BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Uncertainty quantification in Gaussian Graphical Models - Déborah
  Sulem (Università della Svizzera italiana)
DTSTART:20250828T150000Z
DTEND:20250828T153000Z
UID:TALK234526@talks.cam.ac.uk
DESCRIPTION:Co-authors: Jack Jewson and David Rossell.\nGaussian graphical
  models are widely used to infer dependence structures. Bayesian methods a
 re appealing to quantify uncertainty associated with structural learning\,
  that is on the plausibility of conditional independence statements given 
 the data\, and on parameter estimates. However\, computational demands hav
 e limited their application when the number of variables is large\, which 
 prompted the use of pseudo-Bayesian approaches. We propose fully Bayesian 
 algorithms that provably scale well to high dimensions when the data-gener
 ating precision matrix is sparse\, at a similar cost to the best available
  pseudo-Bayesian methods. Our examples show that the methods extend the ap
 plicability of exact Bayesian inference from roughly one hundred to roughl
 y one thousand variables (equivalently\, from 5\,000 edges to 500\,000 edg
 es) if one desires a solution within a few seconds or minutes. All methods
  are implemented in the R package mombf.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR