On some information-theoretic aspects of non-linear statistical inverse problems
Richard Nickl
University of Cambridge, Faculty of Mathematics, Cambridge CB3 0WA, UKGabriel P. Paternain
University of Cambridge, Faculty of Mathematics, Cambridge CB3 0WA, UK
![On some information-theoretic aspects of non-linear statistical inverse problems cover](/_next/image?url=https%3A%2F%2Fcontent.preview.ems.press%2Fassets%2Fpublic%2Fimages%2Fbooks%2Fcover-279.png&w=3840&q=90)
This book chapter is published open access.
Abstract
Results by van der Vaart (1991) from semi-parametric statistics about the existence of a non-zero Fisher information are reviewed in an infinite-dimensional non-linear Gaussian regression setting. Information-theoretically optimal inference on aspects of the unknown parameter is possible if and only if the adjoint of the linearisation of the regression map satisfies a certain range condition. It is shown that this range condition may fail in a commonly studied elliptic inverse problem with a divergence form equation (‘Darcy’s problem’), and that a large class of smooth linear functionals of the conductivity parameter cannot be estimated efficiently in this case. In particular, Gaussian ‘Bernstein von Mises’-type approximations for Bayesian posterior distributions do not hold in this setting.