Relational Bayesian Networks

TL;DR

This paper introduces a novel approach to representing probabilistic relations among multiple random events using Bayesian networks, enhancing expressiveness and flexibility over previous rule-based methods.

Contribution

It presents a new formalism for probabilistic relations with a direct distribution representation via Bayesian networks, allowing complex constraints and nested functions.

Findings

More expressive formalism for probabilistic relations

Ability to specify complex, nested combination functions

Supports constraints on event equalities

Abstract

A new method is developed to represent probabilistic relations on multiple random events. Where previously knowledge bases containing probabilistic rules were used for this purpose, here a probability distribution over the relations is directly represented by a Bayesian network. By using a powerful way of specifying conditional probability distributions in these networks, the resulting formalism is more expressive than the previous ones. Particularly, it provides for constraints on equalities of events, and it allows to define complex, nested combination functions.