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:Graphons and Machine Learning: Modeling and Estima
tion of Sparse Massive Networks - Part I - Jennife
r Chayes (Microsoft (UK))
DTSTART;TZID=Europe/London:20161212T133000
DTEND;TZID=Europe/London:20161212T143000
UID:TALK69453AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/69453
DESCRIPTION:There are numerous examples of sparse massive netw
orks\, in particular the Internet\, WWW and online
social networks. \; How do we model and learn
these networks? \; In contrast to conventiona
l learning problems\, where we have many independe
nt samples\, it is often the case for these networ
ks that we can get only one independent sample.&nb
sp\; How do we use a single snapshot today to lear
n a model for the network\, and therefore be able
to predict a similar\, but larger network in the f
uture? \; In the case of relatively small or m
oderately sized networks\, it&rsquo\;s appropriate
to model the network parametrically\, and attempt
to learn these parameters. \; For massive net
works\, a non-parametric representation is more ap
propriate. \; In this talk\, we first review t
he theory of graphons\, developed over the last de
cade to describe limits of dense graphs\, and the
more the recent theory describing sparse graphs of
unbounded average degree\, including power-law gr
aphs. \; We then show how to use these graphon
s as nonparametric models for sparse networks.&nbs
p\; Finally\, we show how to get consistent estima
tors of these non-parametric models\, and moreover
how to do this in a way that protects the privacy
of individuals on the network. \; \;

Part I of this talk reviews the theo
ry of graph convergence for dense and sparse graph
s. \; Part II uses the results of Part I to mo
del and estimate sparse massive networks.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR