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:Inference in Stochastic Processes - Javier Antoran
(University of Cambridge)\, Matthew Ashman (Unive
rsity of Cambridge)\, Stratis Markou (University o
f Cambridge)
DTSTART;TZID=Europe/London:20210224T110000
DTEND;TZID=Europe/London:20210224T123000
UID:TALK156730AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/156730
DESCRIPTION:In parametric models\, probabilistic inference is
most often approached by computing a posterior dis
tribution over model weights. These weights are th
en marginalised to obtain a distribution over func
tions and make predictions. If our goal is solely
to make good predictions\, an appealing alternativ
e is to directly perform inference over the ‘funct
ion-space’ or predictive posterior distribution of
our models\, without considering the posterior di
stribution over the weights. Using Gaussian Proces
ses (GPs) as motivation\, this talk starts by intr
oducing a method for constructing more general sto
chastic processes based on combining basis functio
ns with random weights. We discuss recent research
on performing approximate inference in the functi
on space of neural networks. Finally\, we provide
a brief introduction to Stochastic Differential Eq
uations (SDEs). We discuss the connection of linea
r SDEs to GPs and Kalman filtering and smoothing\,
and present a recent method for performing infere
nce and learning in nonlinear SDEs.\n\nRecommended
reading\n\n# Rasmussen & Williams\, “Gaussian pro
cess for Machine Learning”\, Chapter 2.2: "Functio
n space view"\, pages 13-18\n# Burt et. al. "Under
standing Variational Inference in Function-Space"
2020\n# Archambeau\, Cédric\, et al. "Variational
inference for diffusion processes.” 2008\n
LOCATION:https://eng-cam.zoom.us/j/86068703738?pwd=YnFleXFQ
OE1qR1h6Vmtwbno0LzFHdz09
CONTACT:Elre Oldewage
END:VEVENT
END:VCALENDAR