BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Non-reversible Markov chain Monte Carlo and hypocoercivity [CRC 17
 20: Early Career Seminar]
DTSTART:20260127T131500Z
DTEND:20260127T141500Z
DTSTAMP:20260716T203800Z
UID:indico-event-836@math-events.uni-bonn.de
CONTACT:sfb1720-seminar@uni-bonn.de
DESCRIPTION:Speakers: Francis Lörler\n\n \n\n\nHow can one construct a Ma
 rkov process that converges to a prescribed invariant measure as fast as p
 ossible? This question lies at the heart of Markov chain Monte Carlo\, a c
 ornerstone of modern computational mathematics. We will explore how non-re
 versibility and degenerate noise can overcome the slow converge of traditi
 onal reversible algorithms. We then show how these ideas are connected wit
 h the theory of hypocoercivity\, which explains rapid convergence in the a
 bsence of direct dissipation\, through the concept of second-order lifts.\
 n\n\n\n\nhttps://math-events.uni-bonn.de/event/836/
LOCATION:Endenicher Allee 60/1-016 - Lipschitzsaal (Mathezentrum)
URL:https://math-events.uni-bonn.de/event/836/
END:VEVENT
END:VCALENDAR