Lecture Notes in Computer Science vol:8754 pages:350-356
Probabilistic Graphical Models edition:7 location:Utrecht date:17-19 September 2014
An important goal of statistical relational learning formalisms is to develop representations that are compact and expressive but also easy to read and maintain. This is can be achieved by exploiting the modularity of rule-based structures and is related to the noisy-or structure where parents independently influence a joint effect. Typically, these rules are combined in an additive manner where a new rule increases the probability of the effect. In this paper, we present a new language feature for CP-logic, where we allow negation in the head of rules to express the inhibition of an effect in a modular manner. This is a generalization of the inhibited noisy-or structure that can deal with cycles and, foremost, is non-conflicting. We introduce syntax and semantics for this feature and show how this is a natural counterpart to the standard noisy-or. Experimentally, we illustrate that in practice there is no additional cost when performing inference compared to a noisy-or structure.