Posterior consistency in Bayesian inference with exponential priors

20.11.2020, 10:00  –  online

Masoumeh Dashti (University of Sussex, UK)


We consider the problem of recovering the underlying truth in a nonparametric Bayesian inference setting with p-exponential priors. These priors are a class of infinite dimensional product measures with tails between exponential and Gaussian. We will discuss some results on the rates of contraction of the posterior around the truth for the white noise and density estimation models.

This is joint work with S. Agapiou and T. Helin.

invited by Jana de Wiljes

zu den Veranstaltungen