A Bayesian Approach to Learning Bayesian Networks with Local Structure

Abstract

Recently several researchers have investigated techniques for using data to learn Bayesian networks containing compact representations for the conditional probability distributions (CPDs) stored at each node. The majority of this work has concentrated on using decision-tree representations for the CPDs. In addition, researchers typically apply non-Bayesian (or asymptotically Bayesian) scoring functions such as MDL to evaluate the goodness-of-fit of networks to the data. In this paper we investigate a Bayesian approach to learning Bayesian networks that contain the more general decision-graph representations of the CPDs. First, we describe how to evaluate the posterior probability that is, the Bayesian score of such a network, given a database of observed cases. Second, we describe various search spaces that can be used, in conjunction with a scoring function and a search procedure, to identify one or more high-scoring networks. Finally, we present an experimental evaluation of the search spaces, using a greedy algorithm and a Bayesian scoring function.

Appears in Proceedings of the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI1997)

References in corpus (2)

• Context-Specific Independence in Bayesian Networks
• Score and Information for Recursive Exponential Models with Incomplete Data

Cited by in corpus (39)

• Empirical Analysis of Predictive Algorithms for Collaborative Filtering
• The Bayesian Structural EM Algorithm
• Learning Module Networks
• Using Temporal Data for Making Recommendations
• A Bayesian Approach to Tackling Hard Computational Problems
• Prediction, Expectation, and Surprise: Methods, Designs, and Study of a Deployed Traffic Forecasting Service
• Learning Arithmetic Circuits
• Models and Selection Criteria for Regression and Classification
• A Decision Theoretic Approach to Targeted Advertising
• Graphical Models and Exponential Families
• Labeled Directed Acyclic Graphs: a generalization of context-specific independence in directed graphical models
• Structure and Parameter Learning for Causal Independence and Causal Interaction Models
• Dependency Networks for Collaborative Filtering and Data Visualization
• Learning Bayesian Networks with Restricted Causal Interactions
• Embedded Bayesian Network Classifiers
• The Libra Toolkit for Probabilistic Models
• Staged Mixture Modelling and Boosting
• Web-Based Question Answering: A Decision-Making Perspective
• Finding Ground States of Sherrington-Kirkpatrick Spin Glasses with Hierarchical BOA and Genetic Algorithms
• Fast Learning from Sparse Data
• Chi-square Tests Driven Method for Learning the Structure of Factored MDPs
• Structure Learning of Contextual Markov Networks using Marginal Pseudo-likelihood
• A Bayesian Decision Tree Algorithm
• Analysis of Estimation of Distribution Algorithms and Genetic Algorithms on NK Landscapes
• Coordinates: Probabilistic Forecasting of Presence and Availability
• Markov random fields factorization with context-specific independences
• Analysis of Evolutionary Algorithms on the One-Dimensional Spin Glass with Power-Law Interactions
• Document Clustering using Sequential Information Bottleneck Method
• Distance-Based Bias in Model-Directed Optimization of Additively Decomposable Problems
• Parameter-less hierarchical BOA
• iBOA: The Incremental Bayesian Optimization Algorithm
• Latent Dependency Forest Models
• A Score-and-Search Approach to Learning Bayesian Networks with Noisy-OR Relations
• Learning Distributional Programs for Relational Autocompletion
• Scalable Bayesian Network Structure Learning with Splines
• Context-Specific Likelihood Weighting
• A New Perspective on Learning Context-Specific Independence
• Transfer Learning, Soft Distance-Based Bias, and the Hierarchical BOA
• Multiobjective hBOA, Clustering, and Scalability