BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Variational Bayes and Beyond: Foundations of Scalable Bayesian Inf
 erence - Tamara Broderick (MIT)
DTSTART:20200115T140000Z
DTEND:20200115T160000Z
UID:TALK136285@talks.cam.ac.uk
CONTACT:J.W.Stevens
DESCRIPTION:Bayesian methods exhibit a number of desirable properties\nfor
  modern data analysis---including (1) coherent quantification of uncertain
 ty\, (2) a modular modeling framework able to capture complex phenomena\, 
 (3) the ability to incorporate prior information from an expert source\, a
 nd (4) interpretability. In practice\, though\, Bayesian\ninference necess
 itates approximation of a high-dimensional integral\, and some traditional
  algorithms for this purpose can be slow---notably at data scales of curre
 nt interest. The tutorial will cover the foundations of some modern tools 
 for fast\, approximate Bayesian inference at scale. One increasingly popul
 ar framework is provided by\n"variational Bayes" (VB)\, which formulates B
 ayesian inference as an optimization problem. We will examine key benefits
  and pitfalls of using VB in practice\, with a focus on the widespread "me
 an-field variational Bayes" (MFVB) subtype. We will highlight properties t
 hat anyone working with VB\, from the data analyst to the theoretician\, s
 hould be aware of. And we will discuss a number of open challenges.
LOCATION:MR15 (GL.02)\, Pavilion G\, CMS
END:VEVENT
END:VCALENDAR