Invariant Bayesian estimation on manifolds
From MaRDI portal
Publication:2388349
Abstract: A frequent and well-founded criticism of the maximum a posteriori (MAP) and minimum mean squared error (MMSE) estimates of a continuous parameter gamma taking values in a differentiable manifold Gamma is that they are not invariant to arbitrary ``reparameterizations of Gamma. This paper clarifies the issues surrounding this problem, by pointing out the difference between coordinate invariance, which is a sine qua non for a mathematically well-defined problem, and diffeomorphism invariance, which is a substantial issue, and then provides a solution. We first show that the presence of a metric structure on Gamma can be used to define coordinate-invariant MAP and MMSE estimates, and we argue that this is the natural way to proceed. We then discuss the choice of a metric structure on Gamma. By imposing an invariance criterion natural within a Bayesian framework, we show that this choice is essentially unique. It does not necessarily correspond to a choice of coordinates. In cases of complete prior ignorance, when Jeffreys' prior is used, the invariant MAP estimate reduces to the maximum likelihood estimate. The invariant MAP estimate coincides with the minimum message length (MML) estimate, but no discretization or approximation is used in its derivation.
Recommendations
- A Bayesian approach to the estimation of maps between Riemannian manifolds
- Intrinsic analysis of statistical estimation
- An invariant form for the prior probability in estimation problems
- scientific article; zbMATH DE number 3998952
- A Bayesian approach to the estimation of maps between Riemannian manifolds. II: examples
Cites work
- An Information Measure for Classification
- An information measure for single link classification
- Assessing the accuracy of the maximum likelihood estimator: Observed versus expected Fisher information
- Differential-geometrical methods in statistics
- scientific article; zbMATH DE number 4066122 (Why is no real title available?)
- scientific article; zbMATH DE number 4084717 (Why is no real title available?)
- scientific article; zbMATH DE number 3667770 (Why is no real title available?)
- scientific article; zbMATH DE number 3599182 (Why is no real title available?)
- scientific article; zbMATH DE number 1055955 (Why is no real title available?)
- scientific article; zbMATH DE number 217456 (Why is no real title available?)
- scientific article; zbMATH DE number 3106666 (Why is no real title available?)
- Preferred point geometry and the local differential geometry of the Kullback-Leibler divergence
- Riemannian center of mass and mollifier smoothing
- The Selection of Prior Distributions by Formal Rules
Cited in
(9)- A Riemannian Bayesian framework for estimating diffusion tensor images
- A Bayesian approach to the estimation of maps between Riemannian manifolds. II: examples
- Equi-affine invariant geometry for shape analysis
- Kernel Methods for Bayesian Elliptic Inverse Problems on Manifolds
- A nonlocal denoising algorithm for manifold-valued images using second order statistics
- Variational Bayesian data analysis on manifold
- Generalized fiducial inference on differentiable manifolds
- Intrinsic Bayesian active contours for extraction of object boundaries in images
- A Bayesian approach to the estimation of maps between Riemannian manifolds
This page was built for publication: Invariant Bayesian estimation on manifolds
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2388349)
