DeepAI AI Chat
Log In Sign Up

Exploiting Causal Independence in Bayesian Network Inference

by   N. L. Zhang, et al.

A new method is proposed for exploiting causal independencies in exact Bayesian network inference. A Bayesian network can be viewed as representing a factorization of a joint probability into the multiplication of a set of conditional probabilities. We present a notion of causal independence that enables one to further factorize the conditional probabilities into a combination of even smaller factors and consequently obtain a finer-grain factorization of the joint probability. The new formulation of causal independence lets us specify the conditional probability of a variable given its parents in terms of an associative and commutative operator, such as "or", "sum" or "max", on the contribution of each parent. We start with a simple algorithm VE for Bayesian network inference that, given evidence and a query variable, uses the factorization to find the posterior distribution of the query. We show how this algorithm can be extended to exploit causal independence. Empirical studies, based on the CPCS networks for medical diagnosis, show that this method is more efficient than previous methods and allows for inference in larger networks than previous algorithms.


page 1

page 2

page 3

page 4


Inter-causal Independence and Heterogeneous Factorization

It is well known that conditional independence can be used to factorize ...

Independence of Causal Influence and Clique Tree Propagation

This paper explores the role of independence of causal influence (ICI) i...

Exploiting Contextual Independence In Probabilistic Inference

Bayesian belief networks have grown to prominence because they provide c...

Exploiting Functional Dependence in Bayesian Network Inference

We propose an efficient method for Bayesian network inference in models ...

Computation of Variances in Causal Networks

The causal (belief) network is a well-known graphical structure for repr...

A Robust Independence Test for Constraint-Based Learning of Causal Structure

Constraint-based (CB) learning is a formalism for learning a causal netw...

Counterexamples to "The Blessings of Multiple Causes" by Wang and Blei

This brief note is meant to complement our previous comment on "The Bles...