Bayesian linear inverse problems in regularity scales
arXiv:1802.08992 · doi:10.1214/19-AIHP1029
Abstract
We obtain rates of contraction of posterior distributions in inverse problems defined by scales of smoothness classes. We derive abstract results for general priors, with contraction rates determined by Galerkin approximation. The rate depends on the amount of prior concentration near the true function and the prior mass of functions with inferior Galerkin approximation. We apply the general result to non-conjugate series priors, showing that these priors give near optimal and adaptive recovery in some generality, Gaussian priors, and mixtures of Gaussian priors, where the latter are also shown to be near optimal and adaptive. The proofs are based on general testing and approximation arguments, without explicit calculations on the posterior distribution. We are thus not restricted to priors based on the singular value decomposition of the operator. We illustrate the results with examples of inverse problems resulting from differential equations.
arXiv comment34 pages
References in corpus (12)
- Rates of contraction of posterior distributions based on Gaussian process priors
- Convergence rates of posterior distributions for noniid observations
- Bayesian inverse problems with Gaussian priors
- Nonparametric Bernstein-von Mises theorems in Gaussian white noise
- Posterior Contraction Rates for the Bayesian Approach to Linear Ill-Posed Inverse Problems
- Nonparametric Bayesian posterior contraction rates for discretely observed scalar diffusions
- Bayesian inverse problems with non-conjugate priors
- Bayesian recovery of the initial condition for the heat equation
- Bayesian optimal adaptive estimation using a sieve prior
- Reconstruction from Radon projections and orthogonal expansion on a ball
- Bayesian inverse problems with unknown operators
- Nonparametric Bayesian model selection and averaging