Reasoning about unpredicted change and explicit time

TL;DR

This paper introduces a framework for explaining unpredicted changes in observations through surprises, which are events that change the truth value of a fluent, combining model-based diagnosis with probabilistic methods.

Contribution

It presents a novel approach to reasoning about unexpected changes by formalizing surprises and integrating probabilistic surprise minimization.

Findings

Framework for explaining observations via surprises

Minimal surprise sets with time intervals

Probabilistic surprise minimization approach

Abstract

Reasoning about unpredicted change consists in explaining observations by events; we propose here an approach for explaining time-stamped observations by surprises, which are simple events consisting in the change of the truth value of a fluent. A framework for dealing with surprises is defined. Minimal sets of surprises are provided together with time intervals where each surprise has occurred, and they are characterized from a model-based diagnosis point of view. Then, a probabilistic approach of surprise minimisation is proposed.