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:Microsoft Research Cambridge\, public talks
SUMMARY:Politics\, Preferences and Permutations: Probabili
stic Reasoning with Rankings - Jonathan Huang
DTSTART;TZID=Europe/London:20110414T093000
DTEND;TZID=Europe/London:20110414T103000
UID:TALK30769AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/30769
DESCRIPTION:Permutations arise fundamentally in a plethora of
real world applications from multi-person tracking
to preference ranking and election analysis. Rea
l world data\, often being noisy and incomplete\,
necessitates a probabilistic approach to learning
and reasoning with permutations. However\, repres
enting arbitrary probability distributions over th
e space of permutations has been notoriously intra
ctable due to the factorial number of permutations
.\n\nIn this talk\, I will present methods for eff
iciently representing and reasoning with such dist
ributions.\nThe main idea that I set forth is that
distributions over permutations can be decomposed
additively or multiplicatively into a series of s
impler functions which can be dealt with more easi
ly. As I show\, additive decompositions turn out
to correspond to generalized Fourier analysis on t
he symmetric groups\, while multiplicative decompo
sitions correspond to a generalized notion of prob
abilistic independence.\nAlong the way\, I will di
scuss applications of these methods for statistica
lly analyzing political elections in Ireland\, pre
ference surveys for sushi\, as well as for perform
ing multi-person tracking using a networked array
of cameras.\n\nBiography:\nJonathan Huang is a Ph.
D. candidate in the School of Computer Science at
Carnegie Mellon University where he also received
a Masters degree in 2008.\nHe received his B.S. de
gree in Mathematics from Stanford University in 20
05. His research interests lie primarily in stati
stical machine learning and for his dissertation\,
he has developed efficient statistical techniques
for modeling and performing inference with combin
atorial objects such as permutations and rankings.
\nHis research has resulted in a number of publica
tions in premier machine learning conferences and
journals\, receiving a paper award in NIPS 2007 fo
r his work on applying group theoretic Fourier ana
lysis to probabilistic reasoning with permutations
.
LOCATION:Small lecture theatre\, Microsoft Research Ltd\, 7
J J Thomson Avenue (Off Madingley Road)\, Cambrid
ge
CONTACT:Microsoft Research Cambridge Talks Admins
END:VEVENT
END:VCALENDAR