Regensburg 2025 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 22: Poster: Statistical Physics
DY 22.7: Poster
Wednesday, March 19, 2025, 10:00–12:00, P3
Large-deviation simulation of the coupling time distribution for the CFTP method applied to the heat bath Ising process — •Mathis Groenhagen, Alexander K. Hartmann, and Peter Werner — Institut für Physik, Carl von Ossietzky Universität Oldenburg, 26111 Oldenburg, Germany
Coupling from the past (CFTP), introduced by Propp and Wilson [1], is a version of the Markov-chain Monte Carlo method, which is capable of generating exact samples from a finite set with a particular distribution. The performance of the CFTP method for a given application can be characterized by the distribution of the CFTP method’s random running time, the coupling time τ, for this application.
A large-deviation Monte Carlo algorithm, as described for example in [2], is used to sample these coupling times for the application of the CFTP method to a single-spin-update heat-bath process for the ferromagnetic two-dimensional square lattice Ising model without an external field. This yields the coupling time distributions over a wide range of coupling times τ down to probability densities of 10−80 for different lattice sizes L and temperatures T. These results give additional numerical evidence for the analytical results shown in [3].
[1] J. Propp, D. Willson, Random Struct. Algorithms 9, 223-252 (1996).
[2] A. K. Hartmann, Phys. Rev. E 65, 056102 (2002).
[3] A. Collevecchio, E.M. Elçi, T.M. Garoni et al., J Stat Phys 170, 22-61 (2018).
Keywords: computer simulations; large deviation theory; Ising model; ferromagnet; coupling from the past