BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: 'Some advances and applications of robust gradient-ba
 sed Markov chain Monte Carlo'  - Sam Livingstone\, University College Lond
 on
DTSTART:20231003T140000Z
DTEND:20231003T150000Z
UID:TALK202714@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:I will discuss some recent work on gradient-based Markov chain
  Monte Carlo. At their best gradient-based algorithms are often state-of-t
 he-art\, but they can also behave erratically in various ways.  In the fir
 st part of the talk I'll introduce a gradient-based scheme called the Bark
 er proposal that can often match the performance of competitors such as MA
 LA and HMC but is probably more robust. I will in particular highlight the
  benefits this brings in the context of adaptive MCMC.  If time permits\, 
 in the second part of the talk I will talk about ongoing using MCMC for au
 tomatic model selection in poly-hazard models.  The first part is joint wo
 rk with Giacomo Zanella\, Jure Vogrinc & Max Hird\, the second is joint wi
 th Luke Hardcastle and Gianluca Baio.
LOCATION:MRC Biostatistics Unit\, East Forvie Building\, Forvie Site Robin
 son Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR