Intercausal Independence and Heterogeneous Factorization

It is well known that conditional independence can be used to factorize a joint probability into a multiplication of conditional probabilities. This paper proposes a constructive definition of intercausal independence, which can be used to further factorize a conditional probability. An inference algorithm is developed, which makes use of both conditional independence and intercausal independence to reduce inference complexity in Bayesian networks. Key words: Bayesian networks, intercausal independence (definition, representation, inference) 1 INTRODUCTION In one interpretation of Bayesian networks, arcs are viewed as indication of causality; the parents of a random variable are considered causes that jointly influence the variable (Pearl 1988). The concept intercausal independence refers to situations where the mechanism by which a cause influences a variable is independent of the mechanisms by which other causes influence that variable. The noisy OR-gate and noisy adder models (Go..