BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Efficient Bayesian Model Comparison with Differential Equations: a
  Population MCMC Approach via the Thermodynamic Integral - Ben Calderhead\
 , University of Glasgow
DTSTART:20100126T140000Z
DTEND:20100126T150000Z
UID:TALK22844@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:*Abstract:* In this talk I shall present some Bayesian methodo
 logy based on Population MCMC and thermodynamic integration\, which simult
 aneously addresses the commonly encountered issue of sampling from multimo
 dal posterior distributions\, as well as accurately estimating the margina
 l likelihoods of statistical models.\n\nI shall characterise the method an
 d compare to other approaches with some results using simple linear models
 \, before looking at more complex models based on nonlinear differential e
 quations.  Such differential equation models are employed in many fields o
 f science to describe processes occurring in the natural world\, and discr
 imination of plausible model hypotheses is vital to the application of the
  scientific method.\n\nFinally\, I shall discuss a sampling scheme in whic
 h inference over differential equation models may be accelerated by employ
 ing auxiliary Gaussian processes to avoid solving the dynamical systems ex
 plicitly.\n
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR