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 Numerics - a snapshot of an emerging
community - Philipp Hennig (Max Planck Institute
for Intelligent Systems\, TÃ¼bingen)
DTSTART;TZID=Europe/London:20140911T110000
DTEND;TZID=Europe/London:20140911T120000
UID:TALK53999AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/53999
DESCRIPTION:Numerical methods for tasks like quadrature\, opti
mization\, linear algebra and the solution of diff
erential equations estimate latent quantities from
the observed result of tractable computations. In
this sense\, they are learning machines\, and acc
essible to the framework of probabilistic inferenc
e. \n\nWhat started as an entertaining observation
has\, over the past few years\, given rise to a s
mall community of researchers studying probabilist
ic numerical methods that has begun to produce non
trivial findings. I will give a brief (and biased)
overview of recent developments\, emphasizing a s
tring of results identifying classic numerical met
hods -- Gaussian quadrature\, conjugate gradients\
, BFGS\, Runge-Kutta -- with maximum a posteriori
estimators. \n\nStirring potential applications\,
coupled with a stack of fundamental questions up f
or grabs\, make probabilistic numerics an exciting
area at the boundary between mathematics and comp
uter science. Machine learning is ideally position
ed to both contribute and benefit from these devel
opments.
LOCATION:Engineering Department\, CBL Room BE-438
CONTACT:Dr Jes Frellsen
END:VEVENT
END:VCALENDAR