Inference with minimal Gibbs free energy in information field theory

  title={Inference with minimal Gibbs free energy in information field theory},
  author={Torsten A. Ensslin and Cornelius Weig},
  journal={Physical review. E, Statistical, nonlinear, and soft matter physics},
  volume={82 5 Pt 1},
  • T. EnsslinCornelius Weig
  • Published 16 April 2010
  • Computer Science
  • Physical review. E, Statistical, nonlinear, and soft matter physics
Non-linear and non-gaussian signal inference problems are difficult to tackle. Renormalization techniques permit us to construct good estimators for the posterior signal mean within information field theory (IFT), but the approximations and assumptions made are not very obvious. Here we introduce the simple concept of minimal Gibbs free energy to IFT, and show that previous renormalization results emerge naturally. They can be understood as being the gaussian approximation to the full posterior… 

Comment on "Inference with minimal Gibbs free energy in information field theory".

It is shown explicitly that MGFE indeed incorporates the MAP principle, as well as the MDI (minimum discrimination information) approach, but not the well-known ME principle of Jaynes, which is important for Bayesian inference problems.

Reconstruction of Gaussian and log-normal fields with spectral smoothness

It is shown that the minimization of the Gibbs free energy, corresponding to a Gaussian approximation to the posterior marginalized over the power spectrum, is equivalent to the empirical Bayes ansatz, in which the power Spectrum is fixed to its maximum a posteriori value.

Reconstruction of signals with unknown spectra in information field theory with parameter uncertainty

The general problem of signal inference in the presence of unknown parameters within the framework of information field theory is formulated and a generic parameter-uncertainty renormalized estimation (PURE) technique is developed.

Reconstructing signals from noisy data with unknown signal and noise covariance.

A method to reconstruct Gaussian signals from linear measurements with Gaussian noise using the principle of minimum Gibbs free energy and extending this algorithm to simultaneously uncertain noise and signal covariances using the same principles in the derivation.

Denoising, deconvolving, and decomposing photon observations - Derivation of the D3PO algorithm

The D 3 PO algorithm addresses the inference problem of denoising, deconvolving, and decomposing photon observations and successfully denoised, deconvolved, and decomposed the data into a diuse and a point-like signal estimate for the respective photon flux components.

Information Theory for Fields

  • T. Ensslin
  • Computer Science, Physics
    Annalen der Physik
  • 2019
Information field theory is built upon the language of mathematical physics, in particular, on field theory and statistical mechanics and permits the mathematical derivation of optimal imaging algorithms, data analysis methods, and even computer simulation schemes.

Generic inference of inflation models by non-Gaussianity and primordial power spectrum reconstruction

We present a generic inference method for inflation models from observational data by the usage of higher-order statistics of the curvature perturbation on uniform density hypersurfaces. This method

Signal inference in Galactic astrophysics

In this thesis we present a combination of methodological work with very wide focus and some specific astrophysical applications. We advance the knowledge on the Galactic interstellar medium by

Information Field Theory and Artificial Intelligence

This paper reformulated the process of inference in IFT in terms of GNN training, suggesting that IFT is well suited to address many problems in AI and ML research and application.

Tomography of the Galactic free electron density with the Square Kilometer Array

We present a new algorithm for reconstructing the Galactic free electron density from pulsar dispersion measures. The algorithm performs a nonparametric tomography for a density field with an



Bayesian field theory

Bayesian Field Theory offers practical examples of Bayesian analysis for the physicist working in such areas as neural networks, artificial intelligence, and inverse problems in quantum theory, and includes nonparametric density estimation problems, including, as special cases,nonparametric regression and pattern recognition.

Water Resources Research

Water Resources Research (WRR) is a peer-reviewed scientific journal published by the American Geophysical Union. AGU states that WRR is an "interdisciplinary journal integrating research in the


Maximum Entropy and Bayesian Methods.

Abstract : This volume contains selections from among the presentations at the Thirteenth International Workshop on Maximum Entropy and Bayesian Methods- MAXENT93 for short- held at the University of

Physical Review

  • Art
  • 1965


  • 1994