Abstract : We analyze the convergence properties of the Wang-Landau algorithm. This sampling method belongs to the general class of adaptive importance sampling strategies which use the free energy along a chosen reaction coordinate as a bias. Such algorithms are very helpful to enhance the sampling properties of Markov Chain Monte Carlo algorithms, when the dynamics is metastable. We prove the convergence of the Wang-Landau algorithm and an associated central limit theorem.
Type de document :
Article dans une revue
Applied Mathematics Research eXpress, Oxford University Press (OUP): Policy H - Oxford Open Option A, 2014, 2014, pp.275-311. 〈10.1093/amrx/abu003〉
https://hal.inria.fr/hal-00721886
Contributeur : Tony Lelievre
<>
Soumis le : mardi 31 juillet 2012 - 08:19:43
Dernière modification le : jeudi 10 mai 2018 - 02:03:58
G. Fort, B. Jourdain, E. Kuhn, T. Lelievre, G. Stoltz. Efficiency of the Wang-Landau Algorithm: A Simple Test Case. Applied Mathematics Research eXpress, Oxford University Press (OUP): Policy H - Oxford Open Option A, 2014, 2014, pp.275-311. 〈10.1093/amrx/abu003〉. 〈hal-00721886〉
