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 Reading Group @ CUED
SUMMARY:Integrated Nested Laplace Approximation (INLA) - S
ara Wade (University of Cambridge)
DTSTART;TZID=Europe/London:20131128T150000
DTEND;TZID=Europe/London:20131128T163000
UID:TALK49164AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/49164
DESCRIPTION:Integrated nested Laplace approximation (INLA) is
an algorithm for approximate Bayesian inference in
a class of latent Gaussian models. This class of
models is characterized by linking the possibly no
n-Gaussian outputs to the inputs through a latent
Gaussian field controlled by few hyperparameters a
nd includes\, among others\, generalized linear mo
dels\, additive models\, smoothing splines\, state
space models\, spatial and spatiotemporal models\
, and log-Gaussian Cox processes. The main advanta
ge of INLA over other Bayesian inference methods\,
such as MCMC\, is computation time. In this talk\
, we will describe the algorithm in detail and pro
vide a demo of the R-INLA package.
LOCATION:Engineering Department\, CBL Room 438
CONTACT:
END:VEVENT
END:VCALENDAR