BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Scaling Machine Learning for the Internet - Prof. Alexander Smola 
 (Yahoo!)
DTSTART:20111207T110000Z
DTEND:20111207T120000Z
UID:TALK34614@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:In this talk I will give an overview over an array of highly s
 calable techniques for both observed and latent variable models. This make
 s them well suited for problems such as classification\, recommendation sy
 stems\, topic modeling and user profiling. I will present algorithms for b
 atch and online distributed convex optimization to deal with large amounts
  of data\, and hashing to address the issue of parameter storage for perso
 nalization and collaborative filtering. Furthermore\, to deal with latent 
 variable models I will discuss distributed sampling algorithms capable of 
 dealing with tens of billions of latent variables on a cluster of 1000 mac
 hines.\n\nThe algorithms described are used for personalization\, spam fil
 tering\, recommendation\, document analysis\, and advertising.\n
LOCATION: Cambridge University Engineering Department\, Lecture Room 4
END:VEVENT
END:VCALENDAR