Abstract. Bayesian networks are efficient tools for probabilistic reasoning over large sets of variables, due to the fact that the joint distribution factorises according to the structure of the network, which captures conditional independence relations among the variables. Beyond conditional independence, the concept of asymmetric (or context specific) independence makes possible the definition of even more efficient reasoning schemes, based on the representation of probability functions through probability trees. In this paper we investigate how it is possible to achieve a finer factorisation by decomposing the original factors for which some conditions hold. We also introduce the concept of approximate factorisation and apply this methodology to the Lazy-Penniless propagation algorithm.

Probability trees are a powerful data structure for representing probabilistic potentials. However, their complexity can become intractable if they represent a probability distribution over a large set of variables. In this paper, we study the problem of decomposing a probability tree as a product of smaller trees, with the aim of being able to handle bigger probabilistic potentials. We propose exact and approximate approaches and evaluate their behaviour through an extensive set of experiments. * Corresponding author