BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:New Advances in Face Recognition - Tadas Baltrusaitis (University 
 of Cambridge)
DTSTART:20140509T130000Z
DTEND:20140509T140000Z
UID:TALK52334@talks.cam.ac.uk
CONTACT:Tadas Baltrusaitis
DESCRIPTION:Machine learning technologies are entering our lives at a grea
 t pace. A\ngrowing number of universities and businesses are employing mac
 hine learning\nto problems such as speech recognition\, visual object clas
 sification\, and\nface recognition to name a few.\n\nDeep learning forms a
  particularly successful set of machine learning\nalgorithms that can solv
 e previously intractable problems with a high degree\nof accuracy\, outper
 forming most previous approaches. In this lecture I will\nprovide a summar
 y of deep learning algorithms with a particular focus on\nsupervised learn
 ing. I will illustrate the algorithms with a real-world case\nstudy of Fac
 ebook's DeepFace system that reaches human level performance in\nthe task 
 of face recognition.  By the end of the lecture you will understand\nthe s
 teps needed to train and use such algorithms.
LOCATION:LT2\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR