BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Faster log-concave sampling via algorithmic warm starts - Sinho Ch
 ewi (Institute for Advanced Study)
DTSTART:20231006T130000Z
DTEND:20231006T140000Z
UID:TALK206002@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:The problem of sampling from a log-concave distribution is a k
 ey algorithmic component of fields such as Bayesian inference\, yet non-as
 ymptotic computational guarantees for this task have only emerged recently
 \, within the last decade. In this talk\, I'll discuss recent progress on 
 understanding one of the most popular samplers\, the Metropolis-adjusted L
 angevin algorithm (MALA)\, by first showing refined mixing time bounds und
 er a warm start\, and then showing how to algorithmically obtain the warm 
 start via the underdamped Langevin process.
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR