BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Making Sense of Data - A Research Agenda - Prof Bob Williamson (AN
 U and Scientific Director of NICTA)
DTSTART:20100311T110000Z
DTEND:20100311T120000Z
UID:TALK23355@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:The talk will comprise two parts. In the first (non-technical)
  part I will outline my long term research agenda which can be summed up a
 s "making machine learning into an engineering discipline". I will explain
  why i think it is not at present\, and what needs to be done in order to 
 make it so. A crucial part of this is to be able to describe machine learn
 ing _problems_ in a declarative way (as opposed to merely describing candi
 date _solutions_).\n\nIn the second (technical) part\, I will present an e
 xample of the sort of theory that I believe will help by showing how seemi
 ngly different problems are actually different perspectives of the same pr
 oblem. Specifically I will discuss connections between the problem of esti
 mating similarity of two distributions and binary classification problems.
LOCATION: Cambridge University Engineering Department\, Lecture Room 6
END:VEVENT
END:VCALENDAR