BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Probabilistic Numerics - a snapshot of an emerging community - Phi
 lipp Hennig (Max Planck Institute for Intelligent Systems\, Tübingen)
DTSTART:20140911T100000Z
DTEND:20140911T110000Z
UID:TALK53999@talks.cam.ac.uk
CONTACT:Dr Jes Frellsen
DESCRIPTION:Numerical methods for tasks like quadrature\, optimization\, l
 inear algebra and the solution of differential equations estimate latent q
 uantities from the observed result of tractable computations. In this sens
 e\, they are learning machines\, and accessible to the framework of probab
 ilistic inference. \n\nWhat started as an entertaining observation has\, o
 ver the past few years\, given rise to a small community of researchers st
 udying probabilistic numerical methods that has begun to produce nontrivia
 l findings. I will give a brief (and biased) overview of recent developmen
 ts\, emphasizing a string of results identifying classic numerical methods
  -- Gaussian quadrature\, conjugate gradients\, BFGS\, Runge-Kutta -- with
  maximum a posteriori estimators. \n\nStirring potential applications\, co
 upled with a stack of fundamental questions up for grabs\, make probabilis
 tic numerics an exciting area at the boundary between mathematics and comp
 uter science. Machine learning is ideally positioned to both contribute an
 d benefit from these developments.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR