On Interactive Explanations as Non-Monotonic Reasoning

TL;DR

This paper models interactive explanations as non-monotonic reasoning processes, addressing inconsistencies and improving the reliability of explanations across multiple inputs in user-system interactions.

Contribution

It introduces a formal model of interactive explanations as non-monotonic reasoning, enabling analysis and resolution of explanation inconsistencies.

Findings

Addresses explanation inconsistency issues

Proposes a formal non-monotonic reasoning framework

Provides insights into interactive explanation properties

Abstract

Recent work shows issues of consistency with explanations, with methods generating local explanations that seem reasonable instance-wise, but that are inconsistent across instances. This suggests not only that instance-wise explanations can be unreliable, but mainly that, when interacting with a system via multiple inputs, a user may actually lose confidence in the system. To better analyse this issue, in this work we treat explanations as objects that can be subject to reasoning and present a formal model of the interactive scenario between user and system, via sequences of inputs, outputs, and explanations. We argue that explanations can be thought of as committing to some model behaviour (even if only prima facie), suggesting a form of entailment, which, we argue, should be thought of as non-monotonic. This allows: 1) to solve some considered inconsistencies in explanation, such as via a specificity relation; 2) to consider properties from the non-monotonic reasoning literature and discuss their desirability, gaining more insight on the interactive explanation scenario.