Region selection in Markov random fields: Gaussian case

DOI10.1016/J.JMVA.2023.105178MaRDI QIDQ6097551zbMATH OpenOpenAlexFDO

Authors I. Soloveychik, Vahid Tarokh

Publication date 5 June 2023

Published in Journal of Multivariate Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1802.03848

zbMATH Keywords

enumeration of polyominoesFano's inequalityGaussian graphical modelsMarkov random fieldsmodel selection

Mathematics Subject Classification ID

Polyominoes (05B50) Large deviations (60F10) Probabilistic graphical models (62H22) Random fields; image analysis (62M40)

Abstract: We consider the problem of model selection in Gaussian Markov fields in the sample deficient scenario. The benchmark information-theoretic results in the case of d-regular graphs require the number of samples to be at least proportional to the logarithm of the number of vertices to allow consistent graph recovery. When the number of samples is less than this amount, reliable detection of all edges is impossible. In many applications, it is more important to learn the distribution of the edge (coupling) parameters over the network than the specific locations of the edges. Assuming that the entire graph can be partitioned into a number of spatial regions with similar edge parameters and reasonably regular boundaries, we develop new information-theoretic sample complexity bounds and show that a bounded number of samples can be sufficient to consistently recover these regions. Finally, we introduce and analyze an efficient region growing algorithm capable of recovering the regions with high accuracy. We show that it is consistent and demonstrate its performance benefits in synthetic simulations.

Recommendations

• Markov random fields in image segmentation
• Fitting Gaussian Markov Random Fields to Gaussian Fields
• Statistical Gaussian model of image regions in stochastic watershed segmentation
• scientific article; zbMATH DE number 1750487
• Data-driven neighborhood selection of a Gaussian field
• Conditional-mean least-squares fitting of Gaussian Markov random fields to Gaussian fields
• Gaussian Markov Random Fields
• A General Bayesian Markov Random Field Model for Probabilistic Image Segmentation

Cites work

• A consistent model selection procedure for Markov random fields based on penalized pseudolikelihood
• A method for the enumeration of various classes of column-convex polygons
• Approximating discrete probability distributions with dependence trees
• Asymptotic enumeration by degree sequence of graphs with degrees \(o(n^{1/2})\)
• Concentration of the spectral measure for large matrices
• Efficiently learning Ising models on arbitrary graphs (extended abstract)
• Generating functions for column-convex polyominoes
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• High-dimensional Gaussian graphical model selection: walk summability and local separation criterion
• High-dimensional graphs and variable selection with the Lasso
• High-dimensional Ising model selection using \(\ell _{1}\)-regularized logistic regression
• scientific article; zbMATH DE number 5957391 (Why is no real title available?)
• scientific article; zbMATH DE number 1064667 (Why is no real title available?)
• scientific article; zbMATH DE number 1134987 (Why is no real title available?)
• scientific article; zbMATH DE number 1526219 (Why is no real title available?)
• Information-Theoretic Limits of Selecting Binary Graphical Models in High Dimensions
• Large deviation principle for the border of a random Young diagram
• Large deviations in the geometry of convex lattice polygons
• Large deviations of convex polyominoes
• Large deviations techniques and applications.
• Learning Markov networks: Maximum bounded tree-width graphs
• Model selection and estimation in the Gaussian graphical model
• On the angular defect of triangulations and the pointwise approximation of curvatures
• Polygons, polyominoes and polycubes
• Prescribing a System of Random Variables by Conditional Distributions
• Reconstruction of Markov random fields from samples: some observations and algorithms
• Sparse inverse covariance estimation with the graphical lasso
• The Complexity of Distinguishing Markov Random Fields
• The generating function of convex polyominoes: The resolution of a \(q\)- differential system
• The limit shape of convex lattice polygons and related topics
• The Schur complement and its applications
• The two-dimensional Ising model

Cited in

(3)

• Reconstruction of Markov random fields from samples: some observations and algorithms
• Model selection for Markov random fields on graphs under a mixing condition
• Data-driven neighborhood selection of a Gaussian field

This page was built for publication: Region selection in Markov random fields: Gaussian case

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6097551)