BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Path-space information metrics for uncertainty quantification and 
 coarse-graining of molecular systems - Markos A. Katsoulakis (University o
 f Massachusetts)
DTSTART:20160614T140000Z
DTEND:20160614T150000Z
UID:TALK66443@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:We present path-space\, information theory-based\, sensitivity
  analysis\, uncertainty quantification and variational inference methods f
 or complex high-dimensional stochastic dynamics\, including chemical react
 ion networks with hundreds of parameters\, Langevin-type equations and lat
 tice kinetic Monte Carlo. We establish their connections with goal-oriente
 d methods in terms of new\, sharp\, uncertainty quantification inequalitie
 s that scale appropriately at both long times and for high dimensional sta
 te and parameter space.  &nbsp\;  The combination of proposed methodologie
 s is capable to (a) tackle non-equilibrium processes\, typically associate
 d with coupled physicochemical mechanisms or boundary conditions\, such as
  reaction-diffusion problems\, and where even steady states are unknown al
 together\, e.g. do not have a Gibbs structure. The path-wise information t
 heory tools\,&nbsp\; (b) yield a surprisingly simple\, tractable and easy-
 to-implement approach to quantify and rank parameter sensitivities\, as we
 ll as&nbsp\; (c) provide reliable parameterizations for coarse-grained mol
 ecular systems based on fine-scale data\, and rational model selection thr
 ough path-space (dynamics-based) variational inference methods.
LOCATION:Seminar Room 2\, Newton Institute
END:VEVENT
END:VCALENDAR