Dear Allstat,
I have the great pleasure of announcing a special seminar by Professor Matt Wand from University of Technology Sydney, Australia with details as below. Matt has a long and distinguished research career (many people will know his 1995 book on kernel smoothing with Chris Jones, for example). Matt is in Leeds for a two-week research visit and will give a seminar based on his 2017 JASA paper -- link below.
Date: Wednesday 25th July 2018
Time: 11:00-12:00
Venue: School of Mathematics, Level 8, Room MALL 1
Matt Wand (University of Technology Sydney, Australia)
Fast Approximate Inference for Arbitrarily Large Statistical Models via Message Passing
We explain how the notion of message passing can be used to streamline the algebra and computer coding for fast approximate inference in large Bayesian statistical models. In particular, this approach is amenable to handling arbitrarily large models of particular types once a set of primitive operations is established. The approach is founded upon a message passing formulation of mean field variational Bayes that utilizes factor graph representations of statistical models. The notion of factor graph fragments is introduced and is shown to facilitate compartmentalization of the required algebra and coding.
This seminar is based on Matt's 2017 JASA paper which can be downloaded from http://matt-wand.utsacademics.info/
For further information see:
https://www.uts.edu.au/staff/matt.wand (official webpage) http://matt-wand.utsacademics.info/ (personal webpage)
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|