Physics-inspired forms of the Bayesian Cramér-Rao bound

07/09/2020
by   Mankei Tsang, et al.
0

Using differential geometry, I derive a form of the Bayesian Cramér-Rao bound that remains invariant under reparametrization. With the invariant formulation at hand, I find the optimal and naturally invariant bound among the Gill-Levit family of bounds. By assuming that the prior probability density is the square of a wavefunction, I also express the bounds in terms of functionals that are quadratic with respect to the wavefunction and its gradient. The problem of finding an unfavorable prior to tighten the bound for minimax estimation is shown, in a special case, to be equivalent to finding the ground state of a Schrödinger equation, with the Fisher information playing the role of the potential. To illustrate the theory, two quantum estimation problems, namely, optomechanical waveform estimation and subdiffraction incoherent optical imaging, are discussed.

READ FULL TEXT
research
02/22/2019

A Family of Bayesian Cramér-Rao Bounds, and Consequences for Log-Concave Priors

Under minimal regularity assumptions, we establish a family of informati...
research
11/19/2021

Non-NP-Hardness of Translationally-Invariant Spin-Model Problems

Finding the ground state energy of the Heisenberg Hamiltonian is an impo...
research
11/24/2020

Minimax rate of estimation for the stationary distribution of jump-processes over anisotropic Holder classes

We study the problem of the non-parametric estimation for the density of...
research
01/15/2018

Information Geometric Approach to Bayesian Lower Error Bounds

Information geometry describes a framework where probability densities c...
research
12/30/2022

PAC-Bayesian-Like Error Bound for a Class of Linear Time-Invariant Stochastic State-Space Models

In this paper we derive a PAC-Bayesian-Like error bound for a class of s...
research
03/01/2023

On Parametric Misspecified Bayesian Cramér-Rao bound: An application to linear Gaussian systems

A lower bound is an important tool for predicting the performance that a...

Please sign up or login with your details

Forgot password? Click here to reset