BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:"Optimal Transport for Machine Learning" by Dr. Gabriel Peyré
DTSTART:20190212T160000
DTEND:20190212T170000
DTSTAMP:20260407T145519Z
UID:b9ae2b73e9b95566b8fa41e5e32b971de81a89dfe7a66daa57896ddb
CATEGORIES:Conferences - Seminars
DESCRIPTION:Gabriel Peyré (Ecole Normale Supérieure)\nAbstract\nOptimal 
 transport (OT) has become a fundamental mathematical tool at the interface
  between calculus of variations\, partial differential equations and proba
 bility. It took however much more time for this notion to become mainstrea
 m in numerical applications. This situation is in large part due to the hi
 gh computational cost of the underlying optimization problems. There is a 
 recent wave of activity on the use of OT-related methods in fields as dive
 rse as image processing\, computer vision\, computer graphics\, statistica
 l inference\, machine learning.\n\nIn this talk\, I will review an emergin
 g class of numerical approaches for the approximate resolution of OT-based
  optimization problems. This offers a new perspective for the application 
 of OT in high dimension\, to solve supervised (learning with transportatio
 n loss function) and unsupervised (generative network training) machine le
 arning problems.\n\nMore information and references can be found on the we
 bsite of our book "Computational Optimal Transport".
LOCATION:Idiap\, Martigny\, room 106 http://goo.gl/maps/9uhoa
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR