BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:An Elementary Introduction to Sequential Monte Carlo Samplers - Xi
 anda Sun (University of Cambridge)
DTSTART:20241127T110000Z
DTEND:20241127T123000Z
UID:TALK224992@talks.cam.ac.uk
CONTACT:Xianda Sun
DESCRIPTION:Sequential Monte Carlo (SMC) methods provide a compelling alte
 rnative to traditional Markov Chain Monte Carlo (MCMC) approaches for samp
 ling from unnormalized probability distributions while also delivering unb
 iased estimates of the normalizing constant. Unlike MCMC\, which generates
  a single chain of samples\, SMC methods operate by evolving a population 
 of weighted particles through a sequence of intermediate distributions\, u
 ltimately converging on the target density.\n\nDespite their flexibility a
 nd strong theoretical underpinnings\, SMC samplers are still underutilized
  in practice. This talk offers a conceptual introduction to SMC samplers\,
  following the framework outlined by Dai\, Heng\, Jacob\, and Whiteley in 
 their paper "_An Invitation to Sequential Monte Carlo Samplers_." Instead 
 of focusing on state-space models\, the presentation will unpack the essen
 tial components of SMC samplers\, making them more accessible to beginners
  and researchers in probabilistic machine learning and related disciplines
 . The goal is to provide a stepping stone into the broader\, rich literatu
 re on SMC methods.
LOCATION:Cambridge University Engineering Department\, CBL Seminar room BE
 4-38.
END:VEVENT
END:VCALENDAR