A Credal Extension of Independent Choice Logic

TL;DR

This paper introduces a credal extension to the independent choice logic, allowing for modeling dependencies among probabilistic choices and enabling conservative probabilistic reasoning with multiple joint distributions.

Contribution

It extends Poole's independent choice logic by relaxing independence assumptions and adopting a credal semantics with multiple joint probability distributions.

Findings

Preliminary tests on object ranking show informative inferences.

The approach handles tasks with unknown dependencies among probabilistic choices.

Conservative probabilistic reasoning is achieved through multiple joint distributions.

Abstract

We propose an extension of Poole's independent choice logic based on a relaxation of the underlying independence assumptions. A credal semantics involving multiple joint probability mass functions over the possible worlds is adopted. This represents a conservative approach to probabilistic logic programming achieved by considering all the mass functions consistent with the probabilistic facts. This allows to model tasks for which independence among some probabilistic choices cannot be assumed, and a specific dependence model cannot be assessed. Preliminary tests on an object ranking application show that, despite the loose underlying assumptions, informative inferences can be extracted.