The pace of research on MCMC methods is so quick that any survey of advanced topics is immediately obsolete. The highly eclectic and decidedly biased coverage in our final chapter begins with a discussion of Markov random fields. Our limited aims here are to prove the Hammersley-Clifford theorem and introduce the Swendsen-Wang algorithm, a clever form of slice sampling. In the Ising model, the Swendsen-Wang algorithm is much more efficient than standard Gibbs sampling.
Anzeige
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten
