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:Isaac Newton Institute Seminar Series
SUMMARY:The geometry of optimal experiment design for vect
or-valued Ornstein-Uhlenbeck processes. - Mohamed-
Ali Belabbas (University of Illinois at Urbana-Cha
mpaign)
DTSTART;TZID=Europe/London:20161124T140000
DTEND;TZID=Europe/London:20161124T150000
UID:TALK69149AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/69149
DESCRIPTION:Ornstein-Uhlenbeck processes are commonly used as
models in engineering\, biology and finance. Consi
der the estimation of the state of such processes
from linear\, noisy measurements\; \;the Kalma
n filter is known to be the minimum mean square er
ror estimator when the measurement noise is Gaussi
an. We address here how to design the measurements
that minimize the error afforded by the Kalman fi
lter. This problem of optimal experiment design\,
which is almost as old as the Kalman filter itself
\, \;is however \;not convex. As a conseq
uence\, many ad hoc methods have been used over th
e years to solve it. We show in this talk how a ge
ometric approach allows us to characterize and obt
ain the optimal designs exactly. This optimal desi
gn yields the lowest possible estimation error fro
m linear measurements with a fixed signal to noise
ratio. \;

LOCATION:Seminar Room 2\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR