BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: "Predictive resampling for scalable Bayes" - Edwin Fo
 ng\, Department of Statistics and Actuarial Science\, University of Hong K
 ong
DTSTART:20251104T100000Z
DTEND:20251104T110000Z
UID:TALK240121@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:The martingale posterior is a framework for Bayesian inference
  in which posterior uncertainty is generated through predictive imputation
 . A key advantage of this approach is that the Bayesian model can be direc
 tly specified using a sequence of predictive\n distributions. This elimina
 tes the need for explicit likelihood and prior specifications\, thereby av
 oiding the computational demands of MCMC. Instead\, posterior sampling for
  martingale posteriors relies on predictive resampling—a parallelizable\
 , bootstrap-like\n procedure that is highly efficient. This talk will high
 light recent computational advances in martingale posteriors\, enabling sc
 alable posterior inference for both nonparametric and parametric models.
LOCATION:Online Seminar
END:VEVENT
END:VCALENDAR