Wanrong Sun MSc Defence

Title: Learning NAT-Modeled Bayesian Networks from Data with Extended BD Scores

Chair: Dr. Mark Wineberg
Advisor: Dr. Yang Xiang
Advisory: Dr. Ayesha Ali [Math & Stats]
Non-Advisory: Dr. Stefan Kremer

Abstract:

Bayesian networks (BNs) are widely used for concise knowledge representation and
probabilistic inference in uncertain environments. By applying conditional probability
tables (CPTs) associated with variables, a BN can encode conditional independence (CI)
between variables. However, the complexity of CPTs is exponential on the number of
parents per variable. Non-impeding noisy-AND tree (NAT) models are local structures
that can be applied to BNs to significantly improve the efficiency. The complexity of NATmodeled
BNs is linear on the number of parents per variable.
To take advantage of representation and inference efficiency by NAT-modeled BNs,
this work studies Bayesian approach for learning NAT-modeled BN structures from data.
We extend the meta-networks to encode NAT local structures and parameters. By
applying the extended meta-networks, we develop a Bayesian Dirichlet (BD) scoring
function to evaluate the candidate structures. We present a heuristic search to reduce
the search complexity due to huge alternative combinations of global and local
structures. An experiment is conducted to evaluate the extended BD score and heuristic
search algorithms for learning NAT-modeled BN structures. It demonstrates that the
inference with learned NAT-modeled BNs is sufficiently accurate and significantly more
efficient than the equivalent tabular BNs.