BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian coresets - Tor Fjelde and Will Tebbutt\, University of Ca
 mbridge
DTSTART:20240207T110000Z
DTEND:20240207T123000Z
UID:TALK212020@talks.cam.ac.uk
CONTACT:Isaac Reid
DESCRIPTION:The proliferation of automated inference algorithms in Bayesia
 n\nstatistics has provided practitioners access to fast\, reproducible dat
 a\nanalysis and powerful statistical models. But designing automated\nmeth
 ods that are both scalable in the number of observations in addition\n to 
 theoretically sound is still a challenging task.\n\nIn this talk we'll dis
 cuss a family of recent works trying to address this referred to as Bayesi
 an coresets. We'll discuss the basic idea behind Bayesian coresets\, and t
 hen look at some of the literature together.\n\nRecomended reading: Campbe
 ll & Beronov (2019) Sparse Variational Inference: Bayesian Coresets From S
 cratch. URL: http://arxiv.org/abs/1906.03329
LOCATION:Cambridge University Engineering Department\, CBL Seminar room BE
 4-38.
END:VEVENT
END:VCALENDAR