BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Uncertainty-Driven Construction of Markov Models from Accelerated 
 Molecular Dynamics - Dr Thomas D Swinburne\, Aix-Marseille University
DTSTART:20190206T141500Z
DTEND:20190206T151500Z
UID:TALK113914@talks.cam.ac.uk
CONTACT:Lisa Masters
DESCRIPTION:A common way of representing the long-time dynamics of materia
 ls is in terms of a Markov chain that specifies the transition rates for t
 ransitions between metastable states. This chain can either be used to gen
 erate trajectories using kinetic Monte Carlo\, or analyzed directly\, e.g.
 \, in terms of first passage times between distant states. While a number 
 of approaches have been proposed to infer such a representation from direc
 t molecular dynamics (MD) simulations\, challenges remain. For example\, a
 s chains inferred from a finite amount of MD will in general be incomplete
 \, quantifying their completeness is extremely desirable. Second\, making 
 the construction of the chain as computationally affordable as possible is
  paramount. \n\nI will talk about some recent work [1] to address these tw
 o questions. We first quantify the local completeness of the chain in term
 s of Bayesian estimators of the yet-unobserved rate\, and its global compl
 eteness in terms of the residence time of trajectories within the explored
  subspace. We then systematically reduce the cost of creating the chain by
  maximizing the increase in residence time against the distribution of sta
 tes in which additional MD is carried out and the temperature at which the
 se are respectively carried out. Using as example the behavior of vacancy 
 and interstitial clusters in materials\, we demonstrate that this is an ef
 ficient\, fully automated\, and massively-parallel scheme to efficiently e
 xplore the long-time behavior of materials. We also show how accommodation
  of exchange\, rotation\, reflection and translation symmetries can massiv
 ely enhance sampling efficiency.\n\n[1] TD Swinburne and D Perez\, Self-op
 timized construction of transition rate matrices from accelerated atomisti
 c simulations with Bayesian uncertainty quantification\, Physical Review M
 aterials 2018
LOCATION:Department of Chemistry\, Cambridge\, Unilever lecture theatre
END:VEVENT
END:VCALENDAR