Being Bayesian about Network Structure

Abstract

In many domains, we are interested in analyzing the structure of the underlying distribution, e.g., whether one variable is a direct parent of the other. Bayesian model-selection attempts to find the MAP model and use its structure to answer these questions. However, when the amount of available data is modest, there might be many models that have non-negligible posterior. Thus, we want compute the Bayesian posterior of a feature, i.e., the total posterior probability of all models that contain it. In this paper, we propose a new approach for this task. We first show how to efficiently compute a sum over the exponential number of networks that are consistent with a fixed ordering over network variables. This allows us to compute, for a given ordering, both the marginal probability of the data and the posterior of a feature. We then use this result as the basis for an algorithm that approximates the Bayesian posterior of a feature. Our approach uses a Markov Chain Monte Carlo (MCMC) method, but over orderings rather than over network structures. The space of orderings is much smaller and more regular than the space of structures, and has a smoother posterior `landscape'. We present empirical results on synthetic and real-life datasets that compare our approach to full model averaging (when possible), to MCMC over network structures, and to a non-Bayesian bootstrap approach.

Appears in Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence (UAI2000)

References in corpus (2)

• Data Analysis with Bayesian Networks: A Bootstrap Approach
• Learning Bayesian Networks: A Unification for Discrete and Gaussian Domains

Cited by in corpus (15)

• Searching for Bayesian Network Structures in the Space of Restricted Acyclic Partially Directed Graphs
• Network inference and biological dynamics
• Gaussian Process Networks
• Inferring Regulatory Networks by Combining Perturbation Screens and Steady State Gene Expression Profiles
• A Non-Parametric Bayesian Method for Inferring Hidden Causes
• Comparative Benchmarking of Causal Discovery Techniques
• An Efficient Search Strategy for Aggregation and Discretization of Attributes of Bayesian Networks Using Minimum Description Length
• Is a single unique Bayesian network enough to accurately represent your data?
• Making life better one large system at a time: Challenges for UAI research
• Bayesian causal inference via probabilistic program synthesis
• Interactive Causal Structure Discovery in Earth System Sciences
• Self-supervised self-supervision by combining deep learning and probabilistic logic
• Estimating Well-Performing Bayesian Networks using Bernoulli Mixtures
• Markov Chain Monte Carlo using Tree-Based Priors on Model Structure
• Identification of Latent Variables From Graphical Model Residuals