Positive dependence in qualitative probabilistic networks

08/19/2022

∙

by   Jack Storror Carter, et al.

∙

0

∙

share

Qualitative probabilistic networks (QPNs) combine the conditional independence assumptions of Bayesian networks with the `qualitative' properties of positive and negative dependence. They attempt to formalise various intuitive properties of positive dependence to allow inferences over a large network of variables. However, we highlight a key mistake in the QPN literature which means that most inferences made by a QPN are not mathematically true. We also discuss how to redefine a QPN in order to fix this issue.