Project Details
Projekt Print View

Models of percolation based on random walks

Subject Area Mathematics
Term from 2017 to 2021
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 390200145
 

Final Report Abstract

1. It has been proved that the random walk loop soup in dimensions d ≥ 3 satisfies a useful decoupling inequality. It has been shown that in a class of strongly correlated percolation models that satisfy such a decoupling inequality, the unique infinite cluster has properties similar to those of uncorrelated percolation; for example, the random walk on the infinite cluster satisfies the quenched invariance principle and the quenched Gaussian heat-kernel bounds. Both the random walk loop soup and its vacant set are in this class. 2. It has been proved that the Poisson cylinder’s percolation in dimensions d ≥ 3 satisfies a useful decoupling inequality; consequently, it has been shown that the occupied set is almost surely transient. 3. The sharpness of percolation phase transition has been shown for some planar dynamical models of percolation, including a class of opinion dynamics models and the Glauber dynamics for the Ising model.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung