Wolff algorithm

The Wolff algorithm is a Monte Carlo algorithm for the simulation of statistical processes, in particular the Ising model .

The Wolff algorithm is one of the cluster algorithms (an area of ​​the MCMC method ) that are particularly effective in the area of phase transitions . Cluster algorithms require significantly less computing time in the vicinity of the critical temperature than local algorithms, as they counteract the divergence of the autocorrelation times in the vicinity of phase transitions - the critical slowing down .

In contrast to local algorithms such as the Metropolis algorithm, the Wolff algorithm does not carry out individual local updates one after the other, but rather changes entire clusters at once. This reduces the correlations that go far in the vicinity of the critical temperature . For simulations far away from the critical point, however, the Wolff algorithm is less effective than local algorithms.

The algorithm was developed in 1989 by the German physics professor Ulrich Wolff , currently working at the Humboldt University in Berlin .

See also

• Swendsen-Wang algorithm

literature

• U. Wolff: Collective monte carlo updating for spin systems . In: Phys. Rev. Lett. tape , 62 , 1989, pp. 361-364 .

• MEJ Newman, GT Barkema: Monte Carlo Methods in Statistical Physics . Oxford University Press, New York 1999, ISBN 0-19-851797-1 .