Spectral Bayesian network theory
From MaRDI portal
Abstract: A Bayesian Network (BN) is a probabilistic model that represents a set of variables using a directed acyclic graph (DAG). Current algorithms for learning BN structures from data focus on estimating the edges of a specific DAG, and often lead to many `likely' network structures. In this paper, we lay the groundwork for an approach that focuses on learning global properties of the DAG rather than exact edges. This is done by defining the structural hypergraph of a BN, which is shown to be related to the inverse-covariance matrix of the network. Spectral bounds are derived for the normalized inverse-covariance matrix, which are shown to be closely related to the maximum indegree of the associated BN.
Recommendations
- Construction and learning methods of Bayesian networks
- Bayesian network structure learning: hybridizing complete search with independence tests
- Learning Bayesian networks: The combination of knowledge and statistical data
- On open questions in the geometric approach to structural learning Bayesian nets
- Learning Bayesian networks from data: An information-theory based approach
Cites work
- A well-conditioned estimator for large-dimensional covariance matrices
- An Asymptotic Expansion for the Distribution of the Latent Roots of the Estimated Covariance Matrix
- Approximating discrete probability distributions with dependence trees
- Asymptotics of eigenvalues and unit-length eigenvectors of sample variance and correlation matrices
- Bayesian Graphical Models for Discrete Data
- Causality. Models, reasoning, and inference
- High-dimensional learning of linear causal networks via inverse covariance estimation
- scientific article; zbMATH DE number 4174001 (Why is no real title available?)
- Large-sample learning of Bayesian networks is NP-hard
- Non-homogeneous dynamic Bayesian networks with Bayesian regularization for inferring gene regulatory networks with gradually time-varying structure
- Sparse inverse covariance estimation with the graphical lasso
- The asymptotic variance matrix of the sample correlation matrix
Cited in
(6)- Algebraic Bayesian networks: checking backbone connectivity
- A review of Bayesian networks and structure learning
- scientific article; zbMATH DE number 5957295 (Why is no real title available?)
- The marginal factorization of Bayesian networks and its application
- 10.1162/jmlr.2003.3.4-5.709
- Inference on the eigenvalues of the normalized precision matrix
This page was built for publication: Spectral Bayesian network theory
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6173923)
