BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Variational Inference: An Algorithm-Centric Perspective - Kyurae K
 im (University of Pennsylvania)
DTSTART:20251126T110000Z
DTEND:20251126T123000Z
UID:TALK241291@talks.cam.ac.uk
CONTACT:Xianda Sun
DESCRIPTION:This tutorial will introduce variational inference (VI)\, a fa
 mily of algorithms for approximate Bayesian inference. We will begin by mo
 tivating the need for VI\; when is it useful in the age of powerful Markov
  chain Monte Carlo algorithms such as NUTS? The tutorial will\nthen take a
  top-down approach\, introducing the abstract setup of variational inferen
 ce. The effect of each component of the setup\, such as the choice of vari
 ational family and divergence measure\, will be illustrated. The rest of t
 he tutorial will walk through the historical evolution of VI algorithms\, 
 but on a conceptual level\, focusing on intuitions. For instance\, what ar
 e the key ideas of a particular algorithm? What are its requirements and a
 ssumptions? What makes it\neffective\, and what are its limitations?
LOCATION:Cambridge University Engineering Department\, CBL Seminar room BE
 4-38.
END:VEVENT
END:VCALENDAR