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:Machine Learning @ CUED
SUMMARY:Probabilistic Numerical Computation: A New Concept
? - Prof. Mark Girolami (University of Warwick)
DTSTART;TZID=Europe/London:20161117T110000
DTEND;TZID=Europe/London:20161117T120000
UID:TALK69133AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/69133
DESCRIPTION:Ambitious mathematical models of highly complex na
tural phenomena are challenging to analyse\, and m
ore and more computationally expensive to evaluate
. This is a particularly acute problem for many ta
sks of interest and numerical methods will tend to
be slow\, due to the complexity of the models\, a
nd potentially lead to sub-optimal solutions with
high levels of uncertainty which needs to be accou
nted for and subsequently propagated in the statis
tical reasoning process. This talk will introduce
our contributions to an emerging area of research
defining a nexus of applied mathematics\, statisti
cal science and computer science\, called "probabi
listic numerics". The aim is to consider numerical
problems from a statistical viewpoint\, and as su
ch provide numerical methods for which numerical e
rror can be quantified and controlled in a probabi
listic manner. This philosophy will be illustrated
on problems ranging from predictive policing via
crime modelling to computer vision\, where probabi
listic numerical methods provide a rich and essent
ial quantification of the uncertainty associated w
ith such models and their computation. \n
LOCATION:CBL Room BE-438\, Department of Engineering
CONTACT:Zoubin Ghahramani
END:VEVENT
END:VCALENDAR