Thursday, January 23, 2025 - 17:00 in V3-201
Time-inhomogeneous approximations of time-homogeneous Markov chains
A talk in the Oberseminar Probability Theory and Mathematical Statistics series by
Daniel Rudolf from Universität Passau
| Abstract: |
We investigate the approximation of a time-homogeneous Markov chain by a time-inhomogeneous one.
We prove an upper bound of the absolute difference of expectations of coordinate Lipschitz functions along finite length trajectories of both Markov chains. As a special case, we obtain estimates of the absolute expected difference of the sample averages with respect to function values along the trajectories as well as a Wasserstein perturbation bound of the n-th step distributions. We illustrate our result in a stochastic gradient descent scenario.
Within the CRC this talk is associated to the project(s): B10 |
Back
