustrated and discussed in a clustering context.https://tva
 yer.github.io
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20181028T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR