Minimax theory for a class of non-linear statistical inverse problems
arXiv:1512.00218 · doi:10.1088/0266-5611/32/6/065003
Abstract
We study a class of statistical inverse problems with non-linear pointwise operators motivated by concrete statistical applications. A two-step procedure is proposed, where the first step smoothes the data and inverts the non-linearity. This reduces the initial non-linear problem to a linear inverse problem with deterministic noise, which is then solved in a second step. The noise reduction step is based on wavelet thresholding and is shown to be minimax optimal (up to logarithmic factors) in a pointwise function-dependent sense. Our analysis is based on a modified notion of Hölder smoothness scales that are natural in this setting.
arXiv comment37 pages
References in corpus (6)
- Iteratively regularized Newton-type methods for general data misfit functionals and applications to Poisson data
- Nonparametric regression in exponential families
- Asymptotic equivalence of functional linear regression and a white noise inverse problem
- Equivalence theory for density estimation, Poisson processes and Gaussian white noise with drift
- Adaptive function estimation in nonparametric regression with one-sided errors
- Adaptation to lowest density regions with application to support recovery