BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Stefan Grosskinsky (Universität Ausburg)
DTSTART:20211208T160000Z
DTEND:20211208T170000Z
DTSTAMP:20260423T021247Z
UID:SMES/8
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/SMES/8/">Fey
 nman-Kac particle models for cloning algorithms</a>\nby Stefan Grosskinsky
  (Universität Ausburg) as part of Probability and Statistical Mechanics S
 eminar\n\n\nAbstract\nDynamic large deviations for additive path functiona
 ls of stochastic processes have attracted recent research interest\, in pa
 rticular in the context of stochastic particle systems and statistical phy
 sics. Efficient numerical 'cloning' algorithms have been developed to esti
 mate the scaled cumulant generating function\, based on importance samplin
 g via cloning of rare event trajectories. Adapting previous results from t
 he literature of particle filters and sequential Monte Carlo methods\, we 
 use Feynman-Kac models to establish fully rigorous bounds on systematic an
 d random errors of cloning algorithms in continuous time. To this end we d
 evelop a method to compare different algorithms for particular classes of 
 observables\, based on the martingale characterization and related to the 
 propagation of chaos for mean-field models. Our results apply to a large c
 lass of jump processes on locally compact state space\, and provide a fram
 ework that can also be used to evaluate and improve the efficiency of algo
 rithms. This is joint work with Letizia Angeli\, Adam Johansen and Andrea 
 Pizzoferrato.\n
LOCATION:https://researchseminars.org/talk/SMES/8/
END:VEVENT
END:VCALENDAR