The What, the Why, and the How of Artificial Explanations in Automated Decision-Making
This work addresses the need for explainable AI in decision-making procedures, offering a theoretical framework that could improve transparency and trust, though it is incremental as it builds on existing philosophical and technical discussions.
The paper tackles the inadequacy of traditional realist accounts of explanation for automated decision-making systems and proposes an alternative epistemic and context-dependent account, which is then used to review existing AI/ML approaches and suggest desiderata for explainable systems.
The increasing incorporation of Artificial Intelligence in the form of automated systems into decision-making procedures highlights not only the importance of decision theory for automated systems but also the need for these decision procedures to be explainable to the people involved in them. Traditional realist accounts of explanation, wherein explanation is a relation that holds (or does not hold) eternally between an explanans and an explanandum, are not adequate to account for the notion of explanation required for artificial decision procedures. We offer an alternative account of explanation as used in the context of automated decision-making that makes explanation an epistemic phenomenon, and one that is dependent on context. This account of explanation better accounts for the way that we talk about, and use, explanations and derived concepts, such as `explanatory power', and also allows us to differentiate between reasons or causes on the one hand, which do not need to have an epistemic aspect, and explanations on the other, which do have such an aspect. Against this theoretical backdrop we then review existing approaches to explanation in Artificial Intelligence and Machine Learning, and suggest desiderata which truly explainable decision systems should fulfill.