English

\Gamma-convergence of Onsager-Machlup functionals. Part I: With applications to maximum a posteriori estimation in Bayesian inverse problems

Statistics Theory 2022-01-10 v3 Numerical Analysis Numerical Analysis Probability Statistics Theory

Abstract

The Bayesian solution to a statistical inverse problem can be summarised by a mode of the posterior distribution, i.e. a MAP estimator. The MAP estimator essentially coincides with the (regularised) variational solution to the inverse problem, seen as minimisation of the Onsager-Machlup functional of the posterior measure. An open problem in the stability analysis of inverse problems is to establish a relationship between the convergence properties of solutions obtained by the variational approach and by the Bayesian approach. To address this problem, we propose a general convergence theory for modes that is based on the Γ\Gamma-convergence of Onsager-Machlup functionals, and apply this theory to Bayesian inverse problems with Gaussian and edge-preserving Besov priors. Part II of this paper considers more general prior distributions.

Keywords

Cite

@article{arxiv.2108.04597,
  title  = {\Gamma-convergence of Onsager-Machlup functionals. Part I: With applications to maximum a posteriori estimation in Bayesian inverse problems},
  author = {Birzhan Ayanbayev and Ilja Klebanov and Han Cheng Lie and T. J. Sullivan},
  journal= {arXiv preprint arXiv:2108.04597},
  year   = {2022}
}

Comments

30 pages