BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Experimental Design: Principles and Computation - Desi Iv
 anova (University of Oxford)
DTSTART:20250623T104500Z
DTEND:20250623T114500Z
UID:TALK232180@talks.cam.ac.uk
DESCRIPTION:Bayesian experimental design (BED) provides a powerful and pri
 ncipled information-theoretic framework for optimising the design of exper
 iments. In this tutorial\, we will introduce the core principles underlyin
 g BED before discussing computational methods essential for practical impl
 ementation\, with a particular focus on modern approximation techniques su
 ch as (amortised) variational inference. Finally\, will cover the policy-b
 ased BED (PB-BED)\, which enables efficient online deployment of the BED p
 ipeline by learning adaptive design strategies through offline training.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR