BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Polynomial Learning of Distribution Families - Kaushik Sinha
DTSTART:20101123T100000Z
DTEND:20101123T110000Z
UID:TALK28020@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:The study of Gaussian mixture distributions goes back to the l
 ate 19th\ncentury\, when Pearson introduced the method of moments to analy
 ze the\nstatistics of a crab population. They have since become one of the
  most\npopular tools of modeling and data analysis\, extensively used in s
 peech\nrecognition\, computer vision and other fields. Yet their propertie
 s are\nstill not well understood. \n\nIn my talk I will discuss some theor
 etical aspects of the problem of\nlearning Gaussian mixtures. In particula
 r\, I will discuss our recent\nresult with Mikhail Belkin\, which\, in a c
 ertain sense\, completes work on an\nactive recent topic in theoretical co
 mputer science by establishing quite\ngeneral conditions for polynomial le
 arnability of mixture distributions.\n
LOCATION:Small lecture theatre\, Microsoft Research Ltd\, 7 J J Thomson Av
 enue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR