Display Abstract
Title Mathematical foundations of Temperature Accelerated Dynamics (TAD)
Name David G Aristoff
Country USA
Email daristof@umn.edu
Co-Author(s) D. Aristoff and T. Lelievre
Submit Time 2013-11-20 17:59:58
Session
Special Session 88: Stochastic processes and spectral theory for partial differential equations and boundary value problems
Contents
Abstract: We give a mathematical framework for Temperature Accelerated Dynamics (TAD), a popular algorithm due to M.R. Sorensen and A.F. Voter for efficiently generating molecular dynamics in the presence of metastability. Using the notion of quasistationary distributions, we show that, under certain idealizing assumptions and with some small modifications to the algorithm, TAD becomes exact. We hope our framework will allow for a rigorous analysis of the error in TAD.
Go Back