BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Statistics
SUMMARY:Distance Shrinkage and Euclidean Embedding via Reg
ularized Kernel Estimation - Ming Yuan (U. of Wisc
onsin)
DTSTART;TZID=Europe/London:20160524T140000
DTEND;TZID=Europe/London:20160524T150000
UID:TALK65861AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/65861
DESCRIPTION:Although recovering an Euclidean distance matrix f
rom noisy observations is a common problem in prac
tice\, how well this could be done remains largely
unknown. To fill in this void\, we study a simple
distance matrix estimate based upon the so-called
regularized kernel estimate. We show that such an
estimate can be characterized as simply applying
a constant amount of shrinkage to all observed pai
rwise distances. This fact allows us to establish
risk bounds for the estimate implying that the tru
e distances can be estimated consistently in an av
erage sense as the number of objects increases. In
addition\, such a characterization suggests an ef
ficient algorithm to compute the distance matrix e
stimator\, as an alternative to the usual second o
rder cone programming known not to scale well for
large problems.
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberfo
rce Road\, Cambridge.
CONTACT:Quentin Berthet
END:VEVENT
END:VCALENDAR