Efficient Sampling of the Stationary Distribution

This article deals with an efficient sampling of the stationary distri-bution of dynamical systems in the presence of metastabilities. For such systems, standard sampling schemes suffer from trapping problems and critical slowing down. Starting multiple trajectories in different regions of the sampling space is a promising way out. The different samplings rep-resent the stationary distribution locally very well, but are still far away from ergodicity or from the global stationary distribution. We will show how these samplings can be joined together in order to get one global sampling of the stationary distribution. AMS MSC 2000: 62H30, 82B80, 65C40