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:Optimal Link Prediction with Matrix Logistic Regre
ssion - Quentin Berthet (University of Cambridge)
DTSTART;TZID=Europe/London:20180116T140000
DTEND;TZID=Europe/London:20180116T144500
UID:TALK97732AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/97732
DESCRIPTION:We consider the problem of link prediction\, based
on partial observation of a large network and on
covariates associated to its vertices. The generat
ive model is formulated as matrix logistic regress
ion. The performance of the model is analysed in a
high-dimensional regime under structural assumpti
on. The minimax rate for the Frobenius norm risk i
s established and a combinatorial estimator based
on the penalised maximum likelihood approach is sh
own to achieve it. Furthermore\, it is shown that
this rate cannot be attained by any algorithm comp
utable in polynomial time\, under a computational
complexity assumption. (Joint work with Nicolai Ba
ldin)
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR