Figure 3

Probabilistic Model. The probabilistic graphical model represents conditional independence relationships between the data $\mathcal{D}$, the network structure $\mathcal{M}$, and the hyper-parameter of the prior on $\mathcal{M}$. The conditional independence relationships can be obtained from the graph according to the standard rules of factorization in Bayesian networks, as discussed, e.g., in [34]. This leads to the following expansion: $P\left(\mathcal{D},\mathcal{M},\beta \right)=P\left(\mathcal{D}|\mathcal{M}\right)P\left(\mathcal{M}|\beta \right)P\left(\beta \right)$.