Inter-causal Independence and Heterogeneous Factorization
This work addresses computational efficiency for probabilistic inference in Bayesian networks, but appears incremental as it builds on existing factorization methods.
The paper tackled the problem of reducing inference complexity in Bayesian networks by proposing inter-causal independence to factorize conditional probabilities, resulting in an inference algorithm that leverages both conditional and inter-causal independence.
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 inter-causal independence, which can be used to further factorize a conditional probability. An inference algorithm is developed, which makes use of both conditional independence and inter-causal independence to reduce inference complexity in Bayesian networks.