Probabilistic Methods for Bioinformatics

Chapter 8 Learning Bayesian Network Structure

In a Bayesian network, the DAG is called the structure and, as mentioned at the beginning of the previous chapter, the conditional probability distributions are called the parameters. In the previous chapter we discussed learning the parameters. In this chapter we address learning the structure. In Section 8.1, we formalize the notion of structure learning. Section 8.2 concerns score-based structure learning. Constraint-based structure learning is the focus of Section 8.3. In Section 8.4 we apply structure learning to inferring causal influences from data. When we do not have a large amount of ...